The Routledge Handbook of

Language, Gender, and Sexuality

The Routledge Handbook of Language, Gender, and Sexuality provides an accessible and
authoritative overview of this dynamic and growing area of research. Covering cutting-
edge debates in eight parts, it is designed as a series of mini edited collections, enabling the
reader, and particularly the novice reader, to discover new ways of approaching language,
gender, and sexuality.
With a distinctive focus both on methodologies and theoretical frameworks, the Handbook
includes 40 state-of-the-art chapters from international authorities. Each chapter provides a
concise and critical discussion of a methodological approach, an empirical study to model
the approach, a discussion of real-world applications, and further reading. Each section also
contains a chapter by leading scholars in that area, positioning, through their own work and
chapters in their part, current state-of-the-art and future directions.
This volume is key reading for all engaged in the study and research of language,
gender, and sexuality within English language, sociolinguistics, discourse studies, applied
linguistics, and gender studies.

Jo Angouri is Professor and the University-level Academic Director for Education and
Internationalisation at the University of Warwick, UK, and Visiting Distinguished Professor
at Aalto University, School of Business, Finland. She is author of Culture, Discourse, and
the Workplace. Jo’s research areas include leadership and teamwork in high-pressure, high-
risk professional settings; language, politics, and ideology; and migration, mobility, and
multilingualism.

Judith Baxter was Emeritus Professor of Applied Linguistics at Aston University, UK.
Her areas of research specialism include gender and language, discourse of leadership, and
feminist poststructuralist discourse analysis. She has written numerous journal articles on
these topics as well as four acclaimed monographs.
Routledge Handbooks in Applied Linguistics

Routledge Handbooks in Applied Linguistics provide comprehensive overviews of the key

topics in applied linguistics. All entries for the handbooks are specially commissioned and
written by leading scholars in the field. Clear, accessible and carefully edited Routledge
Handbooks in Applied Linguistics are the ideal resource for both advanced undergradu-
ates and postgraduate students.

The Routledge Handbook of Research Methods in Applied Linguistics

Edited by Jim McKinley and Heath Rose

The Routledge Handbook of Language Education Curriculum Design

Edited by Peter Mickan and Ilona Wallace

The Routledge Handbook of Language and Intercultural Communication

Second Edition
Edited by Jane Jackson

The Routledge Handbook of Forensic Linguistics

Second Edition
Edited by Malcolm Coulthard, Alison May and Rui Sousa-Silva

The Routledge Handbook of Corpus Approaches to Discourse Analysis

Edited by Eric Friginal and Jack A. Hardy

The Routledge Handbook of World Englishes

Second Edition
Edited by Andy Kirkpatrick

The Routledge Handbook of Language, Gender, and Sexuality

Edited by Jo Angouri and Judith Baxter

For a full list of titles in this series, please visit www.routledge.com/series/RHAL

 The Routledge
Handbook of Language,
Gender, and Sexuality

Edited by Jo Angouri and Judith Baxter

First published 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2021 selection and editorial matter, Jo Angouri and Judith Baxter; individual
chapters, the contributors
The right of Jo Angouri and Judith Baxter to be identified as the authors of the
editorial material, and of the authors for their individual chapters, has been
asserted in accordance with sections 77 and 78 of the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from the
publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without intent
to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Angouri, Jo, editor. | Baxter, Judith, 1955- editor.
Title: The Routledge handbook of language, gender, and sexuality /
edited by Jo Angouri and Judith Baxter.
Description: Abingdon, Oxon; New York, NY: Routledge, 2021. |
Series: Routledge handbooks in applied linguistics | Includes bibliographical
references and index.
Identifiers: LCCN 2020047347 | ISBN 9781138200265 (hardback) |
ISBN 9781315514857 (ebook)
Subjects: LCSH: Language and sex. | Language and languages–Sex
differences.
Classification: LCC P120.S48 R68 2021 | DDC 306.44–dc23
LC record available at https://lccn.loc.gov/2020047347
ISBN: 978-1-138-20026-5 (hbk)
ISBN: 978-0-367-74683-4 (pbk)
ISBN: 978-1-3155-1485-7 (ebk)
Typeset in Times New Roman
by Deanta Global Publishing Services, Chennai, India
To Judith
 Contents

List of figures xii

List of tables xiv
List of contributors xvi
Acknowledgements xxiv
Foreword xxv

1 Introduction: language, gender, and sexuality: sketching out the field 1

Jo Angouri

PART I
Variationist approaches 23

2 Non-binary approaches to gender and sexuality 25

Penelope Eckert and Robert J. Podesva (Part I leads)

3 Sexuality as non-binary: a variationist perspective 37

Erez Levon

4 Perception of gender and sexuality 52

Kathryn Campbell-Kibler and deandre miles-hercules

5 Gender diversity and the voice 69

Lal Zimman

PART II
Anthropological and ethnographic approaches 91

6 Ethnography and the shifting semiotics of gender and sexuality 93

Kira Hall and Jenny L. Davis (Part II leads)

7 Gender, language, and elite ethnographies in UK political institutions 108

Sylvia Shaw

vii
Contents

8 ‘Gay, aren’t they?’ An ethnographic approach to

compulsory heterosexuality 121
Jodie Clark

9 Anthropological discourse analysis and the social ordering of

gender ideology 136
Susan U. Philips

10 Using communities of practice and ethnography to answer

sociolinguistic questions 150
Ila Nagar

11 Digital ethnography in the study of language, gender, and sexuality 164

Piia Varis

PART III
Interactional sociolinguistic approaches 179

12 Interactional sociolinguistics: foundations, developments, and

applications to language, gender, and sexuality 181
Cynthia Gordon and Deborah Tannen (Part III leads)

13 Leadership and humour at work: using interactional sociolinguistics to

explore the role of gender 197
Stephanie Schnurr and Nor Azikin Mohd Omar

14 More than builders in pink shirts: identity construction

in gendered workplaces 212
Jo Angouri, Meredith Marra, and Shelley Dawson

15 Interactional sociolinguistics in language and sexuality research:

benefits and challenges 226
Corinne A. Seals

PART IV
Ethnomethodological and conversation analytic approaches 241

16 The accomplishment of gender in interaction: ethnomethodological and

conversation analytic approaches to gender 243
Lorenza Mondada (Part IV lead)

17 Feminist conversation analysis: examining violence against women 258

Emma Tennent and Ann Weatherall

viii
 Contents

18 Performance in action: walking as gendered construction practice in

drag king workshops 272
Luca Greco

19 Gender and sexuality normativities: using conversation analysis to

investigate heteronormativity and cisnormativity in interaction 289
Stina Ericsson

20 Examining girls’ peer culture-in-action: gender, stance, and category

work in girls’ peer language practices 304
Ann-Carita Evaldsson

PART V
Sociocultural and critical approaches 321

21 Language, gender, and sexuality: reflections on the field’s ongoing

critical engagement with the sociopolitical landscape 323
Lia Litosseliti (Part V lead)

22 Applying queer theory to language, gender, and sexuality research in schools 339
Helen Sauntson

23 Text trajectories and gendered inequalities in institutions 354

Susan Ehrlich and Tanya Romaniuk

24 ‘I thought you didn’t accept gay marriage Fr’: combining corpus

linguistics and critical discourse analysis to investigate the
representation of gay marriage and the Irish Mammy stereotype
in Mrs Brown’s Boys 368
Bróna Murphy and María Palma-Fahey

25 The impact of language and gender studies: public engagement and

wider communication 382
Deborah Cameron

PART VI
Poststructuralist approaches 397

26 Poststructuralist research on language, gender, and sexuality 399

Bonny Norton (Part VI lead)

27 Analysing gendered discourses online: child-centric motherhood and

individuality in Mumsnet Talk 408
Jai Mackenzie

ix
Contents

28 Leadership language of Middle Eastern women: using feminist

poststructuralist discourse analysis to study women leaders in Bahrain 422
Haleema Al A’ali

29 Feminist poststructuralism: discourse, subjectivity, the body, and

power: the case of the burkini 437
Chris Weedon and Amal Hallak

30 Affect in language, gender, and sexuality research: studying

heterosexual desire 450
Kristine Køhler Mortensen and Tommaso M. Milani

31 Language, gender, and the discursive production of women as leaders 465

Roslyn Appleby

PART VII
Semiotic and multimodal approaches 479

32 Gender and sexuality in discourse: semiotic and multimodal approaches 481

Michelle M. Lazar (Part VII lead)

33 Multimodal constructions of feminism: the transfiguration of

Chimamanda Ngozi Adichie in Vogue 494
Linda McLoughlin

34 Judged and condemned: semiotic representations of women criminals 509

Carmen Rosa Caldas-Coulthard

35 Confident appearing: revisiting Gender Advertisements in contemporary

culture 528
Kirsten Kohrs and Rosalind Gill

36 Doing gender and sexuality intersectionally in multimodal social media

practices 543
Sirpa Leppänen and Sanna Tapionkaski

PART VIII
Corpus linguistic approaches 557

37 Lovely nurses, rude receptionists, and patronising doctors:

determining the impact of gender stereotyping on patient feedback 559
Paul Baker and Gavin Brookes (Part VIII leads)

x
 Contents

38 Investigating gendered language through collocation: the case of mock

politeness 572
Charlotte Taylor

39 The South African news media and representations of sexuality 587

Sally Hunt

40 Women victims of men who murder: XML mark-up for nomination,

collocation, and frequency analysis of language of the law 602
Amanda Potts and Federica Formato

Index 619

xi
 Figures

2.1 Percent negative concord by binary gender 26

2.2 Percent negative concord by binary gender and social category 27
2.3 Negative concord by subcategory. Shown as standard deviation
from class mean, including Jocks, Burnouts, and In-Betweens 28
2.4 H1*–H2* (dB) as a function of speaker age for women (F) and men (M) 31
3.1 Average mean pitch level in Hz for speakers ordered from left to
right by position on the ‘butch’–‘lipstick’ continuum 46
3.2 Average mean pitch level in Hz for speakers ordered from left to right
by position on the ‘butch’–‘lipstick’ continuum and divided by speech topic 47
4.1 Distribution of gay ratings 60
4.2 Distribution of masculine ratings 60
4.3 Distribution of listener slopes between gay and masculine 61
4.4 Perceptual clusters 62
5.1 Mean centre of gravity for /s/ by gender grouping (by group) 78
5.2 Mean centre of gravity for /s/ (AFAB speakers) 80
5.3 Mean centre of gravity for /s/ (AMAB speakers) 81
18.1 The makeup space 278
18.2 The makeup space transformed into a walking space 278
20.1 The four girls constituting the core group are standing in the hallway 308
20.2 Beyan runs up the stairs, followed by Samina 310
20.3a The girls commit themselves to confront Rana 312
20.3b Azra and Yaasmiin perform a catfight 312
20.4 The girls end up chasing Rana 314
27.1 Research design for the Mumsnet study 413
30.1 Configuration of a male dating profile 455
33.1 Adichie in a trendy loft-style apartment 501
34.1 An adolescent is arrested for the seventeenth time for stealing a
motorbike in Anápolis, Goiania 513
34.2 A motorboy is filmed during the robbery and the thief is shot
by the police 514
34.3 Cousins 515
34.4 Grandma with child 516
34.5 Naughty old woman 516

xii
 Figures

34.6 ‘Woman criminal’ 519

34.7 ‘Woman criminal’ 519
34.8 Elize in her lingerie 522
34.9 Elize the murderess 523
34.10 Members of a gang of young women, referred to as ‘The Blondie
Gang’ (‘Guangue das Loiras’) 524
35.1 MiuMiu ‘Subjective Reality Afternoon Sun’ 533
35.2 Gucci 534
35.3 Gucci 535
35.4 Wonderbra (1990s). Image courtesy of The Advertising Archives 536
36.1 Modified image of a post from a Finnish discussion forum
for ‘bronies’ 549
36.2 Kilikali in the video 552
36.3 Bianca’s performance in the video as a blond Finnish woman 553
38.1 Gendered collocates of adjectival mock politeness labels 579
38.2 Distribution of male/female performance of behaviours 580
38.3 Evaluative collocates of adjectival mock politeness labels 581
38.4 Sketch Engine Thesaurus output for ‘sarcastic’ 582
38.5 Sketch Engine Thesaurus output for ‘patronising’ 583
40.1 Example text without any additional mark-up 605
40.2 Example text with XML mark-up 606
40.3 Example output in a Sketch Engine concordance window 606
40.4 Framework developed to operationalise categories of solidarity and
distance between the judge and the victims 610

xiii
 Tables

4.1 Common comment types 59

4.2 Perceptual clusters, by listener gay/masculine pattern 62
5.1 Dimensions of sex and gender 73
5.2 Number of speakers by gender assignment, identity, and sexuality 76
5.3 Results of linear regression Model #1 77
5.4 Results of ANOVA Model #1 78
5.5 Results of Tukey HSD post-hoc testing on ANOVA Model #1 78
5.6 Results of linear regression Model #2 82
5.7 Results of ANOVA Model #2 83
5.8 Results of Tukey HSD post-hoc testing on ANOVA Model #2 83
8.1 Grammatical analysis of some clauses 127
8.2 Investigation of the process types in the clauses of each account 127
8.3 Clauses in Speedo’s account 128
8.4 Material clauses in Speedo’s and Beth’s accounts 128
8.5 Participants in Speedo’s and Beth’s accounts 129
8.6 A comparison of the clauses that serve the ‘justification’ function 129
8.7 Participants in Speedo’s account 130
8.8 Verbal clauses interrogating Speedo’s sexual identity 131
24.1 The Mrs Brown’s Boys dataset 371
24.2 Negative semantic prosody for gay marriage in Mrs Brown’s Boys 373
37.1 Concordances of the collocation of ‘he’ and ‘good’ 563
37.2 Frequency of positive and negative evaluative adjectives as
collocates of ‘he’ and ‘she’ when referring to NHS practitioners 564
38.1 Comparison of collocate ranking according to different measures 575
39.1 Twenty strongest collocates for HOMOSEXUAL by MI: 4L to 4R,
minimum frequency 3 594
39.2 Selected concordance lines for HOMOSEXUAL 595
39.3 Selected concordance lines for *tabane 596
39.4 Selected concordance lines for anti-gay: legal aspect 596
39.5 Selected concordance lines for anti-gay: anti-gay ‘activism’ 596
39.6 Selected concordance lines for corrective rape 598
40.1 Overview of cases included in sentencing remarks corpus 607

xiv
 Tables

40.2 Overview of naming strategy categories with frequencies

and percentages 609
40.3 Type, frequency, and percentage of given name references 610
40.4 Frequency of categorising strategies 612
40.5 Frequency and percentages of (grammatical) case 613
40.6 Collocates of WHV in the genitive case, categorised semantically
and ranked in descending order of frequency of collocation 614

xv
 Contributors

Haleema Al A’ali is Assistant Professor in the Department of English language and Literature
in the University of Bahrain. She received her PhD in Linguistics from Aston University, UK,
in 2013. Her general research interests are: language and gender, leadership language, lan-
guage in the workplace, entrepreneurship and gender, and discourses of ageing and gender.
Her current research focuses on language, gender, and leadership in the Middle East, and the
role of gendered discourses in perpetuating workplace inequalities in the region.

Jo Angouri is Professor and Academic Director for Education and Internationalisation at

the University of Warwick, UK. Jo has published extensively on language and identity,
teamwork and leadership in professional settings, and migration, mobility, and multilin-
gualism. Jo is the author of Culture, Discourse, and the Workplace (Routledge, 2018)
and has co-edited Negotiating Boundaries at Work (EUP, 2017). She is also a National
Teaching Fellow (UK) and fully committed to pedagogic innovation and education for
global citizenship. Jo is Subject Chair for Linguistics, Language, Communication and
Media on the Scopus board.

Roslyn Appleby is Applied Linguistics Researcher and Educator in the Faculty of Arts and
Social Sciences at the University of Technology Sydney, Australia. Her research interests
include the cultural politics of gender and sexuality, and she is the author of ELT, Gender
and International Development (2010), Men and Masculinities in Global English Language
Teaching (2004), and Sexing the Animal in a Posthumanist World (2019).

Paul Baker is Professor of English Language at Lancaster University, UK, and member of
the Corpus Approaches to Social Science ESRC Research Centre where he specialises in
corpus linguistics and discourse analysis. He has written 19 books, 36 journal articles, and
28 book chapters, and is the commissioning editor of Corpora journal. His books include
Using Corpora in Discourse Analysis (2006), Sexed Texts (2008), and American and British
English: Divided by a Common Language (2017).

Gavin Brookes is Senior Research Associate in the Centre for Corpus Approaches to Social
Science at Lancaster University, UK, and Associate Editor of the International Journal
of Corpus Linguistics. His research interests include corpus linguistics, discourse studies,
multimodality, and health communication. His recent books include Corpus, Discourse and
Mental Health (Bloomsbury, with D. Hunt, 2020) and The Language of Patient Feedback:
A Corpus Linguistic Study of Online Health Communication (Routledge, with P. Baker and
C. Evans, 2019).

xvi
 Contributors

Carmen Rosa Caldas-Coulthard is Professor of English Language and Applied Linguistics

at the Federal University of Santa Catarina, Brazil. She is also Senior Research Fellow in the
English Department at the University of Birmingham, UK, where she taught and researched
for many years. She has published widely in the areas of critical discourse analysis, media,
gender studies, social semiotics, and visual communication. Her most recent publication is
the edited volume Innovations and Challenges: Women, Language and Sexism (2020).

Deborah Cameron is Professor of Language and Communication at Oxford University,

UK. In addition to the academic books and articles she has published on language and gen-
der, she has a long record of media/public engagement: she is the author of a general interest
book, The Myth of Mars and Venus (2007), a regular contributor to broadcast discussions
(heard on the BBC, CBC, US National Public Radio, and Radio New Zealand), and the crea-
tor of the blog Language: A Feminist Guide.

Kathryn Campbell-Kibler is Associate Professor in the Department of Linguistics at the

Ohio State University, USA. Her research investigates the social meanings of linguistic
variation. She focuses on how listeners process individual variables and incorporate them
into their social perceptions of speakers. She is also the outreach director for the ‘the Pod’,
a working linguistics lab in the science museum COSI and the director of See Your Speech,
an interactive website that gives users visual displays based on their own speech.

Jodie Clark is Senior Lecturer in English Language at Sheffield Hallam University, UK.
Her research and teaching focus upon gender, sexuality, race, and class. Her work engages
in grammatical analysis of everyday descriptions of the social world as a means of identify-
ing ideas for alternative, transformative social structures. She is the author of Language, Sex
and Social Structure (2012) and Selves, Bodies and the Grammar of Social Worlds (2016).

Jenny L. Davis (Chickasaw) is Associate Professor of Anthropology and American Indian

Studies at the University of Illinois, USA. Her research focuses on Indigenous language
revitalisation; gender/sexuality; and collaborative research methods and ethics. Her work
has received two book prizes: the Beatrice Medicine Award for her 2018 book, Talking
Indian: Identity and Language Revitalization in the Chickasaw Renaissance, and the Ruth
Benedict Book Prize for her 2014 co-edited volume, Queer Excursions: Retheorizing
Binaries in Language, Gender, and Sexuality.

Shelley Dawson is a Research and Teaching Fellow in the School of Linguistics and Applied
Language Studies at Te Herenga Waka – Victoria University of Wellington, New Zealand.
Her doctoral research examined exchange students’ identity negotiations around gender,
sexuality and nationality in study abroad contexts. Her work makes use of interdisciplinary,
critical approaches, focusing on the influence of ideological structures in interaction. Most
recently she has been investigating the instantiation of implicit bias at a research funding
agency and the implications for inclusive practices.

Penelope Eckert is Albert Ray Lang Professor of Linguistics and (by courtesy) Anthropology
at Stanford University, USA. Her research, based on ethnographic studies of sociolinguistic
variation among adolescents and preadolescents, examines the construction of meaning in
stylistic practice.

xvii
Contributors

Susan Ehrlich is Professor of Linguistics at York University, Canada. She has written exten-
sively on language, sexual violence, and the law and is currently working on a project that
investigates intertextual practices in the legal system in order to shed light on broader patterns
of social inequalities. Recent books include The Handbook of Language, Gender, and Sexuality
(co-edited with Miriam Meyerhoff and Janet Holmes, 2014) and Discursive Constructions of
Consent in the Legal Process (co-edited with Diana Eades and Janet Ainsworth, 2016).

Stina Ericsson is Professor of Swedish at the University of Gothenburg, Sweden. She is the
co-editor of a book on sociolinguistic methods, and her research interests include gender,
sexuality, disability, interaction, multimodality, and technology. Currently, she is involved
in two interdisciplinary research projects, one where she investigates interactions between
pregnant Arabic speakers and Swedish-speaking midwives in Swedish antenatal care, and
one project on categorisations of people through language and public space in relation to
Universal Design.

Ann-Carita Evaldsson is Professor of Education at Uppsala University, Sweden. Her

research draws on multimodal interactional approaches to children’s peer language prac-
tices, morality, emotions and identity work (gender, class, ethnicity, disability) in culturally
diverse settings. She has published extensively in the Journal of Pragmatics, Text & Talk,
Childhood, ROCSI, Multilingua, Emotional and Behavioral difficulties.

Federica Formato is Senior Lecturer in Sociolinguistics at the University of Brighton,

UK. She quantitatively and qualitatively investigates sexist language in the media towards
female politicians, the gendered crime of femminicidio (intimate partner homicide) and con-
structions of fatherhood on social networks, in the context of Italy. Her first monograph
Gender, Discourse and Ideology in Italian was published in 2019.

Rosalind Gill is Professor of Social and Cultural Analysis at City, University of London,
UK. She is author or editor of ten books including Gender and the Media (2007) and New
Femininities: Postfeminism, Neoliberalism and Subjectivity (2011). Her most recent book
(with Meg-John Barker and Laura Harvey) is Mediated Intimacy: Sex Advice in Media
Culture (2018). She is currently completing a monograph on confidence for Duke University
Press (with Shani Orgad).

Cynthia Gordon is Associate Professor in the Department of Linguistics at Georgetown

University, USA. Her research interests include interactional sociolinguistics; discourse
in family, health-related, and digital contexts; and intertextuality and metadiscourse. She
is author of Making Meanings, Creating Family (Oxford University Press, 2009) and co-
editor (with Alla Tovares) of Identity and Ideology in Digital Food Discourse (Bloomsbury,
2020). She is on the editorial boards of Language in Society, Journal of Sociolinguistics,
and Journal of Language and Social Psychology.

Luca Greco is Professor in Sociolinguistics at the Université de Lorraine, France, and edi-
tor of the francophone journal Langage et Société. They is the author of Dans les coulisses
du genre: la fabrique de soi chez les Drag Kings (2018) and of various articles, special
issues, and edited books on gender and language studies, queer linguistics, and membership
categorisation analysis. Their research focuses on gender and multimodality, performance
in contemporary art and everyday practices, and on borders in action.

xviii
 Contributors

Kira Hall is Professor of Linguistics and Anthropology at the University of Colorado

Boulder, USA. Currently the President of the Society for Linguistic Anthropology, she has
published on such topics as language and sexuality in India, Hinglish, embodied sociolinguis-
tics, identity, mass hysteria, and Trump’s use of comedic gesture. In addition to the early vol-
umes Gender Articulated (1995) and Queerly Phrased (1997), she is co-editor of the journal
Gender and Language and The Oxford Handbook of Language and Sexuality (forthcoming).

Amal Hallak is a Freelance Researcher. She did her BA in English Language and Literature
at Aleppo University, Syria. She is currently finishing a PhD in Critical and Cultural Theory at
Cardiff University, UK. Since coming to the UK Amal has worked as a Research Fellow on a
four-year AHRC funded research project on Translation and Translanguaging: Investigating
Linguistic and Cultural Transformation in Superdiverse Wards in Four UK Cities.

Sally Hunt is Staff Tutor and Lecturer in English Language and Applied Linguistics at the
Open University, UK, after many years at Rhodes University, South Africa. Her research and
supervision interests centre on the representation of identity, especially gender and sexual-
ity, in the media and in fiction, using corpus linguistics within a critical discourse analysis
approach. The construction of the gendered body in children’s fiction is a particular focus.

Kirsten Kohrs is currently Senior Lecturer at the University of Greenwich, UK, following
an extensive and stellar career creating commercial communication. Her research interests
focus around visual communication and discourse analysis. She holds a PhD in Culture,
Media and Creative Industries from King’s College London, UK, and her academic writ-
ing appears in the International Journal of Social Research Methodology, the Journal of
Fashion Marketing and Management and in the Routledge series New Directions in Public
Relations & Communication Research.

Michelle M. Lazar is Associate Professor and the Head of the Department of English
Language and Literature at the National University of Singapore. A critical discourse ana-
lyst by academic training, her research centres on gender, sexuality, (post)feminism, multi-
modality, media, politics, and southern praxis. She has published widely in all these areas,
and is the founding editor of the Routledge Critical Studies in Discourse monograph series.

Sirpa Leppänen is Professor in the Department of Language and Communication Studies

at University of Jyväskylä, Finland. She has published widely on (1) multilingualism and
semiotic diversity as a resource for social interaction in informal and interest-driven social
media, (2) identifications, communality, and gender online, and (3) transgression as a means
for cultural production and digital work.

Erez Levon is Professor of Sociolinguistics and Director of the Center for the Study of
Language and Society at the University of Bern, Switzerland. His work focuses on how peo-
ple produce and perceive socially meaningful patterns of variation in language. He is par-
ticularly interested in variation as it relates to gender and sexuality, and how these intersect
with other aspects of lived experience (notably race, class, and national belonging). Erez
is the author of Language and the Politics of Sexuality (2010) and co-editor of Language,
Sexuality and Power (2016).

xix
Contributors

Lia Litosseliti is Associate Dean International at City, University of London, UK. She has
authored and edited eight books on gender, language/discourse, and research methodologies,
including Gender and Language: Theory and Practice (2006), Gender and Language Research
Methodologies (2008), and Research Methods in Linguistics (2018, 2nd ed). Lia has served
as President of the International Gender and Language Association (IGALA) and Associate
Editor of Gender and Language, and acts as reviewer for several funding bodies and journals.

Jai Mackenzie is British Academy Postdoctoral Fellow at the University of Nottingham,

UK. Her primary research expertise lies in explorations of language, gender, sexuality,
and parenthood, especially in new media contexts. She is the Principle Investigator for the
Marginalised Families Online project, which explores the role of digital media for sin-
gle and same-sex parent families in the UK, and recently published her first monograph:
Language, Gender and Parenthood Online (2019).

Meredith Marra is Professor in Linguistics at Te Herenga Waka – Victoria University of

Wellington, New Zealand. Since 2015 she has been Director of the Wellington Language in
the Workplace Project (LWP), a long-standing sociolinguistic research project investigat-
ing effective workplace communication in a range of contexts. Meredith’s research focuses
on gender, ethnic, and leadership identities, particularly in the setting of meetings. Most
recently her work has turned to empowering newcomers, including skilled migrants and
refugees transitioning into the New Zealand workplace.

Linda McLoughlin is Principal Lecturer at Liverpool Hope University, UK, where she
teaches courses on language, gender, and sexuality, language and law, and sociolinguis-
tics. Her most recent publications, A Critical Discourse Analysis of South Asian Women’s
Magazines (2017), deals with issues of gender, race, class, diaspora, and globalisation and
‘The Nirbhaya who lived’ (Gender and Language, 2019) examines representations of sexual
violence in Femina. She is an ordinary member of IGALA and a senior fellow of the HEA.

Tommaso M. Milani is Professor of Multilingualism at the University of Gothenburg,

Sweden, and holds Visiting Professorships at Umeå University, Sweden, and the University
of the Witwatersrand Johannesburg, South Africa. His areas of research interests include
discourse analysis, language ideologies, language policy and planning, linguistic landscape,
as well as language, gender, and sexuality. He has published extensively in international
journals and edited volumes. He has recently been appointed general co-editor of the journal
Language in Society.

deandre miles-hercules is a PhD student in the Department of Linguistics at the University of

California, Santa Barbara, USA. They hold a BA in Linguistics with minors in Anthropology
and African American Studies from Emory University, USA. deandre specialises in socio-
cultural linguistics and sociophonetics with particular research interests in identity, Black
Language, Blackfemme-inist Theory, queer theory, embodiment, poetics, and pedagogy.
Their current project employs discourse analysis to study the physico-discursive linguistic
construction of Blackfemme-ininities.

Nor Azikin Mohd Omar is Senior Lecturer at the Faculty of Languages and
Communication, Universiti Sultan Zainal Abidin, Malaysia. Her main research interests
include language in the workplace, particularly in humour, leadership discourse, and

xx
 Contributors

decision-making. She has published several works on humour in workplace contexts and
identity construction.

Lorenza Mondada is Professor of Linguistics at the University of Basel, Switzerland. Her

research deals with social interaction in ordinary, professional, and institutional settings,
within an ethnomethodological and conversation analytic perspective. Her specific focus
is on video analysis and multimodality, integrating language and embodiment in the study
of human action. She has extensively published in J. of Pragmatics, Discourse Studies,
Language in Society, ROLSI, Journal of Sociolinguistics, and co-edited several collective
books (for CUP, De Gruyter, Benjamins, Routledge).

Kristine Køhler Mortensen is a PhD fellow at the Department of Swedish, University of

Gothenburg, Sweden. Her research focuses on language, gender, and sexuality in relation to
such topics as desire and romance, coloniality, and nationalism. Mortensen has published in
international journals, handbooks, and edited volumes within the field of sociolinguistics.
Mortensen is currently the secretary of the International Gender and Language Association.

Bróna Murphy lectures in Language Education at the University of Edinburgh, UK. Her
research areas explore spoken corpus linguistics, Irish English pragmatics, and language
and gender in fictional media. Her work involves looking at small specialised sociolinguis-
tic-oriented corpora to explore functionally motivated linguistic variation.

Ila Nagar is Associate Professor at the Department of Near Eastern Languages and Cultures
at the Ohio State University, USA. She is a sociolinguist and works on language, sexual-
ity, power, and meaning. Her first monograph Being Janana: Language and Sexuality in
Contemporary India (2019) examines how jananas, who are men who desire men but can
have heteronormatively masculine positions in society, make meaning of the marginalisa-
tion of their sexuality and desire.

Bonny Norton, FRSC, is University Killam Professor in the Department of Language and
Literacy Education, University of British Columbia, Canada. Her primary research interests
are identity and language learning, critical literacy, and international development. Recent
publications include a 2017 MLJ special issue on language teacher identity. A Fellow of
the Royal Society of Canada and the American Educational Research Association, she was
awarded BC CUFA 2020 Academic of the Year for her work on the Global Storybooks project
(https://globalstorybooks.net/).

María Palma-Fahey lectures in Spanish language and in Intercultural Communication for

Business at Shannon College of Hotel Management, National University of Ireland. Her
research interests include culture and identity, Irish English pragmatics, language and gen-
der in fictional media, and pragmatic/regional variation across Spanish(es).

Susan U. Philips is Professor Emerita of Anthropology at the University of Arizona, USA.

She is the co-editor of Language, Gender and Sex in Comparative Perspective (1987) and
author of The Invisible Culture: Communication in Classroom and Community on the Warm
Springs Indian Reservation (1983) and Ideology in the Language of Judges: How Judges
Practice Law, Politics and Courtroom Control (1998). She is currently interested in the lives
of women in retirement communities.

xxi
Contributors

Robert J. Podesva is Associate Professor of Linguistics at Stanford University, USA. His

research examines the social significance of phonetic variation and its role in the construc-
tion of identity, most notably gender, sexuality, and race.

Amanda Potts is Senior Lecturer in the Centre for Language and Communication Research
at Cardiff University, UK. Her specialism is in corpus-based critical discourse analysis of
texts and topics in public and professional communication. Her main interest is representa-
tions of ideology and identity, most recently in (social) media discourse, medical communi-
cation, and language of the law.

Tanya Romaniuk is Adjunct Instructor in the Department of Communication at Portland

State University, USA. She has previously published on the ‘double bind’ faced by women
politicians (with Susan Ehrlich), on questioning practices (with Steve Clayman), and
on laughter in broadcast news interviews. Her primary interest in gender concerns how
researchers make claims about its role in qualitative research, specifically in relation to the
discursive construction and representation of women in politics.

Helen Sauntson is Professor of English Language and Linguistics at York St John University,
UK. Her research areas are language in education and language, gender, and sexuality. She
has published a range of book chapters and journal articles in these areas and has authored
and edited nine books including Language, Sexuality and Education (2018). She is co-editor
of the Palgrave Studies in Language, Gender and Sexuality book series.

Stephanie Schnurr is Associate Professor at the University of Warwick, UK. Her main
research interests are professional and medical communication with a particular focus on
leadership discourse. Stephanie has published widely on gender, humour, identity, and (im)
politeness. She is the author of The Language of Leadership Narratives (with Jonathan
Clifton and Dorien van de Mieroop, 2019), Language and Culture at Work (with Olga
Zayts, 2017), Exploring Professional Communication (Routledge, 2013), and Leadership
Discourse at Work (2009).

Corinne A. Seals is Senior Lecturer of Applied Linguistics at Victoria University of

Wellington, New Zealand, and the director of the Wellington Translanguaging Project and
its resource branch Translanguaging Aotearoa. She holds a PhD and MS in Linguistics
from Georgetown University, USA, and a BA (Hons) in Sociocultural Linguistics from the
University of California, USA. Her recent publications include Choosing a Mother Tongue:
The Politics of Language and Identity in Ukraine (2019) and Embracing Multilingualism
Across Educational Contexts (2019).

Sylvia Shaw is Senior Lecturer in English language and Linguistics at the University of
Westminster, UK. Her research focus is on language, gender, and politics. This includes
research projects in the House of Commons and an ESRC funded project investigating
gender, language, and participation in the devolved political institutions of the UK. She has
published research on televised political debates and political interviews, and a monograph,
Women, Language and Politics (2020).

Deborah Tannen is University Professor in the Department of Linguistics at Georgetown

University, USA. Her 26 books (13 authored, 13 edited or co-edited) and over 100 articles

xxii
 Contributors

address such topics as conversational interaction, cross-cultural communication, frames the-

ory, conversational vs. literary discourse, gender and language, and social media discourse.
Her books include Conversational Style: Analyzing Talk Among Friends; Talking Voices:
Repetition, Dialogue, and Imagery in Conversational Discourse; Gender and Discourse;
and Talking from 9 to 5: Women and Men at Work.

Sanna Tapionkaski (née Lehtonen), is Senior Lecturer in Applied Linguistics/Discourse

Studies at the Department of Language and Communication Studies, University of
Jyväskylä, Finland. Her research is broadly situated in the area of language, identity, and
children and youth cultures, and draws on discourse studies, cultural studies, and narrative
analysis. She is the author of Girls Transforming: Invisibility and Age-Shifting in Children’s
Fantasy Since the 1970s (2013).

Charlotte Taylor is Senior Lecturer in English Language and Linguistics at University of

Sussex, UK, and editor of CADAAD Journal. Her research interests include impoliteness,
political and media discourses, migration discourses, and corpus methodologies. She is
author of Mock Politeness in English and Italian (2016) and co-author/co-editor of Corpus
Approaches to Discourse (2018), Exploring Absence and Silence in Discourse (2018), Patterns
and Meanings in Discourse (2013), and The Language of Persuasion in Politics (2017).

Emma Tennent is Lecturer in Communication at Te Herenga Waka – Victoria University of

Wellington, New Zealand. She examines how experiences, identities, and relationships are built
through social interaction. Her doctoral research examined victim support helpline calls and
current research interests include gendered violence and digital communication technologies.

Piia Varis is Associate Professor at the Department of Culture Studies and Deputy Director
of Babylon, Centre for the Study of Superdiversity, at Tilburg University, the Netherlands.
Her research interests include digital culture, social media, surveillance and privacy, and the
role of digital media and technologies in knowledge production.

Ann Weatherall is Professor in Psychology at Victoria University of Wellington, New

Zealand. She has published on the intersection of discursive psychology and feminism includ-
ing studies of motherhood, sex work, and sexism. Increasingly her work explores the possibil-
ities of conversation analysis for understanding and challenging a social and moral order that
functions to systematically disadvantage women. A current project is a video study of self-
defence classes to identify how talk and the body are used to prevent gender-based violence.

Chris Weedon is Professor Emerita at Cardiff University, UK. She has published widely on
feminist theory, cultural politics, culture and identity, women’s writing, British Black and
Asian writing, and multi-ethnic Britain. She recently completed a book of life stories from
the GDR and she is currently co-investigator on an AHRC project on memory, identity, and
trauma among refugees from war.

Lal Zimman is Associate Professor of Linguistics at the University of California, USA.

His research focuses on the linguistic practices of transgender speakers, employing a range
of methodologies from sociophonetics to discourse analysis. He is also General Editor of
Oxford University Press’s series Studies in Language and Gender.

xxiii
 Acknowledgements

Academic books are long, transformational journeys. They leave a mark on the author or
the editor, and, hopefully, the reader, and always require the efforts and generosity of many.
This Handbook has been made possible thanks to friends and colleagues who made a sig-
nificant contribution on multiple levels.
Special thanks are due to the contributors to this volume for their support throughout
the project and to me personally after Judith’s passing. I am most grateful for your mes-
sages, warmth, and friendship and, of course, patience in the whole process. Handbooks are
always collective efforts but typically involve little interaction on the way; our project has
very much felt to me like the product of a closely knit community based on the principles of
mutual respect and collegiality.
I am particularly grateful to Janet Holmes who has been particularly influential in my
work on workplace sociolinguistics and my thinking on all aspects of language, gender, and
sexuality. I vastly appreciate Janet’s time and contribution in the early stages of the work
and the framing of the Handbook, and in her generosity for reviewing the, admittedly unu-
sually long, introduction. I am deeply indebted for the most insightful feedback and for all
the support over the years.
The Handbook would not have been completed without the excellent work of my
Research Assistant, Polina Mesinioti, our Editorial Assistant for the submission stage of
the project. Polina joined the team two years ago and her commitment, loyalty, and perse-
verance has made an invaluable contribution to the level of the work. I cannot thank you
enough, Polina, for meticulously cross-checking every aspect, reviewing with rare dedica-
tion, chasing elegantly and tactfully and your amazing ability to keep all the files and folders
in order!
My thanks are also due to Jai Mackenzie, our first Editorial Assistant and Judith’s former
PhD student, for joining our adventure from the start and supporting the earlier stages of
the project. Thank you, Jai, for keeping the project organised and us synchronised. I would
also like to express my appreciation to all my undergraduate and PhD students for playing
a role through their stimulating questions on methodologies and for their critical comments
on established orthodoxies. It is a continuous pleasure to be learning with you all.
My appreciation goes to the British Association of Applied Linguistics (BAAL) for
allowing me to reprint a passage from the BAAL newsletter (2018).

xxiv
 Foreword

The end of a major academic project comes with the energising effect of the submission, the
fulfilling and gratifying effect of the completeness, the hindsight of what could have been
done better or differently, and the impact of the physical effort. The strong emotions are
difficult to describe but, I believe, are shared in the lived experience of our academic com-
munity. The completion of this Handbook is particularly bittersweet. It involves the work
of over 55 colleagues and it is the outcome of our collective effort to reflect and provide a
repository of experience on theory and method. This with a view to empower those coming
new to the field to use our tools and bring news ideas and insights.
At the same time, it will always have a special place in our collective memory and remind
us of the legacy of Judith Baxter, an influential scholar, wonderful educator, and close friend
to many of us. Judith’s career path is reflective of her continuous self-development, explo-
ration, and substantial contribution to the study of gender and language and her deep com-
mitment to education. Starting as an English and History graduate in the late 1970s, Judith
turned to education immediately after and completed a PGCE and an MSc in Educational
Studies. Her thesis in 1999 brought together a number of areas Judith continued unpacking
in her career. Her thesis’ short title ‘Teaching girls to speak out’ is indicative of her commit-
ment to empowering women, her being a true applied linguist applying theory to practice
through education, and her interest in leadership in institutional and public life. She contin-
ued publishing extensively in those areas and her 2014 and 2017 books (Double-voicing:
Power, Gender and Linguistic Expertise; Women Leaders and Gender Stereotyping in the
UK Press: A Poststructuralist Approach) will, without doubt, continue influencing future
scholars.
Judith and I started discussing the Handbook in 2015; we shared an anti-orthodoxy
stance which was solid common ground to design what, we hoped, will complement the
rich scholarship already existing in the field. We also wanted to react to the feedback of
the many students over the years asking why there is so little focus on methodology in
academic writing. It is indeed true; despite putting so much emphasis on the training of
students on research methods and heavily assessing dissertations and theses on methodol-
ogy, our own writing is far less detailed and transparent. Judith and I also shared the belief
that the existing divide between ‘education’ and ‘research’ is artificial and damaging for
the future of academia in general and our field is not an exception. We therefore wanted
to provide the tools to our students to design their own projects and find their voice from
day one.
Judith passing in 2018 was a shock and left a hole. We had just completed the first cycle
of reviews and the Handbook had started taking shape. Knowing how excited Judith was

xxv
Foreword

about the overall quality and breadth of the work, I am absolutely certain that our collective
effort is the best tribute to her memory and a lasting memorial for new scholars in the field.
For me our Handbook has been one of the most difficult and fulfilling projects in my
academic life so far. It is also a project I will always be most proud of and which will remind
me of Judith’s true commitment to a polyphonic approach to discourse analysis and the
question she has been addressing throughout her work: ‘how can we best represent the suc-
cesses, contradictions and diversity of female experiences whilst still working to transform
the appalling inequities that so many girls and women continue to face?’ (2003: 198).
In closing, I am reproducing here a piece that has been published in BAAL News (Issue
113: 6) as a memorial to Judith. It is co-authored by myself and Helen Sauntson. It is
included here with BAAL’s permission.

Looking Back, Looking Forward: In Memory of Judith Baxter by Jo Angouri

(University of Warwick) and Helen Sauntson (York St John University)

It was with great sadness that we lost our dear colleague and friend, Professor Judith
Baxter, in February. Judith was Professor of Applied Linguistics and Head of English
at Aston University and, although retired since 2015, remained an active author and
key contributor to the field of language, gender and sexuality, and to applied linguistics
more broadly, right up until her untimely death. She has been an inspiration to many
students and colleagues, and her work leaves a valuable legacy for current and future
scholars.
Judith was known internationally for her work in the field of language and gender
and particularly for her research into the language of business and language and gender
in different professional settings. She has addressed the issue of the lack of women in
senior leadership and specifically the role of language in negotiating power at work.
She is also known for developing the analytical framework of Feminist Poststructuralist
Discourse Analysis (FPDA). In this piece, we provide brief reflections on Judith’s key
contributions in these areas and evaluate their continued future relevance for applied
linguistics.
Judith described FPDA as a ‘supplementary’ approach, following a Derridaean line
of enquiry. FPDA values multiple perspectives and is conceived to be used in conjunc-
tion with other methods. In her own words, Judith defined FPDA as:

an approach to analysing intertextualised discourses in spoken interaction and

other types of text. It draws upon the poststructuralist principles of complexity,
plurality, ambiguity, connection, recognition, diversity, textual playfulness, func-
tionality and transformation.

FPDA looks into ‘what is happening right now, on the ground, in this very conversa-
tion’ and connects this with societal power structures particularly in relation to gender.
However, rather than focusing upon issues around power alone and explicitly advocat-
ing an emancipatory agenda, FPDA does not presume that women as a social group
category are necessarily going to emerge as powerless. Instead, it views female subject
positions as complex, shifting and multiply located. There are competing discourses of
gender in any given context and the interplay between discourses means that speakers
can continually fluctuate between subject positions on a matrix of powerfulness and
powerlessness. Having a degree of agency, individuals can recognise how and through

xxvi
 Foreword

which discourses they are being ‘positioned’, and can then take up or resist particular
subject positions.
FPDA is influenced by the social constructionist agenda and it seeks to bring together
the micro and macro levels of analysis. It shares with CDA a focus on power and the
ways in which it is negotiated in different contexts. FPDA is eclectic in nature and it
looks into syntax, lexis, prosody, metaphor, topic, framing, visual semiotics, and more,
at micro level. This provides the basis for a wider perspective, looking into a group or a
community of practice and how, for instance, male-dominated or female-friendly they
are. FPDA aims to provide tools for analysing what is going on in core events, particu-
larly in business contexts like for instance the well-studied ‘meeting’, and to address
wider questions such as whether women have a steeper road than men to do leadership
effectively. FPDA is interested in revealing power imbalances but also aims to chal-
lenge theoretical and methodological ‘orthodoxies’ associated with CA and CDA.
FPDA aims to ‘release the words of marginalised or minority speakers’ and entails
giving space to marginalised or silenced voices in localised interactions. Judith’s own
analyses have focused, for example, upon certain girls who say little in classroom set-
tings. Her work has also shown that women leaders are often perceived to use leader-
ship language ‘effectively’ but this perception varies depending on whether and the
extent to which a senior team is ‘female-friendly’.
This ultimately allows for a greater richness and variety of debate, discussion and
freedom to speak from all social groups, not just those who are heard more often. And
this is important in the field of language, gender and sexuality. This means that FPDA
is ultimately concerned with equality but this concern arises from an epistemological
perspective rather than an ideological one. Judith’s work has shown that female speak-
ers cannot be categorised homogenously as “powerless, disadvantaged or as victims”.
To the contrary, female speakers (in positions of leadership or not) have agency to resist
practices that position them as powerless and occasionally, though unfortunately not
always, they succeed.
As well as Judith’s applications of the framework, other studies which have made
use of it include: Budach’s (2005) study of language, gender and community in French
Ontario; Castenada’s (2007) insightful analysis of competing gender discourses in pre-
school EFL lessons in Brazil; Kamada’s (2010) work on hybrid identities in teenage
girls’ conversations; Baker’s (2013) investigations into media representations of gen-
der; Sauntson’s (2012; 2018) studies of the experiences of marginalised LGBT+ class-
room participants.
FPDA is not simply concerned with identifying discourses, but focuses more on how
discourses interact, how they variously position speakers in relation to other discourses
and how speakers shift between subject positions. Judith, for example, examined how
competing gender discourses are sometimes produced in the context of public speaking
in the classroom. Looking forward, FPDA will continue to be used to release the voices
of the marginalised across a variety of settings – a key aim of critical applied linguistics.
FPDA and Judith’s work more broadly played a pioneering role in making linguistic
research relevant and applied for professionals outside the academy. Well before impact
agendas became defined and part of the HE institutional discourse, Judith was con-
cerned with crossing boundaries and contributing to the wellbeing of men and women
particularly by addressing inequality and underrepresentation. This agenda is as timely
and relevant as it has ever been; we are certain many of us will continue following
Judith’s path.

xxvii
Foreword

As well as being a key contributor to the field of language and gender, Judith was a
founding member of the BAAL Language, Gender and Sexuality special interest group,
and was the SIG’s convenor until 2015.
Judith was a fantastic colleague and was always very generous with her time. She
offered invaluable support and advice, provided feedback on drafts of work and was
always available for chats. She was working tirelessly and until the last days she had
that drive and enthusiasm that anybody who has worked with her would recognise.
Judith was a good friend to us and her wonderful sense of humour meant laughing a lot
and enjoying companionship and warmth apart from the stimulating conversations and
exciting debates. These are times we treasure and will miss very much.

xxviii
 1
Introduction
Language, gender, and sexuality: sketching
out the field

Jo Angouri

The LGS journey from essentialism to poststructuralism

A field’s footprint and vitality are typically assessed through volume of scholarly activity,
ongoing societal interest, and significant evolution of thought; language, gender, and sexu-
ality (LGS) is a case in point. From the study of grammatical gender to the descriptive use
of language by biologically defined wo/men, to critical analyses of ideologies and power
asymmetries, linguists turned to gender from the very early days of the discipline as a whole
and in all its branches, from general linguistics to psycholinguistics, sociolinguistics, and
pragmatics. Since the 1990s, sexuality has also become a core part of the field.
The aim of this introduction is to set the scene, providing a commentary on the scope
of this volume, its organisation and contents. I start with a short overview of developments
within the field since its inception in the mid-1970s to establish the rationale for the dis-
tinctiveness of this volume and our decision to take a research methodological focus. I
will sketch a blueprint for the novice or experienced reader. The volume aims to help LGS
researchers to theorise the whole research process from inception to impact with the empha-
sis on the design and affordances of theories and methods. These are not neutral products of
academic labour; they are products of their time and hence need to be contextualised. The
volume provides a permanent resource and complements other works in the field.
Gender is the central node in the LGS triad and undoubtedly a core social category for
lay people and scholars alike. This simultaneous coexistence of first (lay people) and sec-
ond-order (scholars) concerns has been central in LGS scholarship, although not always
addressed in those terms (Ehrlich et al. 2014; Freed 2003). It is, however, a particularly
relevant angle as lay people and scholars engage with the same phenomena but through dif-
ferent conceptual and analytical processes and for different purposes.1 In 1953 Schutz raised
the issue of a mismatch between the common-sense orientation of a group’s first-order con-
structs and the second-order constructs of the expert (social scientist). Schutz writes: ‘The
thought objects constructed by the social scientists refer to and are founded upon the thought
objects constructed by the common-sense thought of man [sic] living his [sic] everyday life
among his fellowmen [sic]’ (1953: 3).

1
Jo Angouri

The ideologies and hegemonies associated with ‘common sense’ have been deeply
embedded in LGS since its inception. Since the pioneering work of Lakoff’s Language and
Woman’s Place (1975), describing, analysing, and critiquing ‘common -sense’ positions,
such positions of asymmetry and struggle have preoccupied linguists interested in gender,
and, more recently, sexuality.
Lakoff’s work is particularly significant, for it firmly established the interconnection
between language, gender, and power struggle; it is justifiably cited as a core milestone in
the establishment of the field. The 1975 publication opens with the following statement:
‘language uses us as much as we use language’ (1975: 3). This, in my reading, foregrounds
the dynamic balance between (sociopolitical) structure and (individual) agency. Lakoff’s
work has been criticised for its methodological approach and essentialist stance (e.g. in
response to the 1973 publication, Dubois and Crouch 1975). She drew on anecdotal data
and a process of contrasting the categories of ‘women’ and ‘men’ as homogenous entities.
Those criticisms reflect the evolution of thinking in our field given that Lakoff’s position
was consistent with the training of linguists of the time and of the political context. Lakoff,
who started as a syntactician, is writing in the 70’s in a political environment where the
feminist movement (women’s liberation or rights movement) (for an overview see Schulz
2017) is raising awareness of the impact of patriarchy in everyday life and is calling on the
oppressed to challenge the status quo by making the private public, and the personal collec-
tive. ‘The personal is political’, as the famous saying goes. Lakoff’s work provided the field
with an agenda which is still relevant (see Lakoff 2004), addressing issues of epistemology,
methodology, and application to real-world concerns.
Early LGS work is usually placed somewhere under the widely cited ‘deficit– differ-
ence–dominance’ axis (typically used to refer to work between the 1970s and the 1990s)
and juxtaposed to ‘discourse’ approaches (the 1990s onwards and under the influence of the
‘discourse turn’ in social sciences – Angouri and Piekkari 2018 for an overview). This time-
line and labelling are analytical artefacts which have been useful in illustrating the field’s
move from essentialism to poststructuralism and the meta-language that continued emerg-
ing as the field developed. Labels, and terminology more generally, are useful artefacts for a
retrospective reading of earlier scholarship; they do not neatly correspond to distinct phases
of the field. They also imply a linearity, which is misleading as the ‘edges’ between, and
within, those stages are fuzzy. They, however, provide a useful chronological order and ease
of reference and help to position developments on a temporal axis.
The period of the field from the 1970s to the 1990s was concerned with explaining styl-
ised linguistic behaviour associated with biologically defined women and men. Scholarship
of the time took gender as a predetermined category on the basis of biological sex and
looked into differences which were associated with the societal expectations projected onto
wo/men. Gender, and sexual orientation to some extent, were framed as social categories
defined on a set list of predetermined characteristics and typically represented as independ-
ent from other facets of peoples’ multiple identities. This conceptualisation of identity
drawing on the enactment of binary realities (e.g. gender seen as either ‘male’ or ‘female’
biological sex) took a rather essentialised approach which became a point of debate with
later studies in the field. While some studies looked at data from the prism of variation in
socialisation (difference), others looked at it as an indication of the patriarchal social order
(dominance). The difference/dominance debate is a product of social and political thinking
of its time and it is well covered in the literature. Chapters of this Handbook will provide
their readings of the relevance of the beginning of the field (see Litosseliti, this volume).

2
 Sketching out the field

At the same time, this early scholarship set the foundation for doing much more than
merely describing linguistic features associated with wo/men; it provided the tools for
looking into the ways in which power is enacted in and through language and challenged
the social ‘consequences when grown women were “girls” and when the masculine pro-
noun was “normal” to refer to everybody’ (Lakoff 2004: 18). The field’s strong orientation
towards making research relevant beyond academia is also noted with publications that
bring the relationship between gender and language to the public eye. Tannen’s You Just
Don’t Understand: Men and Women in Conversation (1990), is a significant landmark of its
time. As the field grew, LGS turned to poststructuralism and embraced social construction-
ism. Scholars criticised approaches to identity in general and gender identity in particular.
I turn to this next.

Identity to desire and beyond

Projecting generalised characteristics onto different groups, be they women, students, or
squash players, is a common topos in categorisation processes. Categorisation is ubiquitous
in daily life and, despite being demonised for stereotyping, it corresponds to dominant mas-
ter narratives that circulate in different sociopolitical contexts, popular media, and domains
of human activity. As such, it is a complex process and one that provides users with useful
‘shorthands’ to an otherwise complex picture. From a second-order viewpoint however,
identity scholars quickly turned to critiquing a static and narrow view of identity on the
basis of fixed and decontextualised characteristics. Instead, they put forward a view of iden-
tity as situated, fluid, and dynamic (Baxter 2003; Holmes 2006). This approach provided
conceptual tools to look into the different layers of identity work, involving both the situated
interactional order as well as the interrelationship with societal, political, and historical con-
texts (Angouri 2015). Seeing identity as a ‘construct’ brought individual agency to the fore
which enabled the field to look into the micro-moment of interaction. This ‘construction’
metaphor however also brought to the fore perceptions of equality in (self/other) identity
ascription – ‘I (the speaker) am empowered to make an identity claim’.
Identity ascription, however, is very rarely down to the agency of the speaker, alone, to
achieve. Depending on where we are, who we are with, the expected performances of our
roles and so on, we negotiate identity claims and navigate differences. What is unmarked
(seen as ‘normal’) in a particular community is associated with a set of behaviours visible
to its member. Each context we find ourselves in comes with power asymmetries which pre-
date the individual encounter. Systems of power circulate and are perpetuated in the media,
work, education, family/caring, etc. Achieving unmarked identities, therefore, is subject to
a delicate equilibrium between the social order and individual agency. Deviating from the
social norms – and social imaginary – has immediate tangible and situated consequences
for the individual. Consider women migrating to countries where gender equality is suppos-
edly greater compared to in the country of origin; Piller (2016: 138) provides a compelling
account of professional women who saw their positioning reverting to the traditional gender
order after migrating to Australia, drawing on research with women from Asia, Eastern
Europe, and Latin America. These women are homogenously positioned and portrayed in
the media as agentless, subordinate to their male family members, and devoid of other pro-
fessional or personal identities. This strips away individual self, excludes women from the
labour market and other symbolic capital (language resource and citizenship), and results in
women being in a worse position compared to their pre-migration state.

3
Jo Angouri

In the early part of the poststructuralist period of the field, the work of the theorist Judith
Butler made a significant impact on gender studies and on LGS in particular. Butler draws
on Austin’s (1962) and Searle’s (1969) theory of performative acts. To Butler, performativity
is an authoritative power discourse created through stylised acts that are repeated; through
the process, a relationship between the individual and the audience is constructed in reaf-
firming or challenging norms which are socially conditioned and in permanent flux. Butler’s
theory of performativity has been significant; it moved the field from focusing on gender
identity through the study of difference to the complex ways in which interactants mobilise
resources in context. This can challenge or perpetuate the social order and dominant norms;
our practices are always read as ‘gendered’ (on the omnirelevance of gender, Butler 1990
and Holmes 1995). The concept of performativity has also been used widely in the field,
sometimes erroneously interchangeably, with the Goffmanian performance. Despite their
both sitting under the poststructuralist spectrum, their interchangeable use causes lack of
conceptual clarity. Goffman’s theory of performance (1959), according to which society
is analogous to a theatre scene where people perform different public acts, is often used in
theorising on gender roles in different domains of activity. Goffman’s emphasis on stigma
and the social cost of non-conforming to societal expectation also found fertile ground in
LGS, and particularly as the field started becoming more inclusive and opening up issues
around, and research on, sexuality.
Although language and gender scholarship had not turned its gaze beyond heteronorma-
tivity before that time, in the 1990s sexuality started becoming a core part of the research
agenda (see Cameron 2005; Livia and Hall 1997). Research on language and sexuality con-
tinued to attract growing interest, first through engaging with lesbian and gay language
(Kulick 2000), and subsequently critiquing binary oppositions and the concept of fixed
sexual identities and turning to an identity-view vs. a desire-view of sexuality at the turn
of the century (e.g. Bucholtz and Hall 2004; cf. Cameron and Kulick 2003). Desire, both
as an abstract concept and as erotic desire, started preoccupying the field. Primarily associ-
ated with psychoanalysis, the turn to desire provided an opportunity to bring to the field
new tools in the transition beyond the structural approach to the study of language, still
dominant for linguists at that time. Psychoanalysis, and specific approaches such as the
Lacanian line of thought, provided a theoretical toolkit and meta-language to move beyond
the legacy of the Saussurian tradition, which is based on the linguistic sign, the relationship
between a signified (concept) and the signifier (sound image or linguistic form). According
to Saussure this relationship is arbitrary (a rose is called a rose by a convention) and social.
Saussure’s work made a significant impact on the development of linguistic thought in the
twentieth century, which is well documented in the field (Sanders 2004). The approach to
the linguistic sign, however, implies a linearity and isomorphism (similarity in form and
structure) between the signifier and the signified and does not address the ideological nature
of the sign. The Lacanian approach (Angermuller, Maingueneau, and Wodak 2014), with
the combined interest in the complex relationship between the language and unconscious
and the concept of desire at the forefront, is a good example of the ways in which the turn to
desire enriched the LGS toolkit. Interestingly, psychoanalysis has influenced certain schools
of discourse analysts (see, for example, the Essex School, Glynos et al. 2009) but not oth-
ers within the LGS disciplinary spectrum and hence provides an opportunity for theoretical
development.
Overall, although the identity vs. desire focus was read as a rather polemical debate
between the two approaches, it actually marked another firm shift in the field from static
categories – this time, gay/lesbian – with all the limitations associated with homogeneity,

4
 Sketching out the field

to a more dynamic engagement of identity and desire as macro-concepts looking into the
affordances and limitations in separating sex from gender. Some of those issues were not
new; for instance, Davies (1990: 501) made a case for using the term sex/gender to chal-
lenge the distinction between ‘physical’ and ‘social’ and writes on the aspiration to turn to
‘an understanding of the person as in process, and of words coming not from an essential
core of the person but from the discursive practices through which the person is constituting
themselves and being constituted’. Some 30 years later Fausto-Sterling (2019) uses ‘gen-
der/sex identity as an autopoietic [self-reproducing] system’ and a process emergent in and
through the situated encounter. Nevertheless, zooming in on sexuality and engaging with
queer linguistics enabled the field to achieve a deeper and fuller engagement with diversity
and nuances of language practices in different contexts and the ways in which femininities
and masculinities need to be decoupled from a one-to-one relationship with biologically
define wo/men and from sexual desire. Ensuing research on transsexual communities and
individuals (Tanupriya and Pannicot 2018 for a recent overview) has looked into the pro-
cess by which hegemonic masculinities and femininities are reproduced as wo/men claim
‘normality’; a condition for being recognised and accepted as belonging to the powerful
majority; in Goffmanian terminology a condition for ‘passing’ in a hegemonic heteronor-
mative world. Passing involves concealing–constructing an ‘acceptable’ persona to gain
recognition and avoid stigma.
As individuals claim membership in communities, professional or social, behaviours
need to be indexed and ratified as ‘one of us’ for membership to be granted. This is, not
exclusively but largely, enacted linguistically. The axiomatic position of discourse as
socially constitutive, socially conditioned, and socially consequential (Wodak and Meyer
2009) is well aligned with practice theories in general and with a framework that this line of
work, and the whole of the LGS field, has fully embraced: namely communities of practice
(CofP) (Lave and Wenger 1991). The framework provided scholars with a way to study
language-use locally and to link the micro-context to the broader socio-cultural environ-
ment and the range of practices that warrants membership to a community. It usefully put
the spotlight on the processes by which communities emerged through membership, instead
of on the community as a whole.
Eckert and McConnell-Ginet (1992: 464) in an early influential definition refer to a
CofP as:

An aggregate of people who come together around mutual engagement in an endeavor.

Ways of doing things, ways of talking, beliefs, values, power relations – in short, prac-
tices – emerge in the course of this mutual endeavor. As a social construct, a CofP is
different from the traditional community, primarily because it is defined simultaneously
by its membership and by the practice in which that membership engages.

The framework provided a useful vocabulary for describing the relationship between ‘cen-
tre’ and ‘periphery’ and the multi-layered relationship between social structures and human
agency. These strengths largely account for its popularity (for a critique see Amin and
Roberts 2008).
Overall, poststructuralist work highlighted further the importance of multiple identities
which interest and continuously interact with one another, forming complex matrixes that
take different meanings in different contexts. Intersectionality (e.g. Acker 2006; McCall
2005), has influenced feminist studies significantly (e.g. Brah and Phoenix 2004), as it con-
vincingly showed that individuals are (self/other) positioned in more than one category

5
Jo Angouri

simultaneously. So a professional may be placed in the category, as an example, ‘woman’

as well as ‘young’, ‘old’, ‘Black’, ‘white’, ‘non-native speaker’, ‘with/without children’,
and so on. Individual lived reality cannot be understood without addressing the com-
plexity of multiple identities and the corresponding social hierarchies and hegemonies.
Post-structuralist scholarship emphasised the ‘multitude of dynamic power relations and
the historically and socially constituted mosaic of intersecting differences’ (Metcalfe and
Woodhams 2012: 134). Drawing on intersectionality enables researchers to show how risks
of exclusion increase from the intersections of race/ethnicity, class, gender, sexual orienta-
tion, age, body ableness, etc. rather than being an effect of one single mark or category of
difference (Harcourt, Icaza, and Vargas 2016). This, evidently, does not mean that individu-
als are passive bearers of those identities; the same research has provided evidence that
individual agency can succeed in challenging, and sometimes changing, the status quo.
Further, how and under what conditions the status quo is perpetuated or subverted became
relevant to studies looking into the relationship between language, gender, and sexuality in
different domains of professional and economic activity. This was particularly the case in
one disciplinary subfield, often labeled as ‘workplace sociolinguistics’, which emerged and
grew, drawing on the work of researchers in different parts of the world from the 1970s to
the 1990s (Holmes and Stubbe 2003 for the genealogy of language in the workplace).

Gender, sexuality, and the workplace

‘Work’ and the ‘workplace’ are at the very heart of social mobility, economic stability and
growth, mental and physical well-being for the individual, and also a site of struggle where
power hierarchies are negotiated, perpetuated, challenged, and resisted. Looking into differ-
ent professional contexts has been significant for putting the magnifying glass to the lived
experience of professionals who struggled with gender order in their daily lives.
Linguists turned to the workplace and looked systematically into the subtle ways in
which exclusion from power centres has been enacted in daily life through gendered prac-
tices since the 1990s. Drawing on the ethnographic tradition and the analysis of interac-
tion (see below), scholars focused on the situated micro-moment in exploring the ways in
which abstract ideologies are made relevant. Using recordings of non-researcher–elicited
authentic workplace interaction, sociolinguists produced detailed descriptive and interpre-
tive accounts of the strategies employed to sanction and exclude from participating in meet-
ings and decision-making (e.g, Baxter 2010).
Leading sociolinguists and critical scholars (Holmes 2006; Holmes and Marra 2004;
Holmes and Stubbe 2003; Tannen 1994; Wodak 2005) provide robust evidence on gendered
practice in professional settings ranging from corporate and blue-collar to institutional and
public bodies in different sociopolitical contexts which lead to persistent patterns frequently
reported in the media, e.g. the gender pay gap, absence of female (and LGBTQ+) profes-
sionals in leadership positions, persistence of sex-segregated occupations, and so on.
Even in the case where female professionals reach senior positions, studies on voice
in the workplace (Baxter 2010) have shown that women are consistently undermined and
‘often ignored, interrupted, mocked, criticised or not given a space to speak’ (Baxter 2018:
405). Baxter refers to ‘double-voicing’ as a strategy women use in pre-empting criticism and
attempting to interactionally achieve recognition. The lack of parity of esteem and parity of
reward between the powerful majority and those who deviate from the norm is consistent.
Occupations that are typically feminine are seen as having lower status than those regarded
as typically masculine and those seen as ‘different’ are typically paid less for the same jobs.

6
 Sketching out the field

More generally, workplace sociolinguists (Holmes 2006) and applied linguists (Hua
2014; Piller 2016) have shown that common ideals of the global workplace, such as effi-
ciency, growth, and global value, also come with narratives that re-/emphasise gendered
normative behaviours. Work on hegemonic masculinities (Connell 1987; cf. Elias 2008;
Plester 2015; Wetherell and Edley 1999) has explored transnational normatively mascu-
line business ideals (e.g. competitiveness based on narratives and metaphors of aggressive
behaviour, gendered division of labour, etc).
Linguists write systematically for a general audience (Baxter 2014; Cameron 2008),
raising awareness and challenging ‘the straightjacket of stereotypes which have generated
apprehension and suspicion of women who do not conform to the expected mould’ (Holmes
2006: 220). All this work aimed to contribute to positive change – well before the highly
valorised current ‘impact’ agenda (see Cameron, this volume).
These developments in the second order, however, did not correspond to major shifts in
first-order conceptualisations of gender. Although empirical research has repeatedly shown
that ‘gender lines are anything but clear-cut’ (Holmes 2006: 24) the ideology of ‘difference’
is still prevalent, especially in popular literature, and has been resilient to change (Freed
2003). In the workplace, for instance, despite years of empirical work and policy interven-
tions little seems to have changed (for a recent account, Holmes 2020).
Regimes of difference, in their turn, continuously reaffirm the power status quo of those
in majority and ratify the social order. Power imbalance is perpetuated on boundaries and
binaries – such as majority/minority, male/female, young/old, and so on. Discourses of gen-
der difference, sexism, ageism, or racism influence and affect meanings, actions, and points
of reference through representations in media, literature, and everyday stories we tell and
are told. Wodak in 1997 (p. 12) writes:

in our societies biological sex is still used as a powerful categorization device […] in
many occupations women are still paid less than men for the same achievements and
positions. In these contexts, biological sex as a ‘natural factor’ is still salient and cer-
tainly not only a variable social construct.

In 2006, this is echoed by Holmes, who argues that ‘many professional workplaces are still
rife with examples of systemic gender discrimination in which language plays an important
role’ (Holmes 2006: 53). Fast forward to 2020, these statements remain true – although
social media and the affordances of technology may be providing new spaces for resistance.
I will revisit this at the very end of the chapter.
To summarise the discussion so far, although the fluidity of identity and significance of
individual agency has generated rich scholarship in our field, the first-/second-order gap
requires us to revisit our agenda. It also has theoretical implications for our engagement
with the essentialism–constructionism binary. ‘Strategic essentialism’ has argued for some
time now that accepting that women (or any oppressed group) exist as a category allows
work on behalf of the rights of the group; if on the other hand, there is no group we can label
‘women’ (or any other), and no acceptance that members of the group share at least some
characteristics, then the group level is abolished and the spotlight is on the individual. This
allows discriminatory and sexist practices to continue. The anti–anti-essentialism move-
ment has taken on some prominence, but not as much as one would expect, in recent LGS
debates (Stone 2004). Challenging the status quo has certainly been on the agenda of many,
if not all, contributors to this volume. Although there is evidence of change, a societal shift

7
Jo Angouri

is yet to come. Whether and to what extent our field should act in an activist and politi-
cal manner is a matter of debate. I discuss this further in the context of current topics and
debates below.

Ongoing debates and re/visiting LGS’s political agenda

Various ‘turns’ have kept appearing in social sciences since the turn of the century and
LGS continues to grow and to diversify its agenda. I have been sceptical on labelling
scholarly interest as a ‘turn’ which implies collective action and paradigm shift (Angouri
2018a); however certain areas undoubtedly grew in prominence. This includes materiality
and embodiment, which I here touch upon briefly, as the reader will find these concepts
under future directions in many chapters in the volume (e.g. Leppänen and Tapionkaski;
Mackenzie; McLoughlin).
As the field’s interest turned to understanding identity holistically, scholarly attention
also started to emphasise materiality and the body. The influence of the Bourdieusian work
on habitus – embodied dispositions that are learned and are stable (but not static) – draws
attention to the ‘somatization of the social relations of domination’ (Bourdieu 2001 [1990]:
23), and Butler’s critique of the biologically predetermined ‘sexed body’ emphasises body
as a social product; these are examples of work that contributed to a shift from the body
being divorced from or secondary to speech, to being, progressively, co-constitutive. This
shift also drew on theoretical and methodological tools interaction analysts have developed
over the years. Conversation analysts and interactional sociolinguists have been looking
into interaction as embodied enactment from the start of the field (Goodwin 1981; Heath
1986). Influenced by ethnography and ethnomethodology, researchers have looked into how
semiotic resources of meaning are sequentially organised in interaction and, through that,
how the social order is re-/enacted. Early interest in conversation analysis (CA) on gestures
and gaze are indicative of the significance attributed to multiple resources of meaning.
As Goodwin and Goodwin (2004: 240) note, ‘multiple parties build action together while
both attending to, and helping to construct, relevant action and context’. The importance of
video analysis is well documented in the field (Mondada 2008; Rusk et al. 2015). Although
most earlier studies draw on audio material this is often because of limitations of access
and confidentiality rather than the researchers’ disposition. Nevertheless, current writing
(Mondada 2019; this volume, Evaldsson; Greco; Mortensen and Milani), re-emphasises
interaction not as the product of disembodied actors but as a process of interacting bodies
which co-construct meaning. The CA move from an interest in gestures (Goodwin 1986;
Streeck and Hartge 1992) to gestures in interaction to embodiment, is a good example of
a shift from looking into specific features, to a set of interactional practices, to position-
ing practices in the embodied material environment and approaching them from multiple
angles. This simultaneous zooming in and out can be described as a shift from a 2D to a 3D
perspective in the field, the latter being a metaphor for depth and complexity.
Overall, a renewed emphasis on the body is noted in the last ten years, responding to
calls for more work on the relationship between language and embodiment in sociocultural
linguistics (Bucholtz and Hall 2016), applied linguistics (Canagarajah 2018), variationist
sociolinguistics (Zimman 2018), and metaphor and critical discursive analyses (El Refaie
2019). As brief illustrations, variationist studies turned to voice as a site of identity strug-
gle (Levon/Zimman), moving the field from seeing ‘sound’ as separate from the socially
contexted body. Current work on trans voices (e.g. Campbell-Kibler and miles-hercules,

8
 Sketching out the field

and Zimman, this volume; see also Zimman and Hall 2010 on the third sex) provide good
examples of ways in which positions discussed earlier on identity and sexuality are being
further elaborated. Studies have expanded work on ideology and body perception and rep-
resentation – see, for instance, Mason (2012) on the difference in discrimination systems
against obese wo/men; and applied linguists are calling for a general turn to the situated and
material locus of the communicative encounter which is spatiotemporally conditioned and
contextualised.
A firm focus on the body and the material environment within which speakers operate
expands and extends the foci in the LGS project. Despite the growing interest, this is not dis-
sonant with the approach of classic work influenced by ethnography and anthropology and
particularly the Hymesian tradition to the communicative event. This pays attention to ‘the
various available channels, and their modes of use, speaking, writing, printing, drumming,
blowing, whistling, singing, face and body motion as visually perceived, smelling, tasting,
and tactile sensation’ (Hymes 1964: 13). We know in our work that communication involves
more than disembodied voices, gestures, or static texts; meaning-making is achieved in and
through interacting bodies who engage in contextualised, situated encounters and are sub-
jected to the systems of power that predate them – and which require more than individual
agency to be challenged and changed.
The embodiment agenda is tightly aligned with the work of critical scholars and multi-
modal discourse analysis. Multimodality constitutes a field comprising different approaches
and schools of thought, drawing on classic work from Barthes (1957) and Saussure (1974
[1916]) to Halliday and Hasan (1985), and has been strongly influenced by systemic func-
tional linguistics (Halliday and Hasan 1985). Scholarly research led to the emergence of
multimodal discourse analysis (O’Halloran 2011) as a concrete field, pushing from a focus
on language to other resources, including embodied interaction and all other aspects of life,
from architecture and music to fashion and advertising. And despite the various schools of
thought, there is agreement that choices people make drawing on all semiotic resources is
what makes meaning. Multimodal discourse analysis has been particularly influential for
showing meaning-making practices that perpetuate the gender order (Lazar 2005; Leppänen
and Tapionkaski, and McLoughlin, this volume) across media and domains of activity.
Evidence over the years and to date shows how the relationship between language and
semiotic representations of oppressed groups perpetuates the power hierarchies in socie-
ties. Caldas-Coulthard’s recent collection (2020) provides ample evidence of how sexism is
deeply embedded in all spheres of human activity and is reaffirmed in politics with a small
‘p’ and with a capital ‘P’ (Wodak 2015).
Overall current scholarship makes a strong case for multilayered, and accordingly
multimethod, enquiry which translates to an agenda influenced by studies on materiality,
embodiment, and multimodality. Beyond the areas discussed above, a turn to the senses and
multisensoriality is also emerging in the field (e.g. Mondada 2019). This is in line with the
move towards ‘languaging’ in applied linguistics. Languaging pushes away from a purist
view of language and of a ‘perfect’ idealised speaker to the multiple ways in which indi-
viduals draw on resources of meaning to achieve their goals: a shift, in other words, from
a structuralist view of language to one of agency. The current trend towards posthumanism
is furthering the post-structuralist tradition in its quest to redefine the human and the non-
human (Pennycook 2018). Work on the relationship between the entire body, other bodies,
and object manipulation (Day and Wagner 2019), although from a different theoretical angle
(primarily ethnomethodology), raises important questions on the agency of the individual
and the in/external boundaries of the body and the material surroundings.

9
Jo Angouri

As evident from even this cursory glance at our field, the balance between structure and
agency has, and will, undoubtedly continue to preoccupy our agendas; and although the lat-
ter is unquestionably significant, as linguists we are well aware that language, in first-order
representations, corresponds to rather static national varieties and comes with the full set
of problematic associations attributed to the (non)standard – mapping exactly onto the mis-
match in first/second approaches to the gender order. This gap is again significant.
Piller (2016) provides useful case studies of migrant communities in Australia and the
ways in which migrant women are linguistically excluded from and marginalised by major-
ity language – and, by extension, capital, in a Bourdieusian sense. Critical discourse analytic
research on migration (e.g. Wodak 2013) has produced similar results showing how migrant
communities are excluded from the social web around them through limited or no resources
in their home language and a systematic representation of the ‘other’ as deviant and dif-
ferent to ‘us’. Particularly in countries where a ‘banal national gender equality discourse’
(Vuori 2009: 218), according to which, gender equality, represented as harmonious and
already achieved, is part of the national myth, migrant women are systematically positioned
as oppressed and marginalised. Gendered ideologies are reproduced in multilingual contexts
and language becomes a convenient arena for differentiating between those who are to be
allowed in from those who need to remain out. Studies drawing on the workplace (e.g.
Duchêne et al. 2013; Franziskus 2017; Piller and Lising 2014; Piller and Takahashi 2010)
provide, again, further evidence on the significance of language hierarchies and the ways in
which migrant women are exploited, either for having, or lacking, language skills in major-
ity languages. Research on skilled migration has provided consistent evidence. The deep
paradox in all programmes designed to empower the ‘powerless’ is that their very existence
reinforces the world’s vertical and horizontal ‘centres’ and ‘peripheries’.
To sum up, current work is pushing the boundaries beyond the field’s traditional (and
expected) interest in language as a distinct variety. At the same time, as linguists we need
to keep first-order implications on the radar while also turning a critical eye on our own
repertoires. Further, there is a need to look into the language practices that dominate our
field and shift our gaze from mono- to multi-lingual analyses. In line with most areas in
social sciences, the dominance and status of journals and studies published in English is
unquestionable. Adding to that, studies from particular linguo-geographical environments
draw on samples that cannot cover the diversity of the lived experience and the systems of
power that are implicated when gender and sexuality are made relevant and negotiated in
different sociopolitical contexts – and when language/s come/s into contact. There is a need
for more studies looking into the implications, impacts, and ways in which speakers draw
on the language varieties accessible to them. Our field has been committed to inclusion and
to moving beyond the ‘global North’ (McElhinny 2007). Considerable effort is made by
scholars (Atanga et al. 2012; Bassiouney 2015; McElhinny 2007), and by journal and series
editors. However, there is a long way still to go to achieve plurality of voice. Research in
general – and this does not exclude LGS research – is political; journals, publishers, and the
academic community operates within set frames of reference that exclude those who are not
familiar with our norms. The well-known ‘language penalty’ (Roberts and Campbell 2006),
indicating the price for those who sound and behave linguistically different to the majority,
also applies to the academy.
This takes me to the last point I wish to cover here, a call to revisit our field’s com-
mitment to action and the ways in which we can achieve change. Scholars over the years
have made a case for LGS to ‘never cease to engage actively with and challenge assump-
tions about gender norms and loudly draw attention to the way power, privilege and social

10
 Sketching out the field

authority interact with and are naturalised as properties of independent social categories’
(Holmes and Meyerhoff 2003: 14), while others have argued that the shift to identity and
‘post-feminism’ (for a reflective discussion see Litosseliti et al. 2019) has also brought a
shift from the clearly political ‘what is to be done?’ questions to ‘who am I?’ (Cameron
2005, 2009), moving the focus to the individual. Lazar (2009: 397) expresses the same con-
cern about the shift from, in her terms, ‘we-feminism’ to ‘I-feminism’.
Despite difference in views, all linguists agree that language is political and that language
remains a powerful arena for ideologies to be ratified or resisted. In order, however, to bring
the change we want to see, collective political action is necessary; and it is the case that,
despite effort, volumes of studies, and evidence, our research has not reached, and is not
known to, those who can most benefit from it.
Academic writing is a genre that serves our community but is not designed for public
consumption; and equally, expecting that our books, as artefacts, are enough to bring change
is clearly not supported by any evidence of uptake. We therefore need a voice for making
our work known beyond our own bubble; without going back to essentialising in gender and
sexuality, there is a need to revisit the ‘difference’ they (still) make – as Cameron (1992)
argued a very long time ago. This also implies training for early career scholars and young
academics to engage with media and government, which is very far from the primary tone of
voice that we consider appropriate in our professional worlds. Cameron (this volume) nicely
critiques the implications of that, and provides us with a challenge to consider what it really
takes to make this transition.2 I discuss this further below.
Finally, equipping those coming new to the field with the theoretical and methodologi-
cal tools that we have developed, in order to enable them to move beyond traditions and
orthodoxies, is necessary for studies to provide new models and perhaps really move us to
a ‘post-difference’ era. This is directly related to the rationale and the motivation for this
volume, which is the next and last section of this introduction.

LGS theories, methods, and the ‘inter/multi/post-disciplinary’ turn

LGS is characterised by a rich diversity of approaches which shape and even change the
ways in which the field is conceptualised and approached in our subdisciplinary areas.
Paradigm shifts have deeply impacted the way we frame and study the core notions and
associated phenomena. There is a very rich scholarship, including handbooks (Ehrlich et al.
2014) which over the years have provided a landscape of the field and its core areas. This
Handbook is taking a distinctively different approach from previously published editions
by focusing on the methodologies and theoretical frameworks upon which LGS constructs
itself as a field.
In presenting the structure and content of the volume, there is perhaps space to coin yet
another turn which undoubtedly is noted across the whole of the LGS disciplinary spectrum;
the ‘inter/multi/post-disciplinary’ turn.
Crossing disciplinary boundaries became a new ‘ideal’ of good research, with funders,
journals, and research assessment encouraging researchers to transcend disciplinary bor-
ders. This is reflected not only in publications (e.g. Frodeman et al. 2017) but also the media
(The Guardian, January 2018). Interdisciplinarity goes hand in hand with mixed and multi-
method design, to which we will return; in fact the latter is a sine qua non of the former.
Venturing beyond linguistics has been naturally compatible with the foci of LGS researchers
and collaborations with media, political sciences, public health, migration studies, theatre
studies (to name but a few) bringing useful crosspollination of ideas.

11
Jo Angouri

As post-structuralism put emphasis on the multifaceted relationship between language,

identity, and the body, and approached language as the way in which realities are enacted,
negotiated, and challenged, empirical studies also drew on and problematised the LGS theo-
retical and methodological toolkit. This includes significant methodological pairs such as
researcher/user, quantitative/qualitative tools, self-reported data/ethnographically informed
approaches. All these binaries corresponded to a movement from quantitative/variationist
study of gender and sexuality as distinct variables towards qualitative and discourse-ana-
lytic, mixed-methods approaches in line with the theoretical developments discussed earlier.
While quantitative, qualitative, and mixed methods continue to be used in parallel, there
has been a shift from a purist to a multi-method approach (Angouri 2018b). In the context of
inter/multi/post-disciplinarity, exploring interconnections between different methods which
have different standings in their own disciplinary context was a necessary step for diversi-
fying the LGS foci. Mixed methods, the current main ‘modus operandi’ for research in the
field, comes with its own affordances and limitations (Angouri 2018b); the bottom line,
however, is that multi-method designs are useful for allowing us to explore more angles of
a social phenomenon, but they are not a panacea.
The position we took in this Handbook, from the outset, is that there is no one theory
or method that can give answers to all questions. Theories and methodologies constitute
tools employed by researchers in the process of accessing and studying multifaceted phe-
nomena. Their ‘value’ is dependent on the context of their use. At the same time, not all
methodologies are equal, and different journals, disciplinary fields, or departments and
courses have preferences as to what ‘good’ research looks like. We are all well aware
of the resilience of the quantitative/qualitative divide and the stereotypical representa-
tions of ‘hard’ evidence; and irrespective of how open-minded we want to be on research
designs, we no doubt still teach our students on the basis of a quan/qual axis that matches
our own, and our departments’, dispositions. There is nothing wrong with a researcher’s
own preferences; to the contrary, they are important resources for research designs. In the
spirit of collaborative learning and sharing good practice however, more work is needed
in articulating the implications for LGS projects and opportunities for addressing old
questions with new tools.
Accordingly, one of the aims of this volume is to challenge purist research ideology,
draw on the experiences of multiple projects, and provide a collective reflexive resource for
the affordances and limitations that come with all the established tools and approaches in
our field. In line with the recent recognition of the importance of reflexivity which involves
‘constant interpretation, positioning and realisation of the researcher’s influence on the pro-
cess’ (Angouri 2018a: 85; see also Angouri, Marra, and Dawson, this volume), we intention-
ally provided the leaders and contributors of each part with the opportunity and the space to
position themselves in the field and draw on their selected methodological approaches in a
way they felt is best, and allowed their voices to be heard; we hope that the reader will ben-
efit from the collective experience and debate between different methodologies and schools
of thought.
In more detail, we have capitalised on the unusually wide repertoire of methodological
approaches within LGS as the key conceptualising and structuring device for this Handbook.
As the field has come of age and after years of healthy academic debate on ‘which meth-
odologies are best’, we take an anti-orthodoxy stance here; different approaches provide
access to different layers of meaning and the same data can be read in contrasting ways.
What is ‘right’ or ‘wrong’ depends on the purpose of the enquiry, the researcher’s deci-
sions, and the audience – among other considerations. We focus here, then, upon what each

12
 Sketching out the field

approach allows for, can reveal, and makes possible within LGS research, as well as what
it is less equipped to do.
The theme of this Handbook has been addressed in a publication (Harrington et al. 2008)
charting ‘gender and language research methodologies’ that followed a British Association
of Applied Linguistics (BAAL) special interest group event in 2005. The publication
brought to light the plural and often competing versions of LGS that circulate in feminist
linguistic scholarship. Our Handbook updates, extends, and develops this earlier work, aim-
ing to represent and capture current thinking in our field and theorise the whole research
process from inception to impact.
The Handbook comprises eight parts, each consisting of a set of chapters.
The core theme areas include well-established, highly utilised approaches from vari-
ationist LGS and critical discourse analysis to conversation analysis and corpus linguistics.
We attempt to provide different audiences with a highly flexible volume which comple-
ments all other publications in the field. Each part contains a chapter by leading scholars in
that area, positioning, through their own work and chapters in their part, current state-of-
the-art and future directions. The volume is designed as a series of mini edited collections,
enabling the reader, and particularly the novice reader, to navigate, take a zig-zag approach,
and discover new ways of approaching LGS.
Contributors have provided a concise and critical discussion of their preferred meth-
odological approach in the context of their research topic and/or specific research interests.
They provide a brief explanation of their methodological approach and why they deem it
to be best suited to their research, model the approach by means of an empirical study, and
discuss applications of their approach to the real world. As such, we hope the Handbook is a
resource for all and will become a useful point of reference. I present, briefly, the eight parts
of the Handbook below.
The Handbook is designed to be essential reading for undergraduates, postgraduate stu-
dents, researchers, and scholars with an interest in gender and sexuality but also in sociolin-
guistics and applied linguistics more broadly. It will be useful for postgraduate students on
programmes in related areas of linguistics (e.g. forensic linguistics, CDA) and it will also be
of interest to postgraduate students and scholars in the social sciences more generally and in
other fields with an interest in gender and sexual identity, such as gender studies, women’s
studies, cultural studies, communication studies, and social psychology.

The Handbook’s parts in detail

Variationist approaches
Part I assesses the significance of variationist approaches to the LGS field, indicating how
early research in the 1970s and 1980s originated in the broader field of sociolinguistics,
which viewed gender as one social variable against which to measure language use in large
populations. While much LGS research is now qualitative, this part represents recent vari-
ationist research, which has moved away from large-scale, quantitative, correlational meth-
ods towards more local, contextualised, and ethnographic approaches that explore gender
as intersecting with other social identities such as class or sexuality within particular com-
munities of practice. All the contributing chapters argue in favour of non-binary approaches
to gender and sexuality and cover topics ranging from voice characteristics (creaky voice,
Eckert and Podesva; gendered voice in transgender speech, Zimman) to pitch variation
(Levon) and performances of hyper-masculine styles (Campbell-Kibler and miles-hercules).

13
Jo Angouri

Anthropological and ethnographic approaches

Continuing to another significant area of work for the field, Part II represents the range of
ways in which ethnographic approaches are used in LGS research from ‘full’ ethnography,
involving complete immersion and participation for a long period of time in a given com-
munity, to semi-ethnographic approaches involving partial immersion where standard quali-
tative methods are used, such as observation, field notes, and interviews. This part covers
a broad range of contexts, from communities with non-normative gender identities in the
US and India (Hall and Davis; Nagar), to the investigation of gender in UK political institu-
tions (Shaw), women’s hockey teams (Clark), Tongan courtrooms in Polynesia (Philips),
and YouTube ‘camgirls’ (Varis); all chapters demonstrate ethnography as a multi-method
research approach and ‘enable us to see the workings of language, gender, and sexuality
more clearly’ (Hall and Davis, this volume). The part also reviews research that supplements
ethnography with other qualitative methods: for example, that which incorporates the ‘dia-
chronic’ use of ethnography with more ‘synchronic’ methods such as discourse or conver-
sational analysis. The part makes clear that qualitative methods are not exclusively used in
ethnographic designs but may be mixed with any of the methods presented in the Handbook.

Interactional sociolinguistic approaches

Turning our gaze to interaction, this set of chapters shows how the Gumperzian legacy
has influenced thinking and provided a set of tools that are still widely used. Part III fore-
grounds professional discourse as a site of considerable interest to LGS researchers for
‘indexing’ gender and sexuality relations and identities. The introduction and three chapters
draw mainly from workplace sociolinguistic research and institutional interaction, and pro-
vide an overview of the theories and methods that characterise work in this field (Tannen
and Gordon), as well as the affordances and limitations of interactional sociolinguistics
(IS) as a theoretical framework for investigating gendered talk at work (Angouri, Marra,
and Dawson). Chapters in this part focus on institutional contexts (leadership enactment,
Schnurr and Mohd Omar; normative femininities and masculinities, Angouri, Marra, and
Dawson) and stand-up comedy performances for different audiences (Seals), illustrating the
benefits and challenges associated with applying an IS approach in the study of language
and sexuality.

Ethnomethodological and conversation analytic approaches

Staying in interaction analytic work, we explore the value of CA for LGS and feminist
research, and more generally for any research concerned with the construction of gender
identities and relations as they are reproduced in human interaction. Part IV touches on the
debate between those scholars who consider that CA is ill-suited to this feminist agenda, and
those who consider that its methodological principles and its technical tools can be deployed
within a feminist framework, and, in fact, are better than other approaches. Chapters draw
on telephone calls of female victims of violence (Tennent and Weatherall), performances in
drag king workshops (Greco), families’ conversations on gender and sexuality normativi-
ties (Ericsson), and girls’ gossip disputes (Evaldsson); all chapters together illustrate ‘the
importance of the sequential organisation of talk for how gender is made locally relevant by
the participants’ (Mondada).

14
 Sketching out the field

Sociocultural and critical approaches

Zooming out and turning towards the political agenda, Part V reviews a range of studies
that conceptualise LGS through the lens of critical theory, the heritage of which lies in post-
Marxist and cultural theory. This part discusses the emergence of critical discourse analysis
(CDA) as the predominant critical approach to LGS, and considers why it has adopted a
‘feminist’ (FCDA) version. It reviews why there has been so much debate between those
LGS scholars taking a ‘bottom-up’, ethnomethodological stance through the medium of CA,
and those taking a ‘top-down’, critical-contextual stance (for a discussion, see Litosseliti),
and touches upon the relationship of language and gender research to the, now all-powerful,
‘impact agenda’ (Cameron). The part also includes other critical approaches such as textual
trajectories (Ehrlich and Romaniuk) and queer theory (Sauntson), the latter viewed as a
theoretical approach that utilises methods from other paradigms, as well as the combination
of critical approaches with corpus linguistics which can enhance and support CDA (Murphy
and Palma-Fahey).

Poststructuralist approaches
Continuing this line of argument, the following part (Part VI) explores the origins of posi-
tioning and poststructuralist approaches as well as their key principles and value in relation
to LGS (Norton). It outlines how (feminist) poststructuralist approaches connect with, but
fundamentally differ from, critical approaches and presents studies that model various meth-
ods of data collection and analysis. The chapters of this part draw mainly on online envi-
ronments and news media to analyse motherhood (Mackenzie), the burkini ban (Weedon
and Hallak), heterosexual desire (Mortensen and Milani), and the discursive constructions
of women leaders (Appleby), while Al A’Ali investigates the language of Middle Eastern
women leaders, drawing on meetings and interviews. In line with many of the parts above,
there are chapters on research studies that combine positioning/poststructuralist approaches
with ‘supplementary’ approaches such as IS in order to produce rich, multidimensional anal-
yses of LGS data. We consider this indicative of the multi-method disposition of the field
and the way forward under the interdisciplinarity principle.

Semiotic and multi-modal approaches

Going into more detail on the analysis of multiple sources of meaning, many of the above
approaches are used by LGS scholars to analyse naturally occurring talk or spoken discourse,
and others, such as CDA, are primarily used to analyse written texts. Part VII focuses on
the range of methodological approaches used for multi-modal, digital, or ‘computer-medi-
ated communication’ texts such as multi-modal analysis, semiotics, critical social semiotic
analysis, and genre analysis (for a discussion, see Lazar). This part reflects on how fit for
purpose the range of current methods are (often borrowed from the analysis of spoken or
written discourse) in order to understand gender in multimodal and discursive contexts, and
the possible need for more theorised methods. The chapters in this part look at semiotic
representations on social media (Caldas-Coulthard; Leppänen and Tapionkaski) and mul-
timodal advertisements (Kohrs and Gill; McLoughlin), and make a case for a systematic
analysis of (a) visual resources in critical readings of (institutional) discourse, and (b) the
intersection of gender and sexuality with other identities.

15
Jo Angouri

Corpus linguistic approaches

Finally, the last and eighth part of the Handbook explores the rise and significance of corpus
linguistics as a field and its beneficial combination with CDA for LGS research. It dem-
onstrates how corpus techniques (e.g. word lists and concordances, Baker and Brookes;
collocations, Taylor; XML mark-up, Potts and Formato; and wild cards and lemmatisation,
Hunt) can be fruitfully applied to a range of gendered data as well as specifying the limits of
this approach. The chapters discuss whether corpus linguistics is primarily a data collection-
and-analysis tool or a theoretically informed methodology, and show the affordances and
limitations of their tools for new and experienced researchers.
Together the eight parts of the book benefit from the collective experience of all the 55
colleagues who have contributed. We aspire to providing the tools and the experience that
will make our current students and future graduates, our colleagues from social sciences,
and ourselves better equipped to provide more evidence and a louder voice for making a
positive contribution to the society around us.

Concluding remarks
The time of writing of the final version of this introduction, summer 2020, coincides with
the COVID-19 global pandemic and a global death toll of over 800,000 (as of 23rd August
2020). The COVID-19 impact has starkly revealed the deep inequalities in social structure
with sharp differences in risk to life, poverty, mental health, career prospects disproportion-
ally affecting all minority groups in our societies.
Going back to the hegemonic gender order and gender roles: despite years of scholarship
and activism, reports that have been published from the start of the pandemic (February
2020) show a clear and steady increase in the gap between men and women in unemploy-
ment, work hours, domestic labour, and care demands (Collins et al. 2020; Power 2020). This
is consistent with reports in the media (e.g. BBC, July 2020; The Economist, June 2020).
Women are still the carriers of the bulk of caring work, and are societally expected to pri-
oritise domestic labour. They are, in the majority of cases, the victims of domestic violence,
receiving less support due to disrupted services. School closures have dire consequences for
girls that are exposed to increased sexual and other forms of exploitation (UN Policy Brief
2020) and for boys that are expected to prioritise income generating activities over educa-
tion in many countries, particularly beyond the dominant ‘West’. Similarly, data on sexual
minorities report disproportionate risk associated with access to health care, stigma, poverty
and psycho-emotional wellbeing, and, for LGBTQ+ communities, the risks elderly individu-
als face are heightened in the pandemic due to lack of access to public health and to stigma.
Apart from the fatal consequences and worrying long-term impact, as LGS scholars we
are presented with a need and impetus to revisit our societal impact, the ways in which we
can and should react to the disruption we are witnessing, including the implications for the
future LGS agenda. Although the chapters in the Handbook have been completed prior to
the pandemic, we are providing the reader with an overview of key topics in the areas that
constitute the main DNA of our field, illustrating how theory is connected to real-world
problems, and we hope to provide the tools for innovative multi-method enquiry.
In order to achieve change we need to bring together the voices of academics, practition-
ers, and policymakers to discuss and challenge the current narratives and the still-high levels
of apathy towards the concerning implications of the gender/sexuality order. This implies a
different organisation of our academic institutions, a shift towards an open science agenda,

16
 Sketching out the field

possible outlets where the outputs of this work can be published, and senior leadership on
the subject of the impact on assessment and career development for academics. As senior
scholars we have an opportunity and responsibility to lead change, pushing harder within
our own institutions. Recruitment and promotion criteria, the lack of diversity in leadership
roles, and the pay gap are a few areas that have been on our radar, but change is slow and
uneven.
Finally, this is an opportunity to revisit the deeply unhelpful divide between education
and research. We have tried, with this volume, to provide all students, from first-year under-
graduates to PhDs, with the tools to carry out their own research and make connections
between theory and real-world problems. Despite the hundreds of LGS courses at UG/PG/
PGT levels, many of our students do not see themselves as pro-feminists, question the deep-
seatedness of the gender/sexuality order, and remain under the impression that this is a ‘free-
for-all’ society, open to self-identification.
This, evidently, does not mean blindness or no societal change. Those currently in their
twenties are more aware and talk more openly of the gender/sexuality order while affor-
dances of technology provide platforms for, some, global activism. The power of social
media enables users to amplify their voices and reach large audiences easily and directly.
The microblogging service Twitter and its technological affordances, in particular, are
playing a pivotal role in grassroots movements. It has become a point of reference and
the hashtag (#) is now widely recognisable within and outside the community of its users.
In this context, landmark movements such as #MeToo3 have enabled sexual harassment
survivors and activists to transcend regional, national, and language borders and to bring
to the fore violence and abuse that is hidden in professional settings and concealed under
loose policies.
Overall, exposing violation of women’s and LGBTQ+ rights is more prominent, while
taken-for-granted inequalities are more frequently challenged in professional and everyday
activities. There is, therefore, some space for optimism. To close on this note, the current
deep disruption we are experiencing will require the remodelling of ways the field engages
with the rupture between second- and first-order concerns in the ‘post-pandemic’ society.
Close to our everyday praxis, the changes in the academy and the blended delivery of cur-
ricula provide the opportunity to enable our students to bring together multiple experiences
of language and gender and sexuality, transcend their local learning environment, and work
with others in cognate programmes and with non-academic stakeholders. Global learning
communities would undoubtedly enhance our field’s reach and feed directly and indirectly
into social change. Opening up our curricula and working across institutions and with dif-
ferent stakeholders is not straightforward; but it is within our power to achieve. I certainly
hope this volume contributes to the collective effort.

Notes
1 Conceptualising and contrasting first- and second-order understandings of linguistic enactments of
complex phenomena attracted traction in linguistic literature in relation to politeness (Locher and
Watts 2005).
2 The Linguistics Society of America has a regular section in this area spotlighting linguists who take
their work to the public – ‘Linguistics in the news’.
3 Tarana Burke founded the ‘Me Too’ movement in 2006 to support women and girls, survivors of sex-
ual violence. It grew to a viral awareness campaign in 2017 after several high-profile female actors
brought to light multiple cases of sexual harassment in the film industry. This triggered an avalanche
of cases across industries, exposing the magnitude of the issue.

17
Jo Angouri

References
Acker, J. (2006) ‘Inequality regimes: gender, class, and race in organizations’, Gender & Society,
20(4), pp. 441–464.
Amin, A. and Roberts, J. (2008) ‘Knowing in action: beyond communities of practice’, Research
Policy, 37(2), pp. 353–369.
Angermuller, J., Maingueneau, D., and Wodak, R. (eds) (2014) The discourse studies reader. Main
currents in theory and analysis. Amsterdam: John Benjamins.
Angouri, J. (2015) ‘Studying identity’, in Hua, Z. (ed) Research methods in intercultural
communication. Oxford: Blackwell, pp. 37–52.
Angouri, J. (2018a) Culture, discourse, and the workplace. London: Routledge.
Angouri, J. (2018b) ‘Quantitative, qualitative, mixed or holistic research? Combining methods in
linguistic research’, in Litosseliti, L. (ed) Research methods in linguistics, 2nd edn. London:
Bloomsbury Academic, pp. 35–56.
Angouri, J. and Piekkari, R. (2018) ‘Organising multilingually: setting an agenda for studying
language at work’, European Journal of International Management, 12(1/2), pp. 8–27.
Atanga, L., Ellece, S. E., Litosseliti, L., and Sunderland, J. (2012) ‘Gender and language in sub-
Saharan African contexts: issues and challenges’, Gender and Language, 6(1), pp. 1–20.
Austin, J. L. (1962) How to do things with words. Oxford: Clarendon Press.
Barthes, R. (1957) Mythologies. Paris: Editions du Seuil.
Bassiouney, R. (2015) Language and identity in modern Egypt. Edinburgh: Edinburgh University
Press.
Baxter, J. (2003) Positioning gender in discourse: a feminist methodology. Basingstoke: Palgrave
Macmillan.
Baxter, J. (2010) The language of female leadership. Basingstoke: Palgrave Macmillan.
Baxter, J. (2014) Double-voicing: power, gender and linguistic expertise. Basingstoke: Palgrave
Macmillan.
Baxter, J. (2018) Women leaders and gender stereotyping in the UK press. Cham: Palgrave Macmillan.
BBC Worklife (2020) How COVID-19 is changing women’s lives. Available at: https://www.bbc.com/
worklife/article/20200630-how-covid-19-is-changing-womens-lives (Accessed: 4th July 2020).
Bourdieu, P. (2001 [1990]) Masculine domination. Cambridge: Polity Press.
Brah, A. and Phoenix, A. (2004) ‘Ain’t I a woman? Revisiting intersectionality’, Journal of
International Women’s Studies, 5(3), pp. 75–86.
Bucholtz, M. and Hall, K. (2004) ‘Theorizing identity in language and sexuality research’, Language
in Society, 33(4), pp. 469–515.
Bucholtz, M. and Hall, K. (2016) ‘Embodied sociolinguistics’, in Coupland, N. (ed) Sociolinguistics:
theoretical debates. Cambridge: Cambridge University Press, pp. 173–197.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Caldas-Coulthard, C. (ed) (2020) Innovations and challenges: women, language and sexism. London:
Routledge.
Cameron, D. (1992) Feminism and linguistic theory. London: Palgrave Macmillan.
Cameron, D. (2005) ‘Language, gender and sexuality: current issues and new directions’, Applied
Linguistics, 26(4), pp. 482–502.
Cameron, D. (2008) The myth of Mars and Venus: do men and women really speak different languages?
Oxford: Oxford University Press.
Cameron, D. (2009) ‘Theoretical issues for the study of gender and spoken interaction’, in Pichler, P.
and Eppler, E. (eds) Gender and spoken interaction. Basingstoke: Palgrave, pp. 1–17.
Cameron, D. and Kulick, D. (2003) ‘What has gender got to do with sex? Language, heterosexuality
and heteronormativity’, in Cameron, D. and Kulick, D. (eds) Language and sexuality. Cambridge:
Cambridge University Press, pp. 44–73.
Canagarajah, S. (2018) ‘Translingual practice as spatial repertoires: expanding the paradigm beyond
structuralist orientations’, Applied Linguistics, 39(1), pp. 31–54.

18
 Sketching out the field

Collins, C., Landivar, L. C., Ruppanner, L., and Scarborough, W. J. (2020) ‘COVID‐19 and the gender
gap in work hours’, Gend Work Organ, 2020, pp. 1–12.
Connell, R. W. (1987) Gender and power: society, the person, and sexual politics. Sydney: Allen and
Unwin.
Davies, B. (1990) ‘The problem of desire’, Social Problems, 37(4), pp. 501–516.
Day, D. and Wagner, J. (2019) Objects, bodies and work practice. Bristol: Multilingual Matters.
Dubois, B. L. and Crouch, I. (1975) ‘The question of tag question in women’s speech: they don’t really
use more of them, do they?’, Language in Society, 4, pp. 289–294.
Duchêne, A., Moyer, M., and Roberts, C. (2013) Language, migration and social inequalities: a
critical sociolinguistic perspective on institutions and work. Bristol: Multilingual Matters.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Think practically and look locally: language and gender as
community-based practice’, Annual Review of Anthropology, 21(1), pp. 461–488.
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) (2014) The handbook of language, gender, and
sexuality, 2nd edn. Malden, MA: Blackwell.
Elias, J. (2008) ‘Hegemonic masculinities, the multinational corporation, and the developmental state:
constructing gender in “progressive” firms’, Men and Masculinities, 10(4), pp. 405–421.
El Refaie, E. (2019) Visual metaphor and embodiment in graphic illness narratives. New York:
Oxford University Press.
Fausto-Sterling, A. (2019) ‘Gender/sex, sexual orientation, and identity are in the body: how did they
get there?’, Journal of Sex Research, 56(4–5), pp. 529–555.
Franziskus, A. (2017) ‘“Have you still not learnt Luxembourgish?”: negotiating language boundaries
in a distribution company in Luxembourg’, in Angouri, J., Marra, M., and Holmes, J. (eds)
Negotiating boundaries at work: talking and transitions. Edinburgh University Press, pp. 178–196.
Freed, A. (2003) ‘Epilogue: reflections on language and gender research’, in Holmes, J. and Meyerhoff,
M. (eds) The handbook of language and gender. Oxford: Blackwell, pp. 699–721.
Frodeman, R., Thompson Klein, J., and Dos Santos Pacheco, R. C. (eds) (2017) The Oxford handbook
of interdisciplinarity, 2nd edn. Oxford: Oxford University Press.
Glynos, J., Howarth, D., Norval, A., and Speed, E. (2009) ‘Discourse analysis: varieties and methods’,
ESRC National Centre for Research Methods, NCRM/014.
Goffman, E. (1959) The presentation of self in everyday life. New York: Doubleday Anchor.
Goodwin, C. (1981) Conversational organization: interaction between speakers and hearers. New
York: Academic Press.
Goodwin, C. (1986) ‘Gestures as a resource for the organization of mutual orientation’, Semiotica,
62(1–2), pp. 29–49.
Goodwin, C. and Goodwin, M. H. (2004) ‘Participation’, in Duranti, A. (ed) A companion to linguistic
anthropology. Malden, MA: Blackwell Publishing, pp. 222–244.
Halliday, M. A. K. and Hasan, R. (1985) Language, context and text aspects of language in a social-
semiotic perspective. Geelong: Deakin University Press.
Harcourt, W., Icaza, R., and Vargas, V. (2016) ‘Exploring embodiment and intersectionality in
transnational feminist activist research’, in Biekart, K., Harcourt, W., and Knorringa, P. (eds)
Exploring civic innovation for social and economic transformation. London: Routledge, pp.
148–167.
Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds) (2008) Gender and language:
research methodologies. Basingstoke: Palgrave.
Heath, C. (1986) Body movement and speech in medical interaction. Cambridge: Cambridge
University Press.
Holmes, J. (1995) Women, men and politeness. London: Longman.
Holmes, J. (2006) Gendered talk at work: constructing gender identity through workplace discourse.
Malden, MA: Blackwell.
Holmes, J. (2020) ‘“Until I got a man in, he wouldn’t listen”; evidence for the gender order in New
Zealand workplaces’, in Caldas-Coulthard, C. R. (ed) Innovations and challenges: women,
language and sexism. London: Routledge, pp. 95–112.

19
Jo Angouri

Holmes, J. and Marra, M. (2004) ‘Relational practice in the workplace: women’s talk or gendered
discourse?’, Language in Society, 33(3), pp. 377–398.
Holmes, J. and Meyerhoff, M. (2003) ‘Different voices, different views: an introduction to current
research in language and gender’, in Holmes, J. and Meyerhoff, M. (eds) The handbook of language
and gender. Oxford: Blackwell, pp. 1–17.
Holmes, J. and Stubbe, M. (2003) Power and politeness in the workplace. London: Pearson Education.
Hua, Z. (2014) ‘Piecing together the “workplace multilingualism” jigsaw puzzle’, Multilingua, 33(1–
2), pp. 233–242.
Hymes, D. (1964) ‘Introduction: towards ethnographies of communication’, American Anthropologist,
66(6), pp. 1–34.
Kulick, D. (2000) ‘Gay and lesbian language’, Annual Review of Anthropology, 29, pp. 243–285.
Lakoff, R. (1975) Language and woman’s place. New York: Harper & Row.
Lakoff, R. (2004) Language and woman ’s place: text and commentaries, edited by Bucholtz, M.
Oxford: Oxford University Press.
Lave, J. and Wenger, E. (1991) Situated learning: legitimate peripheral participation. Cambridge:
Cambridge University Press.
Lazar, M. M. (ed) (2005) Feminist critical discourse analysis: gender, power and ideology in
discourse. London: Palgrave Macmillan.
Lazar, M. M. (2009) ‘Entitled to consume: postfeminist femininity and a culture of post-critique’,
Discourse and Communication, 3, pp. 371–400.
Litosseliti, L., Gill, R., and Favaro, L. (2019) ‘Postfeminism as a critical tool for gender and language
(study)’, Gender and Language, 13(1), pp. 1–22.
Livia, A. and Hall, K. (eds) (1997) Queerly phrased: language, gender, and sexuality. Oxford: Oxford
University Press.
Locher, M. A. and Watts, R. J. (2005) ‘Politeness theory and relational work’, Journal of Politeness
Research, 1, pp. 9–33.
Mason, K. (2012) ‘The unequal weight of discrimination: gender, body size, and income inequality’,
Social Problems, 59(3), pp. 411–435.
McCall, L. (2005) ‘The complexity of intersectionality’, Signs: Journal of Women in Culture and
Society, 30, pp. 1771–1800.
McElhinny, V. (2007) A look into the new bank of the South, due to launch in December 2007.
Washington, DC: Bank of the South.
Metcalfe, B. D. and Woodhams, C. (2012) ‘Introduction: new directions in gender, diversity and
organization theorizing – re-imagining feminist post-colonialism, transnationalism and geographies
of power’, International Journal of Management Reviews, 14, pp. 123–140.
Mondada, L. (2008) ‘Using video for a sequential and multimodal analysis of social interaction:
videotaping institutional telephone calls’, Forum: Qualitative Social Research, 9(3), art. 39.
Mondada, L. (2019) ‘Contemporary issues in conversation analysis: embodiment and materiality,
multimodality and multisensoriality in social interaction’, Journal of Pragmatics, 145, pp. 47–62.
O’Halloran, K. L. (2011) ‘Multimodal discourse analysis’, in Hyland, K. and Paltridge, B. (eds)
Companion to discourse analysis. London: Continuum, pp. 120–137.
Pennycook, A. (2018) Posthumanist applied linguistics. Oxford: Routledge.
Piller, I. (2016) Linguistic diversity and social justice: an introduction to applied sociolinguistics.
Oxford: Oxford University Press.
Piller, I. and Lising, L. (2014) ‘Language, employment and settlement: temporary meat workers in
Australia’, Multilingua, 33(1/2), pp. 35–59.
Piller, I. and Takahashi, K. (2010) ‘Language, migration and human rights’, in Wodak, R., Johnston,
B., and Kerswill, P. (eds) Sage handbook of sociolinguistics. London: SAGE, pp. 573–587.
Plester, B. (2015) ‘“Take It Like a Man!”’: performing hegemonic masculinity through organizational
humour’, Ephemera: Theory and Politics in Organization, 15(3), pp. 537–559.
Power, K. (2020) ‘The COVID-19 pandemic has increased the care burden of women and families’,
Sustainability: Science, Practice, and Policy, 16(1), pp. 67–73.

20
 Sketching out the field

Roberts, C. and Campbell, S. (2006) Talk on trial. Job interviews, language and ethnicity. DWP
Report 344.
Rusk, F., Pörn, M., Sahlström, F., and Slotte-Lüttge, A. (2015) ‘Perspectives on using video recordings
in conversation analytical studies on learning in interaction’, International Journal of Research &
Method in Education, 38(1), pp. 39–55.
Sanders, C. (ed) (2004) The Cambridge companion to Saussure. Cambridge: Cambridge University
Press.
de Saussure, F. (1974 [1916]). Course in general linguistics, translated by Wade Baskin. London:
Fontana/Collins.
Searle, J. R. (1969) Speech acts: an essay in the philosophy of language. Cambridge: Cambridge
University Press.
Schulz, K. (ed) (2017) The women’s liberation movement: impacts and outcomes. New York: Berghahn
Books.
Schutz, A. (1953) ‘Common-sense and scientific interpretation in human action’, Philosophy and
Phenomenological Research, 14, pp. 1–38.
Stone, A. (2004) ‘Essentialism and anti-essentialism in feminist philosophy’, Journal of Moral
Philosophy, 1(2), pp. 135–153.
Streeck, J. and Hartge, U. (1992) ‘Gestures at the transition place’, in Auer, P. and di Luzio, A. (eds)
The contextualisation of language. Amsterdam: John Benjamins, pp. 135–157.
Tannen, D. (1990) You just don’t understand: women and men in conversation. New York:
HarperCollins Publishers.
Tannen, D. (1994) Talking from 9 to 5: how women’s and men’s conversational styles affect who gets
heard, who gets credit, and what gets done at work. New York: William Morrow.
Tanu, P. and Pannicot, D. (2018) ‘Production of gender: a study on performativity in female-to-male
transsexuals’, Masculinities & Social Change, 7(3), pp. 255–278.
The Economist (2020) Downturns tend to reduce gender inequality. Not under COVID-19. Available
at: https://www.economist.com/finance-and-economics/2020/06/04/downturns-tend-to-reduce-g
ender-inequality-not-under-covid-19 (Accessed: 4th July 2020).
The Guardian (2018) The university of the future will be interdisciplinary. Available at: https://www
.theguardian.com/higher-education-network/2018/jan/24/the-university-of-the-future-will-be-int
erdisciplinary (Accessed: 4th July 2020).
UN Policy Brief (2020) The impact of COVID-19 on women. United Nations.
Vuori, J (2009) ‘Guiding migrants to the realm of gender equality’, in Keskinen, S., Tuori, S., and Irni,
S. (eds) Complying with colonialism: race and ethnicity in the nordic region. Surrey: Ashgate, pp.
207–223.
Wetherell, M. and Edley, N. (1999) ‘Negotiating hegemonic masculinity: imaginary positions and
psycho-discursive practices’, Feminism & Psychology, 9(3), pp. 335–356.
Wodak, R. (1997) Gender and discourse. London: SAGE.
Wodak, R. (2005) ‘Gender mainstreaming and the European Union: interdisciplinarity, gender studies
and CDA’, in Lazar, M. M. (ed) Feminist critical discourse analysis. London: Palgrave Macmillan.
Wodak, R. (2013) ‘Dis-citizenship and migration: a critical discourse-analytical perspective’, Journal
of Language, Identity, and Education, 12(3), pp. 173–178.
Wodak, R. (2015) The politics of fear: what right-wing populist discourses mean. London: SAGE.
Wodak, R. and Meyer, M. (eds) (2009) Methods for critical discourse analysis, 2nd edn. London:
SAGE.
Zimman, L. (2018) ‘Transgender voices: insights on identity, embodiment, and the gender of the
voice’, Lang Linguist Compass, 12, p. e12284.
Zimman, L. and Hall, K. (2010) ‘Language, embodiment, and the “third sex”’, in Llamas, C. and
Watt, D. (eds) Language and identities. Edinburgh: Edinburgh University Press, pp. 166–178.

21
 Part I
Variationist approaches
 2
Non-binary approaches to gender
and sexuality
Penelope Eckert and Robert J. Podesva (Part I leads)

Introduction
The study of variation is founded on replicable correlations with macrosocial categories,
particularly with class, gender, age, and ethnicity. These correlations provide the larger gen-
eral landscape of social variability – the distribution of variables across society, the social
areas and kinds of distinctions that are at work in variation, and the nature of the networks
that guide the spread of change. At the same time, we need to treat these correlations with
caution, because categories are structural objects that abstract away from, and erase, the
practice that produces them. They provide a global pattern, but no explanation. A purely
structural perspective can all too easily lead to deterministic analyses – women talk the way
they talk because they are women, working class people because they’re working class, gay
people because they are gay. This approach guides the focus to the categories themselves
as objects of indexicality, or to some stereotype associated with membership in particular
categories. It also leads us to think purely in terms of the analytic structure of the categories
– as continua (class, age, attention paid to speech) or binary oppositions (gender, sexual
orientation, race, and even ‘above’ or ‘below’ the level of consciousness).
This is illustrated in the path that gender and sexuality studies in variation has taken.
Labov’s (1966) analysis of variation and socioeconomic class was a radical move in the
1960s, opening the door to serious quantitative work in linguistics and introducing a solid
basis for the social analysis of variability. But the robust finding that linguistic form corre-
lates with class stratification established a class basis for subsequent interpretation of vari-
ation. The view of class as a continuum of global prestige was established as the source
of meaning in variation, with variants viewed as carrying prestige or stigma, and patterns
of style shifting as reflecting orientation to the prestige and stigma of class position. With
class-based prestige and stigma as the fundamental terms of indexicality, correlations
with gender were then taken to reflect binary orientations to prestige and to class position.
Women’s greater use than men of stable variables was attributed to status consciousness
(e.g. Trudgill 1972) – an attribution that has not held up well in the face of the regular
finding that women commonly (but not always1) lead in sound change. At the same time,
the assumption that women are more status conscious than men has led class to bleed into

25
Penelope Eckert & Robert J. Podesva

qualities, with femininity associated with refinement and masculinity with toughness. And
as attention has turned to sexuality, it has been all too easy to treat sexual orientation as a
fractal (Irvine and Gal 2000) within the male/female binary, with the speech of gay women
and men treated as masculine and feminine respectively. Just about any linguistic variable is
likely to show a correlation with class and with binary gender because both are fundamental
to the social order – in fact, they are so fundamental that correlations with them are often in
themselves meaningless. People pursue their lives at the intersections of social categories,
each intersection bringing forth its own conditions and situations, its own constraints and
possibilities. It is in day-to-day practice at this juncture that variation becomes meaningful,
and that correlations are produced.

Variation, style, and persona

We take as illustration the use of negative concord (e.g. ‘I didn’t do nothing’) in the speech
of students at Belten High School in the Detroit suburbs (Eckert 2000). Figure 2.1 shows a
stark version of the typical gender pattern in the use of negative concord at Belten High – the
boys use significantly more negative concord than the girls. But the picture shifts when we
look beyond this binary, as participation in the class-based social categories that dominate
the school social order yields yet another binary that interacts with gender. The ‘Burnouts’
and the ‘Jocks’ (Eckert 1989, 2000) are communities of practice (Eckert and McConnell-
Ginet 1992; Lave and Wenger 1991) constituting working-class and middle-class cultures
respectively. Figure 2.2 shows that while the overall gender difference persists in each com-
munity of practice, it is far greater among the Jocks than among the Burnouts. It is important
to note that the strong correlation is with the speaker’s participation in their peer-based com-
munities of practice, and not with their parents’ socioeconomic class. This indicates that the
use of negative concord is not passively acquired at home and in the neighbourhood, but is
part of the stylistic construction of an adolescent self.
We note that the gender difference is far greater among the Jocks than the Burnouts,
an interaction that clearly indicates that these categories are not simply additive, but

50
45
40
% Negative Concord

35
30
25
20
15
10
5
0
Boys Girls

Figure 2.1 Percent negative concord by binary gender. (x2 = 3124, p = .000).

26
 Non-binary approaches

50

45

40

35
% Negative Concord

30

25

20

15

10

0
Burnout Girls Jock Girls Burnout Boys Jock Boys

Figure 2.2 Percent negative concord by binary gender and social category.

intersectional. Most obviously, while Jock girls are considerably more constrained than
boys to be conservative in their behaviour, Burnout girls are not (see Eckert 1989, 2000).
There is considerable variety of behaviours among Burnout girls, ranging from fairly com-
pliant with adult norms to quite rebellious. This range shows up in a friendship cluster
of urban-oriented and rebellious Burnout girls who pride themselves on being the ‘big-
gest Burnouts’, commonly referred to by their classmates as the ‘Burned-out Burnouts’.
These girls are a quite distinct friendship group from the bulk of burnout girls who embrace
more working-class norms and who are less alienated from school, and less rebellious –
the ‘Regular Burnouts’. Figure 2.3 compares several clusters of Burnouts and Jocks with
the larger student body represented by not only Jocks and Burnouts but students who do
not identify with either. As this figure shows, the Burned-out Burnout girls lead the entire
student body in the use of negative concord. Meanwhile, those who use the least negative
concord are the Jock boys who are engaged in student government (as opposed to those who
are athletes only). In other words, we see that gender and class break down into the kinds
of indexicalities that Trudgill raised, but not in the gender groups for which he raised them.
Rather, the Burned-out Burnout girls are using negative concord to index an autonomous
and anti-establishment stance, while the male student government Jocks are using standard
variants in maintaining their corporate personae. When talking at the macrosocial level, it
is reasonable to say that gender norms constrain women and men to engage in the world
in such a way that it benefits them to use standard and nonstandard forms respectively. But
these norms work not at the binary gender level but at the persona level as constrained, but
not determined, by the gender binary. If negative concord indexes gender, it’s only indirectly
through the relation between what it indexes directly (anti-establishment stance) and what
it ‘means’ to be male or female.

27
 Penelope Eckert & Robert J. Podesva

1.4

1.2

1
STANDARD DEVIATION FROM CLASS MEAN

0.8

0.6

0.4

0.2

-0.2

-0.4

-0.6

-0.8
Burned-out Burnout Regular Athlete Jock Jock girls Non-Athlete
Burnout girls boys Burnout girls boys Jock boys

Figure 2.3 Negative concord by subcategory. Shown as standard deviation from class mean,
including Jocks, Burnouts, and In-Betweens.

In other words, as Levon (this volume, Chapter 3) says, we need to always ask the ‘other
question’. In this case the question is: ‘When I see gender difference, where are the class
interests and when I see class difference, where are the gendered interests?’. When we do
this, we move on to consider the dynamics that unfold at the intersection of class and gender.
And we experience gender – and class – not in isolation, but in the practices and personae
that emerge at their intersection.
The Burned-out Burnouts and the Corporate Jocks are not passively responding to their
place in the social order – in fact, nobody is. Social practices emerge as people in similar
structural locations mutually engage in similar responses – responses that commonly repro-
duce the structure that gave rise to them, but that also carry the potential for change. Butler’s
theory of performativity emphasises the dialectic between structure and agency in the pro-
duction of the gendered individual. Category assignment places the individual in situations
where they learn to engage in ‘a stylised repetition of acts through time’ (Butler 1988: 520),
stylised acts that constitute gender as a central aspect of emergent identities. All aspects of
identity are achieved performatively, as we engage with the affordances of our social world
to construct a self with which we can take a place in that world. Central to Butler’s account
of performativity is that in these repeated acts lies the potential for change:

If the ground of gender identity is the stylized repetition of acts through time, and not
a seemingly seamless identity, then the possibilities of gender transformation are to be
found in the arbitrary relation between such acts, in the possibility of a different sort of
repeating, in the breaking or subversive repetition of that style.
(Butler 1988: 520)

28
 Non-binary approaches

Style is the key to performativity. It is the means by which we mark both pre-existing and
emergent distinctions (Irvine 2001). Emergent distinctions might be particularly tentative,
and a small stylistic move is sufficiently inexplicit to provide the possibility of denial, put-
ting a verbal toe in the water. A young man who has been in the closet might want to make
a tentative move towards sounding gay. Producing a slightly more fronted /s/ than usual
would be both slightly perceptible and quite deniable, and depending on the reactions of
others and his own embodied sensation in making that move, he might decide to move for-
ward in or beyond that situation, or not. When we think of style we focus on the projection
of persona to others, but style is also a bodily act and has a ‘feel’ that connects the speaker
to the outward presentation. A stylistic move, in other words, is not just an external act, but
a small – perhaps temporary – bit of personal transformation. And stylistic practice, the
production of persona, is not a momentary event, but a continuous process that carries us
through life. While there may be moments in which persona is foregrounded, and in which
we make quite intentional moves, most of the time stylistic practice takes place quite unin-
tentionally, even unconsciously. And it takes place not at the level of ‘gender’ or ‘sexuality’,
but at the level of qualities, stances, momentary activities that are about the kind of person
we want to be in the moment. While some of these qualities, stances, and activities may be
overtly associated with gender or sexuality, most of them are quite indirectly (Ochs 1992)
indexical of categories. And certainly, no single stylistic feature – no single variable – can
work alone. Stylistic practice is a process of bricolage (Hebdige 1984; Levi Strauss 1962)
in which meaningful elements are combined to constitute a higher meaning that is more than
the sum of its parts.
Distinctions of voice quality, segmental phonetics, prosody all have meaning potential,
but potential that is underspecified, and unrealised in isolation. /t/ release in flap position
has been found in the speech of Orthodox Jews (Benor 2001), geek girls (Bucholtz 1996),
and gay men (Podesva et al. 2002). And while this single feature can be said to index each
category, it does so only in combination with the other elements that constitute the larger
style in which it is embedded. Podesva (2004, 2007) documented the stylistic changes of a
young gay doctor, Heath, as he moved from the clinic to a barbeque with his friends. In the
clinic, his ‘caring doctor’ persona featured a good deal of /t/ release, indexing a formal and
articulate style. In his ‘gay diva’ persona at a barbeque with his friends, he used fewer stop
releases, but they had significantly longer and more intense bursts, giving them a parodic
quality. And while he used falsetto in the clinic, but almost exclusively on discourse markers
(‘okay’, ‘alright’), his gay diva persona made copious use of falsetto with greater duration,
greater f0 range and maximum. The /t/ release alone would not have contributed much to
his gay diva persona – in fact, without other elements such as falsetto, the /t/ release might
have just sounded weird. It is in this stylistic practice – the combination and recombination
of resources – that variation takes on meaning. And that meaning is not fixed, but changes
as our needs to make meaning change.
Indeed, a fundamental problem with the purely structural view is that it implies a kind
of permanence. Class, gender, and sexuality are certainly here to stay as categorisation
schemes, but the structure of the schemes themselves changes with time. And this change
does not happen in the abstract, but is inseparable from each individual’s trajectory through
life, as each generation hopes to move beyond the previous one – to be more ‘modern’, more
‘advanced’, to move on from the past. New ideas emerge, new issues, and with them new
ways of being in the world, and bricolage makes it possible to use old material to bring these
new ways into the social landscape.

29
Penelope Eckert & Robert J. Podesva

Meaning and use: creaky voice

In the remainder of this chapter, we illustrate the ideas above in an extended discussion of
creaky voice, which has undergone massive indexical change. This single linguistic feature
also illustrates some of the pitfalls of the binary thinking that underlies variationist socio-
linguistic research. We will show that creaky voice is both ideologically associated with
femininity, and more prevalent in the speech of speakers who identify as women. In our
view, these facts are trivial, as they tell us nothing about why women exhibit a preference
for creak, nor do they tell us anything about gender. We advocate for an alternative approach
that is concerned less with the question of who creaks the most and more with the question
of what creak means. An approach centred around social meaning, we argue, can better
explain gendered distributions of sociolinguistic variables while also providing insight into
the social construction of gender.
The other chapters in this section similarly illustrate the importance of taking a meaning-
centric approach to the study of gender and sexuality. Zimman, for example, highlights the
limits of conceptualising gender as having an atomic meaning. His study shows that gender
identity generally predicted the realisation of /s/, but gender role (‘socially and/or institution-
ally recognised gender categories within a given culture’) – a dimension of gender distinct
from gender identity – was necessary to account for outliers. Levon emphasises that sexual-
ity is constructed by means of linguistic features whose meanings are variable depending
on speakers’ gender presentation. Generally, as the degree to which female speakers iden-
tify as ‘lipstick’ increases, so too do their mean pitch levels. But different speakers exploit
higher pitch in distinct ways, with ‘lipstick’ speakers using higher pitch when discussing
gay topics and ‘butch’ speakers using higher pitch for non-gay topics. Campbell-Kibler and
miles-hercules also stress that meanings relating to gender and sexuality are not inherent to
linguistic features, but rather vary as a function of contextual factors, such as who the lis-
tener is. Using perceptual methods, they show that in spite of a dominant ideology whereby
gayness and masculinity are negatively correlated among men, this association does not
hold across the listener sample. Together, these studies indicate that ideologies associating
linguistic forms with gender and sexuality are at once too general (linguistic practice cannot
be explained by unanalysed, monolithic notions of gender) and too specific (gender is sel-
dom articulated independently of other dimensions of identity), and fail to capture the fact
that gender and sexuality are emergent.
That creak has been ideologised as a (young) feminine speech feature is apparent in
media discourse surrounding the feature, also referred to as vocal fry. Creaky voice has been
described as ‘the new way young women talk’ (Weiss 2013) and a ‘female fad’ promulgated
by celebrities like Britney Spears, Kim Kardashian, and Zooey Deschanel (Steinmetz 2011).
In an opinion piece for The Guardian, author Naomi Wolf (2015) goes so far as to issue the
plea: ‘Young women, give up the vocal fry and reclaim your strong female voice’.
There is reason to think that the ideology of the young female creaker is based – at least
in part – on patterns of use. A number of studies have documented the prevalence of creak
in the speech of young women (e.g. Wolk, Abdelli-Beruh, and Slavin 2012) and shown that
women produce more creaky voice than men (Podesva 2013; Yuasa 2010). These previ-
ous studies are limited in that they have employed auditory methods, considered limited
samples of speakers, and, in some cases, been based on analyses of read speech, which
may not satisfy the social conditions required for the extensive production of creak. To
address this issue, Callier and Podesva (2015) carried out a larger-scale acoustic study of
creaky voice among 93 white, cisgender speakers from inland California. About half of the

30
 Non-binary approaches

speakers self-identified as women, and both women and men spanned the entire adult life
course, from 18 to 93 years old. Data were approximately hour-long sociolinguistic inter-
views. Every vowel was labelled as either creaked or not (using Kane, Drugman, and Gobl’s
2013 model, which classifies speech on the basis of acoustic properties) and measured for a
variety of acoustic parameters that capture different dimensions of voice quality variation,
including spectral tilt (e.g. H1*–H2*), or the relative power of higher frequencies compared
to lower frequencies in speech, and periodicity (e.g. CPPS), or how regularly the vocal folds
vibrate. The results are in many ways consistent with previous studies, as women produced
significantly more creaky voice than men.
While we wouldn’t question the robustness of this overall pattern, it is important to
look beyond binary gender difference to identify two other robust patterns. Figure 2.4 plots
H1*–H2* (an acoustic measure of spectral tilt that captures the degree of glottal constric-
tion during phonation) as a function of age for both female and male speakers. Lower val-
ues of H1*–H2* indicate a great degree of glottal constriction, thus creakier phonation.
The first pattern to note is that women exhibit a curvilinear pattern across the age span,
that is, younger women creak the strongest, middle-aged women creak the weakest, and
older women produce stronger creak than middle-aged women. This is a robust quantita-
tive pattern, as a quadratic term for age emerged as significant for female speakers in a
mixed-effects linear model, which tests for the effects of social factors while controlling
for the influence of other social and linguistic factors, as well as random variation from one
speaker to the next. So statements like ‘women creak the most’ erase intragender differ-
ences and falsely characterise women as exhibiting uniform patterns. To return to our earlier

2.0

1.5

1.0

0.5

0.0
M
–0.5
H1*-H2* (dB)

–1.0

–1.5

–2.0

–2.5

–3.0
F
–3.5

–4.0

–4.5

–5.0

–5.5
15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Age (years)

Figure 2.4 H1*–H2* (dB) as a function of speaker age for women (F) and men (M). Lower values
indicate a greater degree of glottal constriction (creakier phonation).

31
Penelope Eckert & Robert J. Podesva

discussion, we need to ask the other question, ‘When I see gender difference, where is age?’
We also note that even though women produce creakier phonation than men overall, young
men in particular produce relatively strong creaky voice. Statements like ‘women creak
the most’ incorrectly erase young men from the narrative of who creaks more and why.
Callier and Podesva (2015) suggest that the noteworthy point about creak isn’t so much that
women are doing it more, but that young people – women and men alike – are doing it more,
because the domain of creak has expanded from its canonical phrase-final position to earlier
in the phrase. A focus on gender alone would have obscured this pattern, yielding mislead-
ing insights about gender and failing to accurately characterise its linguistic patterning.
At this point, we can say on the basis of apparent time data that creaky voice is on the rise,
and that young women (and young men) are leading the change. The question here is why.
To arrive at an answer, we argue, we need to consider its social meaning. Though creaky
voice was once more prevalent in the speech of men (Esling 1978; Henton and Bladon
1988), it is unlikely that women are using it to borrow on men’s authority. Apart from being
easy and far-fetched binary thinking, it fails to consider that it has had a pragmatic use
independent of gender for some time. Although media treatments of creak characterise it as
functionally useless at best and physiologically harmful at worst (Garfield 2013; Steinmetz
2011), social evaluation studies have shown that creak is not always evaluated negatively.
Eckert (2019), for example, has shown that younger listeners are more forgiving than older
speakers in their assessments of very creaky speech samples from National Public Radio
news personalities. Yuasa (2010) further reports that listeners judge speakers who produce
relatively creaky speech as sounding professional. These studies hint at the possibility that
creaky voice could serve a useful interactional or stylistic function.
Work in discourse analysis suggests that speakers can strategically use creaky voice to
show disengagement or express negative affect. Lee (2015) shows that speakers commonly
slip into creaky voice when they go off topic for a bit, suggesting that creak distances paren-
thetical speech from the main thread of a conversation. She goes on to argue that speakers
can draw on this conventional function of creaky voice to distance themselves from the
issue under discussion, even when speech is on topic. Grivičić and Nilep (2004) advance
a similar claim in their analysis of the word ‘yeah’ in telephone conversations. They argue
that creaky ‘yeah’ expresses either a disalignment between interlocutors or a dispreference
to continue on the current topic. Zimman (2017) shows that a transmasculine speaker draws
on creak to index ‘a stance of disaffectation, an aloof persona, or a kind of emotional stoi-
cism’. In all of these studies, creaky voice distances the speaker from either their addressee
or the topic of discussion. The recurrence of this function of creak across studies suggests
that it may carry a more stable, conventionalised meaning, just like pitch and the realisation
of /s/, as discussed in all three of the other chapters in this section.
In contrast to most discourse analytic work, one of the goals of variationist sociolinguis-
tics is to look beyond immediate interactions and draw generalisations across a community.
Yet it is difficult to generalise about the ways that groups employ linguistic features (as
opposed to simply the frequency with which they employ them) without attending to the
interactional concerns at the heart of discourse analytic work, such as stance-taking. To
repeat a phrase from earlier in this chapter, meaning lies in use. Of course, generalisation
requires the analysis of relatively large amounts of data, which preclude line-by-line analy-
ses of turns at talk. Is it possible to operationalise interactional analysis in a quantitative
variationist paradigm?
Podesva (2018) advocates for an approach that quantifies affective stance on the basis of
its expression across modalities. Such an approach makes it possible to test the hypothesis

32
 Non-binary approaches

that creaky voice conveys negative, disengaged affect. Starting with the most direct expres-
sion of affect, we can consider speakers’ own assessments of interactions. We should
observe an inverse correlation between how much speakers enjoy interactions and how
much they creak. Second, we can examine the sentiment conveyed through the lexical or
semantic material of speech. Beginning with the lexicon, if creak conveys negative, disen-
gaged affect, we should observe more creak on words that convey negative valence, low
arousal, and low dominance. Finally, we can consider the least linguistically explicit, though
arguably the most transparently readable, expression of affect – embodied affect. If creak
indexes negative affect, we would expect to see higher rates when people are not smiling.
And if creak conveys disengagement, we would expect to see higher rates when people are
moving their bodies less.
These predictions were tested in a corpus of audio-visual recordings collected in the
Interactional Phonetics Laboratory at Stanford University. The lab’s ‘Living Room’ has
the acoustical specifications of a sound booth (thus enabling the collection of high-quality
audio recordings), but it is staged like a living room (to enable speakers to relax into
the environment) and features inconspicuous video cameras (to enable the analysis of
embodied practices). Approximately 150 speakers – mostly a diverse group of Stanford
undergraduate and graduate students – were recorded in dyadic unscripted conversations.
Of these speakers, 42, all from the Western United States, were chosen for analysis. Data
were acoustically analysed using the methods described above for Callier and Podesva’s
(2015) study. The valence, arousal, and dominance for each word was coded using the
lexicons published by Mohammad (2018). Finally, computer vision methods were used
to identify when speakers were smiling, using the method in Podesva, Callier, Voigt, and
Jurafsky (2015), and how much they were moving, using the method in Voigt, Jurafsky,
and Podesva (2013).
Results largely confirm the hypotheses. Creaky voice is more prevalent in interactions
that speakers rate as less enjoyable. It is also used more commonly on words that con-
vey less dominance. Finally, creaky voice correlates with both forms of embodiment. For
female speakers, creaky voice is more common in phrases where speakers are not smiling.
And regardless of gender, speakers creak more at moments of time when they are moving
less. This recalls a similar finding in Pratt’s (2018) study based on a year-long ethnography
of students at a high school for the performing arts. She found that students in the school
drew a distinction between kids who were considered ‘chill’ and those who were considered
higher-energy ‘louds’. She found that ‘chill’ students not only shifted their sitting positions
less over the course of sociolinguistic interviews with her, but they also used higher rates
of creaky voice.
The data largely support the hypothesis that creaky voice conveys negative, disengaged
affect, which enables us to finally return to the question of why young women (and young
men) do it more often, why the feature is on the rise. In short, creaky voice enables speakers
to disengage or take negatives stances without saying so overtly (cf. Besnier 1990). Eckert
(2016) has argued that a design feature of language is being able to convey social mean-
ing, including affect, without expressing it through lexical content. The social meanings
conveyed by creaky voice, and all sociolinguistic variables for that matter, do not alter an
utterance’s referential meaning. As a result, it would be infelicitous to respond to someone
who is creaking by asking, ‘Why are you disengaging from me?’ The disengagement is in
the room, but it isn’t on the table, so to speak. The plausible deniability of disengagement
is useful, perhaps most useful to women in interactions replete with ‘mansplaining’ and
‘manterruptions’.

33
Penelope Eckert & Robert J. Podesva

Conclusion
Gender is the product more than the explanation for linguistic variation. Viewing gender
through a performative lens encourages variationists to situate gender locally, in interaction,
and in dialogue with other dimensions of identity. In spite of ideologies that cast gender as
a rigid binary structure, gender is constructed in practice in non-binary ways. The remain-
ing chapters in this section offer ways of conceptualising and analysing the non-binarity
of gender and sexuality. Zimman deconstructs gender into four potentially distinct dimen-
sions – gender assigned at birth, gender identification, gender role, and gender presentation
– and argues that doing so is necessary not only for adequately characterising the linguistic
practices of transgender individuals, but also for explaining intragroup diversity among
members of any gender category. Levon surveys a body of work examining the diversity of
linguistic practices among individuals that comprise half of the sexuality binary (‘LGB’ or
‘queer’) and another body of work that considers how sexuality is articulated in conjunction
with other mutually constitutive identity categories. Campbell-Kibler and miles-hercules
quantify how strongly listeners’ masculinity ratings of hypermasculine speech performances
correlate with their gayness ratings. They find not only that listeners vary significantly with
respect to the strength of this correlation, but that the correlation is weaker among African
American listeners. All three chapters emphasise the intersectionality of gender and sexu-
ality. In the present chapter, we have argued for an approach to variation that focuses on
personae, as speakers use linguistic features to perform gendered personae (e.g. Eckert 2006
[1996]; Podesva 2007), and listeners draw on and update representations of these perso-
nae when evaluating and perceiving speech (e.g. Campbell-Kibler 2007; D’Onofrio 2015,
2018). All of this is possible because variation is meaningful. So variationists must continue
to drill down, past gender, to locate the meanings from which gender is constructed, mean-
ings that represent how gender is experienced.

Note
1 We note, also, that it has become commonplace to take women’s lead in the use of a variant as diag-
nostic of sound change – yet another example of reliance on a shaky binary.

References
Benor, S. (2001) ‘Sounding learned: phonological variation in orthodox Jewish English’, in Sanchez,
T., and Johnson, D. E. (eds) Penn working papers in linguistics: selected papers from NWAV 29.
Philadelphia: University of PA, Department of Linguistics, pp. 1–16.
Besnier, N. (1990) ‘Language and affect’, Annual Review of Anthropology, 19, pp. 419–451.
Bucholtz, M. (1996) ‘Geek the girl: language, femininity and female nerds’, in Warner, N., Ahlers,
J., Bilmes, L., Oliver, M., Wertheim, S., and Chen, M. (eds) Gender and belief systems. Berkeley:
Berkeley Women and Language Group, pp. 119–131.
Butler, J. (1988) ‘Performative acts and gender constitution: an essay in phenomenology and feminist
theory’, Theatre Journal, 40, pp. 519–531.
Callier, P. and Podesva, R. J. (2015) Multiple realizations of creaky voice: evidence for phonetic
and sociolinguistic change in phonation. Paper presented at New Ways of Analyzing Variation
(NWAV) 44, Toronto.
Campbell-Kibler, K. (2007) ‘Accent, (ING), and the social logic of listener perceptions’, American
Speech, 82, pp. 32–64.
Campbell-Kibler, K. (2011) ‘Intersecting variables and perceived sexual orientation in men’, American
Speech, 86, pp. 52–68.

34
 Non-binary approaches

D’Onofrio, A. (2015) ‘Persona-based information shapes linguistic perception: valley girls and
California vowels’, Journal of Sociolinguistics, 19, pp. 241–256.
D’Onofrio, A. (2018) ‘Controlled and automatic perceptions of a sociolinguistic marker’, Language
Variation and Change, 30, pp. 261–285.
Eckert, P. (1989) Jocks and burnouts: social categories and identity in the high school. New York:
Teachers College Press.
Eckert, P. (2000) Linguistic variation as social practice. Oxford: Blackwell.
Eckert, P. (2006 [1996]) ‘Vowels and nailpolish: the emergence of linguistic style in the preadolescent
heterosexual marketplace’, in Cameron, D. and Kulick, D. (eds) The language and sexuality
reader. London: Routledge, pp. 189–195.
Eckert, P. (2016) ‘Variation, meaning, and social change’, in Coupland, N. (ed) Sociolinguistics:
theoretical debates. Cambridge: Cambridge University Press, pp. 68–85.
Eckert, P. (2019) ‘Indexical obsolescence’, in Blake, R. and Buchstaller, I. (eds) The Routledge
companion to the work of John Rickford. London: Routledge, pp. 298–307.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Think practically and look locally: language and gender as
community–based practice’, Annual Review of Anthropology, 21, pp. 461–490.
Esling, J. (1978) ‘The identification of features of voice quality in social groups’, Journal of the
International Phonetic Association, 8, pp. 18–23.
Garfield, B. (2013) ‘“Old fart” responds to the great vocal fry outcry of 2013’, Slate. Available at: https
://slate.com/human-interest/2013/01/young-women-and-vocal-fry-slate-podcast-wars-continue.html
Grivičić, T. and Nilep, C. (2004) ‘When phonation matters: the use and function of yeah and creaky
voice’, Colorado Research in Linguistics, 17, pp. 1–10.
Hebdige, D. (1984) Subculture: the meaning of style. New York: Methuen.
Henton, C. and Bladon, A. (1988) ‘Creak as a sociophonetic marker’, in Hyman, L. M. and Lee, C. N.
(eds) Language, speech, and mind: studies in honor of Victoria A. Fromkin. London: Routledge,
pp. 3–29.
Irvine, J. (2001) ‘Style as distinctiveness: the culture and ideology of linguistic differentiation’, in
Eckert, P. and Rickford, J. (eds) Stylistic variation in language. Cambridge: Cambridge University
Press, pp. 21–43.
Irvine, J. and Gal, S. (2000) ‘Language ideology and linguistic differentiation’, in Kroskrity, P. V. (ed)
Regimes of language: ideologies, politics, and identities. Santa Fe, NM: SAR Press, pp.35–83.
Kane, J., Drugman, T., and Gobl, C. (2013) ‘Improved automatic detection of creak’, Computer
Speech and Language, 27, pp. 1028–1047.
Labov, W. (1966) The social stratification of English in New York City. Washington, DC: Center for
Applied Linguistics.
Lave, J. and Wenger, E. (1991) Situated learning: legitimate peripheral participation. Cambridge:
Cambridge University Press.
Lee, S. (2015) ‘Creaky voice as a phonational device marking parenthetical segments in talk’, Journal
of Sociolinguistics, 19, pp. 275–302.
Lévi Strauss, C. (1962) La pensée sauvage. Paris: Plon.
Mohammad, S. (2018) ‘Obtaining reliable human ratings of valence, arousal, and dominance for 20,000
English words’, in Proceedings of the 56th annual meeting of the association for computational
linguistics. Melbourne, Australia.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context.
Cambridge: Cambridge University Press, pp. 335–358.
Podesva, R. J. (2004) On constructing social meaning with stop release bursts. Paper presented at
Sociolinguistics Symposium 15. Newcastle upon Tyne.
Podesva, R. J. (2007) ‘Phonation type as a stylistic variable: the use of falsetto in constructing a
persona’, Journal of Sociolinguistics, 11, pp. 478–504.
Podesva, R. J. (2013) ‘Gender and the social meaning of non-modal phonation types’, Proceedings of
the Berkeley Linguistics Society, 37, pp. 427–448.

35
Penelope Eckert & Robert J. Podesva

Podesva, R. J. (2018) The affective roots of gender patterns in the use of creaky voice. Paper
presented at Experimental and Theoretical Approaches to Prosody (ETAP) 4 at the University of
Massachusetts, Amherst.
Podesva, R. J., Callier, P., Voigt, R., and Jurafsky, D. (2015) ‘The connection between smiling and goat
fronting: embodied affect in sociophonetic variationis’, Proceedings of the International Congress
of Phonetic Sciences, 18.
Podesva, R. J., Roberts, S. J., and Campbell-Kibler, K. (2002) ‘Sharing resources and indexing
meanings in the production of gay styles’, in Campbell-Kibler, K., Podesva, R. J., Roberts, S. J.,
and Wong, A. (eds) Language and sexuality: contesting meaning in theory and practice. Stanford:
CSLI Press, pp. 175–190.
Pratt, T. (2018) Affective sociolinguistic style: an ethnography of embodied linguistic variation in an
arts high school. Ph.D. dissertation, Stanford University.
Steinmetz, K. (2011) ‘Get your creak on: is ‘vocal fry’ a female fad?’, Time. Available at: http://
healthland.time.com/(2011/12/15/get-your-creak-on-is-vocal-fry-a-female-fad/
Trudgill, P. (1972) ‘Sex, covert prestige and linguistic change in the urban British English of Norwich’,
Language in Society, 1, pp. 179–195.
Voigt, R, Jurafsky, D., and Podesva, R. J. (2015) ‘Speaker movement correlates with prosodic
indicators of engagement’, Proceedings of Speech Prosody, 7, pp. 70–74.
Weiss, P. (2013, January 6) Do you have “annoying” girl voice? Yahoo! Shine. Available at: http://us
.lifestyles.qahttp://1p.global.media.yahoo.com/healthy-living/8220-annoying-8221-girl-voice-16
4900387.html
Wolf, N. (2015) ‘Young women, give up the vocal fry and reclaim your strong female voice’, The
Guardian. Available at: https://www.theguardian.com/commentisfree/2015/jul/24/vocal-fry-s
trong-female-voice
Wolk, L, Abdelli-Beruh, N. B., and Slavi, N. D. (2012) ‘Habitual use of vocal fry in young adult
female speakers’, Journal of Voice, 26, pp. e111–e116.
Yuasa, I. P. (2010) ‘Creaky voice: a new feminine voice quality for young urban-oriented upwardly
mobile American women’, American Speech, 85, pp. 315–337.
Zimman, L. (2017) Operationalizing stance: affective stance in a pervasively creaky transgender
speaker. Paper presented at the 91st Annual Meeting of the Linguistic Society of America. Austin,
TX.

36
 3
Sexuality as non-binary
A variationist perspective
Erez Levon

Introduction
The first question I am often asked by non-linguists about my work is whether it is true
that gays and lesbians speak differently than heterosexuals. My answer to this question is
inevitably that ‘it’s complicated’. It is complicated because sexuality is not a homogenous
category, such that we can speak about what all ‘gays and lesbians’ do (in the same way
that we cannot speak about what all of those who are not gay or lesbian do). It is also com-
plicated because there is never a one-to-one correspondence between language and social
identity. Just because someone identifies as a lesbian does not mean that she will necessarily
speak in a particular way any, some, or all of the time. And just because someone speaks in
a particular way any, some, or all of the time does not mean that she is a lesbian. This is not
to say that there is no relationship between sexuality and language. There clearly exist ways
of speaking that are stereotypically associated with different sexual identity categories, and
research over the years has been able to demonstrate that, in some communities at least,
certain linguistic features appear more (or less) frequently in the speech of gays or lesbians
than among their heterosexual counterparts. But identifying surface-level correlations like
these between linguistic features and identity categories can only ever tell us a partial, and
often inaccurate, story. The reality of the connection between language and sexuality is a
more complicated one.
In this chapter, I outline some of the ways in which we can approach this complexity in
our research. Focusing on research within variationist sociolinguistics, I review certain key
studies in the area of language and sexuality and describe the different methodological tools
that have been developed for investigating how variable patterns in language use participate
in the construction of sexuality. In order to do so, I also review some important theoretical
concepts that variationist sociolinguists draw on to frame their analyses, including indexi-
cality (e.g. Ochs 1992), intersectionality (e.g. Crenshaw 1989), and performativity (Butler
1990, 1993). Finally, I close with a brief illustration of some of these theories and methods
‘in action’ by discussing pitch variation within a community of lesbians in London (based
on data collected and analysed by Lawrence 2014). By the end of the chapter, the reader will
understand why it is problematic to treat sexuality as a binary characteristic (i.e. ‘gay’ versus

37
Erez Levon

‘not gay’) in linguistic research. I will also introduce some of the theoretical tools that can
be used to move away from this type of binary thinking, and will provide an illustration of
variationist research on sexuality that places these tools at the centre of its analysis. Before
getting to that, however, I first begin in the next section with a brief overview of the study of
sexuality from a variationist perspective. This is important because it helps to contextualise
the methodological discussion that follows.

Sexuality in variation: from correlation to emergence

We can divide variationist research on sexuality into three basic types based on the kind of
theoretical approach the research takes (see Levon and Mendes 2016 for a more detailed
discussion of this taxonomy). The first type is research using what we can describe as a cor-
relational approach to the subject (Eckert 2012: 94). Correlational research assumes that the
language practices we observe are directly determined by some element of the underlying
social structure. In the case of sexuality, the assumption would be that there exists such a
thing as the ‘gay and lesbian community’, and that membership in this community gives rise
to a set of distinctive social and linguistic practices. This is the perspective that was adopted
by the earliest variationist research on language and sexuality, where studies focused on
identifying the specific phonological, lexical, or discursive features that were believed to
define the unique experiences of lesbians and gays (see Jacobs 1996; Kulick 2000; Queen
2007 for full reviews). Moonwomon (1985), for example, examined pitch differences in
the speech of lesbians and heterosexual women in the US, and found that lesbian speakers
had lower mean pitch levels and lower overall pitch ranges than the heterosexual speakers
she studied. Similarly, Leap (1993, 1996) identified certain conversational features that he
argued were specific to interactions among American gay men. In both cases, the authors
suggested that it was the speakers’ identities as gays and lesbians that caused them to speak
in distinctive ways.
Beginning in the 1990s, a number of developments challenged the assumptions under-
lying this correlational model of language and sexuality. From outside linguistics, Butler
(1990, 1993) popularised the notion of performativity, or the belief that identity is not the
cause of observed behaviour, but rather its result. In other words, we do not act in a certain
way because we are lesbian; we are socially constituted as lesbians because of how we act.
Within linguistics, this concept of performativity was refined by Ochs’ (1992) definition of
indexicality. Arguing that the link between a linguistic feature and a social category is rarely
a direct one, Ochs claimed that features in language index particular stances, acts, and activ-
ities that are then ideologically linked to salient social categories. According to this account,
tag-questions, for example, do not directly index the category ‘woman’. Instead, they are
taken to signal a stance of ‘uncertainty’, which is itself linked to stereotypes of womanhood.
Together, performativity and indexicality gave rise to a new type of constructionist research
on language and sexuality. Rather than attempting to catalogue a characteristic lesbian or
gay way of speaking, research in this paradigm sought to identify how people use language
to actively construct sexual personae. Barrett (1995, 1997), for example, described how
African American drag queens in Texas juxtapose features that are stereotypically linked to
both white women in the US South and African American men in order to variably construct
themselves as gay men, as African Americans, and as drag queens. The crucial point is that
research in this paradigm did not assume that individuals would speak in a particular way
because they were lesbian or gay, but rather that speakers draw on the indexical power of

38
 Sexuality as non-binary

language to construct their sexualities through linguistic practice (see Cameron and Kulick
2003; Livia and Hall 1997 for more detail).
In the same way that the constructionist approach challenged certain underlying tenets of
the correlational perspective, theoretical developments in the mid-2000s began to critique
the assumptions of constructionist work. The crux of this critique was the assertion that the
meaning of variation in constructionist research was very often still reduced to the cultural
formations it was used to construct. ‘Gay language’, for example, though not necessarily
viewed as an inherent correlate of gay identity, was nevertheless understood as that set of
linguistic features used to construct a gay ‘self’. Scholars like Eckert (2008, 2012) argued
that this was problematic because it fails to recognise the multiple possible meanings that a
particular variable can have (and the multiple functions a speaker can use a variable to per-
form). Instead, Eckert developed an emergentist framework for analysing sexuality-linked
variation, in which the focus is on understanding the more local actions speakers use vari-
ation to perform. Podesva (2007), for example, discusses how a man he calls Heath draws
on the ability of falsetto voice to index ‘expressiveness’ to construct distinct personae in
different settings. When at a barbecue with his friends, Heath uses falsetto to help adopt an
expressive stance (i.e. as a person who explicitly expresses his thoughts and emotions) that,
in conjunction with other relevant features, results in the creation of a ‘diva’ style. At the
medical clinic where he works, in contrast, Heath’s use of falsetto serves instead to index
expressivity as part of the creation of a ‘caring doctor’ persona. Crucially, while the same
linguistic feature is deployed in both contexts, the ultimate meaning of the feature, in terms
of the persona it helps to construct, is context-dependent. Moreover, while he acknowl-
edges that the perception of ‘gay identity’ may emerge from Heath’s use of falsetto, Podesva
argues that this is in a sense a potential by-product of Heath’s use of the feature and that the
primary motivation behind Heath’s observed practice is the construction of situationally rel-
evant personae. Research on sexuality within an emergentist paradigm thus does not ignore
the fact that identities may result from variation. But it does not assume that ‘doing identity’
is a speaker’s ultimate aim. Instead, emergentist research looks first at what immediate
interactional goals speakers are trying to achieve and only then attempts to explain how the
linguistic attainment of those goals may link to the emergence of salient social identities
in interaction. In this respect, variationist research within the emergentist paradigm shares
certain similarities with other sociolinguistic traditions for the study of gender (e.g. inter-
actional sociolinguistics, discourse analytic perspectives), though as variationist research
it remains committed to identifying systematic (i.e. quantitative) distributional patterns of
language use.

Sexuality and lived experience

The review in the preceding section is by no means an exhaustive overview of variationist
research on sexuality (for that, see Queen 2013, 2014). It does, however, give a sense of
the kinds of theoretical developments that have taken place, going from seeing language as
the result of identity (correlational), to seeing language as a tool with which to ‘do’ iden-
tity (constructionist), to finally seeing language as an instrument for accomplishing local
interactional goals through which embodied identities also emerge (emergentist). These
developments have allowed us to provide much more nuanced analyses of the relationship
between language and sexuality than would otherwise have been possible. Yet despite these
advances, and as I have argued elsewhere (Levon 2015), our analyses of sexuality have

39
Erez Levon

tended to remain framed in terms of unitary categories of experience (like ‘gay’ or ‘les-
bian’). Thus, while we have developed sophisticated accounts of how particular linguistic
forms take on sexualised meanings and of how those meanings are then recruited by speak-
ers in interaction, we have been somewhat less attentive to the fact that those sexualised
meanings are also simultaneously gendered, classed, raced, and age-, culture-, and region-
specific. Intuitively, we know that individuals do not experience life through the prism of a
single identity category. Each of us maintains multiple affiliations and identifications, and
these different components all influence our own experiences of self. It is therefore both
theoretically and empirically inaccurate to conceive of sexuality in terms of simple binary
contrasts (e.g. homosexual versus heterosexual) since each half of that binary itself encom-
passes a huge range of diverse stances, orientations, and behaviours.
A useful way for dealing with the complexity of sexuality as a lived experience is through
the prism of intersectionality theory. A term originally coined by legal theorist Crenshaw
(1989), and itself drawing on a long history of early Black feminist theorising (e.g. Anthias
and Yuval-Davis 1983; Davis 1981; hooks 1981; Hull, Scott and Smith 1982), intersection-
ality refers to the idea that lived experience cannot be defined in terms of membership in a
single identity category (e.g. ‘woman’, ‘black’). Rather, both our own, inner understandings
of self and the kinds of access, opportunity, and treatment we receive are the product of mul-
tiple and intersecting systems of social classification. Because of this, an intersectionality
perspective argues that no one analytical category is sufficient if we are to provide a rigorous
analysis of the social practices we observe. Instead, we must investigate how a multiplicity
of categories come together in the formation of individual subjectivity.
Since its popularisation in the early 1990s, intersectionality has become the dominant
construct for theorising identity across the humanities and social sciences, including in
fields as diverse as gender studies, sociology, philosophy, and politics (Collins and Bilge
2016; Davis 2008; Lutz, Vivar and Supik 2011; McCall 2005). However, this does not mean
that intersectionality is a unified social theory. There are numerous debates in the relevant
literature about the framework’s key concepts and about how to methodologically imple-
ment an intersectional perspective (cf. for example, Cho, Crenshaw, and McCall 2013;
Choo and Ferree 2010; Weldon 2008). For the purposes of the current chapter, we can
nevertheless identify two main assertions that all forms of intersectional analysis maintain.
The first assertion is that if we assume that lived experience is ultimately intersectional, then
we must place this intersectional complexity at the heart of our analyses. In practical terms,
we can accomplish this by engaging in what Matsuda (1991: 1189) describes as ‘asking the
other question’:

When I see something that looks racist, I ask, ‘Where is the patriarchy in this?’ When
I see something that looks sexist, I ask ‘Where is the heterosexism in this?’ When I see
something that looks homophobic, I ask ‘Where are the class interests in this’.

A deceptively simple method on the surface (Davis 2008), ‘asking the other question’ forces
us to go beyond analyses in terms of categories in isolation to consider how these categories
intersect with equally important others.
The other main assertion of intersectional research is that categories do not only inter-
sect but mutually constitute one another (Choo and Ferree 2010). This is a somewhat more
contentious aspect of the theory (Crenshaw 2011) since it argues that intersections are not
to be viewed as simple points of contact between two (or more) already existing categories
(Shields 2008). Instead, the argument is that intersections are themselves formative of the

40
 Sexuality as non-binary

categories in question. In other words, the idea of mutual constitution suggests that con-
structs such as class, race, and gender do not exist as entities unto themselves. Rather, they
crucially depend for their meaning on their relationship to the other categories with which
they intersect. Thus, there is no ‘gender effect’ to be discovered and analysed in a dataset;
there is only the effect of gender in relation to class, race, etc. This is a strong claim, and
there is debate in the literature as to whether such a strong position is necessary. Without
getting in to the details of this discussion, it is nevertheless important for us to note how
the concept of mutual constitution pushes the envelope of intersectional analysis further,
encouraging us to move beyond seeing things in terms of compartmentalised categories to
focus instead on the relationship between categories as formative of lived experience.

Approaching intersectionality in variationist research

These two assertions – that lived experience is intersectional in nature and that catego-
ries mutually constitute one another – correspond to two main avenues for intersectional
research in sociolinguistics and related fields. In this section, I very briefly review the two
approaches and cite some key examples of work in these areas. For fuller examples of this
type of work, including both non-variationist studies and studies on topics other than sexual-
ity, see Levon (2015) and Levon and Mendes (2016).

Diversity within: intra-categorical intersectionality

In her well-known discussion of methods for doing intersectionality research, McCall (2005)
defined what she terms ‘intra-categorical’ intersectionality, or intersectionality research that
focuses on the diversity of more specific articulations of identity that exist within a given
category. For example, an intra-categorical approach would critique a category such as ‘les-
bian’, arguing that we need to examine the different ways in which the label ‘lesbian’ can
be experienced and lived (e.g. ‘black lesbian’, ‘butch lesbian’, ‘middle-class lesbian’, etc).
This type of intra-categorical approach is central to what intersectionality is about, and it
serves a crucial theoretical, empirical, and political role in bringing to light a variety of
lived experiences that would otherwise be obscured (Crenshaw 1989, 1991; Morgan 2004).
The intra-categorical approach is also the most common form of intersectionality research
within the variationist paradigm (Levon 2015).
I explore intra-categorical intersectionality in some of my own prior research in Israel (e.g.
Levon 2009, 2010, 2011). In that work, which was based on a year-long sociolinguistic eth-
nography of members of 12 gay and lesbian activist groups from across the Israeli politi-
cal spectrum, I demonstrate how groups of Israeli gays and lesbians use pitch differently as
a way of signalling their distinct political affiliations. Those speakers who align with more
centrist political discourses about Israeli society and Israeli nationalism (and whom I refer
to as members of the ‘mainstream’ group) adopt mean pitch levels that align with dominant
Israeli gender norms. In contrast, those who reject normative Israeli conceptualisations of the
nation (who I describe as members of the ‘radical’ group) engage in linguistic practices that
flout Israeli sociolinguistic gender norms. In this way, I illustrate how language varies among
lesbians and gays in Israel as a function of their broader political beliefs and alignments.
Podesva and Van Hofwegen (2016) make a similar argument in their exploration of /s/ varia-
tion among lesbian and gay speakers in rural California. There, they demonstrate that rural gay
male speakers, for example, produce backer articulations of /s/ than gay men in nearby San
Francisco do. Podesva and Van Hofwegen argue that this difference is due to the social context

41
Erez Levon

in which these men live, and the strong pressure to conform to a more normative masculine
style (which includes the use of backer articulations of /s/). The language used by gay men
in rural California is thus not only related to their sexuality (their articulations of /s/ are sig-
nificantly fronter than their heterosexual rural male counterparts). Crucially, it is also affected
by their overall orientations to ‘country’ versus ‘town’ culture. In both of these studies (and
others like them), the main goal of the analysis is to understand how other aspects of the social
context affect the lived experience of (homo)sexuality, and thus demonstrate the impossibility
of thinking in terms of a simple homosexual versus heterosexual binary.
Variationist research has also examined intra-categorical complexity from the perspec-
tive of perception research. Work on this topic attempts to identify the ways in which listen-
ers’ perceptions of sexuality may be affected by other socially salient aspects of a speaker’s
voice. Pharao, Maegaard, Møller, and Kristiansen (2014), for example, discuss how the
identification of a voice as sounding ‘gay’ in Danish depends on its perceived ethnicity,
such that non–white-sounding voices are never perceived as ‘gay’ even when they contain
the same sexuality-linked linguistic features as white-sounding voices (see also Maegaard
and Pharao 2016; Pharao and Maegaard 2017). Similarly, research by Mendes (2016) con-
siders how the use of non-standard plural noun-phrase agreement in Brazilian Portuguese
acts as a marker of femininity in certain voices, but not in others. Like the work by Pharao
and colleagues, Mendes’ results demonstrate that we cannot make broad claims about what
‘sounds gay’ (or ‘lesbian’) in speech. Instead, we need to look at the specific linguistic con-
text in which relevant features occur. Finally, a growing body of work has made the same
type of argument in relation not only to linguistic contexts, but also to social ones. Studies
by Drager (2011) in Hawaii, Mack (2010) in Puerto Rico, and Rácz and Papp (2016) in
Hungary, among others, have all shown that there are specific cultural differences in what
features listeners pick up on when making judgements about a speaker’s sexuality. As in the
production studies described above, research on intra-categorical complexity in perception
serves to highlight the diversity of ways in which one can be (or sound) lesbian or gay. In
doing so, it pushes us to embrace – rather than ignore – this complexity in our own work.

Looking across categories: mutual constitution

The intra-categorical method outlined above succeeds in responding to the first claim of
intersectionality theory, namely that lived experience is itself intersectional and so we must
examine how multiple categories come together to influence observed social practice. It
does not, however, force us to consider how these different categories constitute one another.
For example, in my own work on Israel I was able to describe how the ways in which people
embody identities like ‘gay’ or ‘lesbian’ are influenced by their other social identifications
and affiliations (such as their political beliefs, and particularly how they saw the relation-
ship between sexual politics and national politics more broadly). In doing so, I was able to
identify a point of mutual influence between the categories ‘sexuality’ and ‘political beliefs’,
but I stopped short of describing how these two categories may in fact define one another
(such that sexual identity categories are politicised, and political beliefs are sexualised). To
do this, we need to open up our analytical gaze to look across categories, and allow linguis-
tic features that normally ‘mean’ one thing to be recruited in the service of another. In other
words, one way to embed the idea of mutual constitution in our analyses is to examine how
a feature that we normally think of as related to one identity category (gender, for example)
is used by speakers to help construct a different identity category (e.g. region). I illustrate
below what I mean with two brief examples.

42
 Sexuality as non-binary

Podesva (2011) describes how a speaker named Regan, a 31-year-old gay man in Northern
California, varies his use of certain vocalic features that are all part of the California Vowel
Shift (CVS). The CVS is a coordinated change in a number of different vowels that is cur-
rently underway in California. As its name implies, the CVS is most saliently associated
with region, such that speakers with CVS-shifted vowels are heard as ‘Californian’. In his
study, Podesva examines Regan’s use of CVS vowels in three contexts: out with friends at
a gay bar, in a meeting with his supervisor at work, and at a casual dinner with a friend.
Podesva demonstrates that Regan uses the most advanced realisations of CVS features when
he is out at the bar, the most conservative with his supervisor at work, and an intermediate
level with his friend at dinner. Podesva does not argue, however, that this finding means
that the CVS features index a gay identity. Rather, drawing on the principles of mutual
constitution, Podesva suggests that elements of the CVS system have become enregistered
(Agha 2007) as markers of the ‘fun’, ‘laidback’, and ‘carefree’ lifestyle that is stereotypical
of Californians. Podesva argues that when he is out at the bar, Regan draws on this regional
meaning of the features to help him construct a fun, carefree, and laidback ‘gay partier’
persona. In other words, Regan recruits a regional meaning to help him construct a sexual
one, thus demonstrating how, in Podesva’s research at least, the categories ‘Californian’ and
‘gay’ mutually constitute one another. A similar example can be found in Ilbury’s (2019)
discussion of stylised uses of African American Vernacular English (AAVE) among young
British gay men on Twitter. In that study, Ilbury describes how men who would normally not
use AAVE in their everyday speech employ tokenistic elements of the variety in their tweets
as a way of portraying a ‘sassy queen’ persona. Like Regan, the men in Ilbury’s study recruit
a feature that has come to be stereotypically associated with one social category (in this
case, African Americans) and use it to help construct another (‘sassy queen’). That they do
so relies on various reductive and racist assumptions about the speech of African Americans
and ignores the power imbalances inherent in the appropriation of AAVE by white speakers
(cf., e.g. Hill 2008). Nevertheless, it also illustrates how for these individuals racialised con-
ceptualisations of what it means to be ‘sassy’ partially constitute what it means to be ‘gay’.
Both of these examples illustrate one way of exploring the mutual constitution of catego-
ries by using variationist methods. In both cases, a linguistic feature that we would normally
think of as meaning one thing (‘Californian’, ‘African American’) is strategically deployed
by speakers to mean something else (‘diva’, ‘sassy’). The reason the features can be used in
this way is because it is possible to ideologically elaborate a first-order indexical meaning
to create a new one, such that a feature meaning Californian (first-order) comes to mean
things that we associate with California (e.g. laid-back, carefree) (second-order). Once that
ideological elaboration has taken place, the feature is then available to be used to express
this new meaning (so that if I want to sound ‘laid-back’, I can use that California feature).
What is important for the present discussion is that tracing how these ideological develop-
ments happen – what new meanings are created and how features are put to use in creative
ways – provides us with a window into the internal composition of an identity: it allows us
to see the gendered, regional, classed, racialised, and other dimensions that comprise it. This
is why this type of method is helpful for intersectional analysis.

Adopting an intersectional perspective: lesbians in London

In the previous sections, I outlined some of the basic principles of intersectionality theory
as it has been applied to variationist research and summarised a number of studies that have
engaged with the principles of intersectionality in their analyses. To better enable the reader

43
Erez Levon

to anchor an intersectional perspective in their own work, in this section I provide a more
detailed discussion of an examination of pitch variation in the speech of a group of young
lesbians in London (Lawrence 2014). In my presentation of this work, I highlight the meth-
odological steps the author took to investigate the relevance of intersectionality to her find-
ings. This is not intended as a prescriptive rulebook for how to conduct a study of this kind.
Instead, my aim is to provide a guide for thinking through questions of intersectionality in
relation to a body of data and an illustration of how variationist techniques can be brought
to bear on these issues.

Overall goals
Lawrence (2014) examined various aspects of language use within a lesbian friendship
group in London. Her goal in doing so was to examine the extent to which a salient stereo-
typical divide between ‘butch’ versus ‘lipstick’ articulations of lesbian identity influenced
the ways in which the women understood their sexualities and the kinds of social practices
in which they engaged. Lawrence’s study is thus a clear example of an intra-categorical
approach to intersectionality. What she is interested in is the diversity of lesbian experi-
ences across individuals, even among members of the same friendship group. In this sense,
Lawrence’s work helps us to move beyond simplistic binary assumptions about ‘lesbian’
versus ‘non-lesbian’ ways of speaking, and instead allows us to explore the various other
factors that together with sexuality help constitute individual subjectivity.
The so-called butch–lipstick dichotomy (sometime also termed butch–femme) is a ste-
reotypically very salient one within lesbian communities (and, arguably, within society
at large). The labels themselves refer to two imagined archetypes of lesbian identity that
occupy opposite positions on a spectrum of gendered and sexual presentation (Eves 2004;
Munt and Smyth 1998). While popularly viewed as simple embodiments of more ‘mas-
culine’ versus more ‘feminine’ styles, respectively, scholars have argued that butch and
lipstick identities serve as powerful forms of resistance to heterosexual (and heterosexist)
norms, allowing women to create a lesbian aesthetic within which to reimagine gender
(e.g. Case 1988). Abstracting away from their political potential, the concepts ‘butch’ and
‘lipstick’ denote opposing positions within a complex economy of embodied practices (such
as dress and demeanour), emotional styles, and sexual and romantic roles. In investigat-
ing the butch–lipstick dichotomy, Lawrence (2014) is not aiming to describe distinct and
homogenous ‘butch’ versus ‘lipstick’ styles of speech. Rather, she attempts to understand
how the women she observed orient to these concepts variably in their daily lives, and how
they draw on language as a resource for materialising these orientations in interaction (see
Jones 2011, 2012, 2014 for another discussion of butch–lipstick dynamics among lesbians
in the UK).

Methods
To achieve these goals, Lawrence (2014) focuses on pitch variability among the women.
She considers a number of different acoustic properties of spoken pitch, including mean
pitch (the average pitch level in an utterance) and pitch range (the overall span of pitch
across an intonation phrase). These are both features that have been extensively studied in
the literature on language and sexuality, though the majority of that work has focused on
(gay) men (e.g. Gaudio 1994; Levon 2006, 2007; Smyth, Jacobs and Rogers 2003). In the

44
 Sexuality as non-binary

studies that have examined lesbian speech (e.g. Moonwomon 1985; Pierrehumbert, Bent,
Munson, et al. 2004), results with respect to pitch have been either contradictory or incon-
clusive (see also Munson and Babel 2007). Nevertheless, there is a strong stereotype in the
UK (and elsewhere) that butch women will have lower mean pitch levels and narrower pitch
ranges, while lipstick women will have higher mean pitch levels and wider pitch ranges
(Queen 1997; Van Borsel, Vandaele and Corthais 2013; Waksler 2001). These stereotypes
in relation to the butch–lipstick dichotomy are analogous to the stereotypes that circulate
with regard to the speech of men versus women (as well as the speech of more ‘masculine’
versus more ‘feminine’ men) (e.g. Henton 1989, 1995). For this reason, variation in pitch is
a salient feature to examine among the women in question.
Data are drawn from the speech of 5 women between the ages of 20 and 27 (mean
age: 23) living in the London area who were all members of a London university’s LGBT
society. Lawrence observed and recorded the women both in their casual interactions as a
group (i.e. when out together at a local pub) and in individual sociolinguistic interviews she
conducted with each of them. The analysis of pitch below is based on speech from these
interviews. The interviews themselves were semi-structured, an approach which encour-
ages natural speech and allows interviewees to have some control over the direction and
content of their talk. This helps to encourage more relaxed, informal conversation, and
enables interviewees to speak in more detail about topics they consider important while
simultaneously ensuring comparability across the dataset (e.g. Schilling 2013). All inter-
views included talk about the interviewees’ social backgrounds, schooling, current work/
studies, leisure activities, media consumption habits, and opinions about lesbian life. Once
recorded, interviews were segmented into intonational phrases (IPs) and the mean pitch
level and pitch range of each IP was measured. In addition to these linguistic measure-
ments, IPs were also coded for whether they were taken from talk on ‘gay’ or ‘non-gay’
topics. ‘Gay’ topics include the women’s personal histories of ‘coming out’, their participa-
tion in a lesbian community or ‘scene’, and their opinions about current sexual politics. The
division between ‘gay’ and ‘non-gay’ topics allows Lawrence to investigate whether there
are patterns of topic-linked intraspeaker variation in the dataset (see, e.g. Levon 2009).
Finally, each of the five speakers was also assigned a value on a butch–lipstick index, based
on Lawrence’s observations and ethnographic knowledge of the participants (for further
details of this index, see Lawrence 2014).

Results
Initial results with respect to mean pitch levels are presented in Figure 3.1. Participants are
ordered in terms of their scores on the butch–lipstick continuum, with those rated as more
‘butch’ on the left side of the plot and those rated more ‘lipstick’ on the right side. We see
that the five speakers divide into roughly three groups: Bow with the lowest mean pitch
level (average: 156.4 Hz), followed by Jane and R (average: 186.9 Hz and 184.2 Hz, respec-
tively), and finally Meredith and Lizzie (average: 202.8 Hz and 200.1 Hz, respectively).
What this pattern shows is that, in general, the more ‘lipstick’ a participant’s embodied style
(i.e. the more it conforms to traditional notions of femininity), the higher her average mean
pitch levels. This is confirmed by quantitative regression analysis, which demonstrates that
there is a positive correlation between placement on the butch–lipstick continuum and mean
pitch (p = 0.02). This finding is important because it illustrates that even though all the
women self-identify as ‘lesbians’, their different positionings in relation to the constructs

45
Erez Levon

Figure 3.1 Average mean pitch level in Hz for speakers ordered from left to right by position on
the ‘butch’–‘lipstick’ continuum.

of ‘butch’ versus ‘lipstick’ have an impact on their linguistic practices. This is thus a clear
example of intra-categorical complexity among the women under investigation.
Further investigation of the women’s speech indicates that the pattern observed in
Figure 3.1 is actually more complex than it appears. In Figure 3.2, mean pitch levels are
again shown by speaker, but this time they are also divided by speech topic (‘gay’ versus
‘non-gay’). There, we see that the correlation between orientations to ‘butch’/’lipstick’ and
mean pitch varies depending on the topic of conversation. For gay topics (the dark bars in
Figure 3.2), we again see the relationship evident in Figure 3.1, with the dividing into three
groups (Bow, Jane, and R, Meredith and Lizzie) and a general increase in mean pitch levels
the more a speaker orients to the ‘lipstick’ end of the continuum. However, on non-gay
topics (lighter bars in Figure 3.2), no such relationship exists. Instead, all of the women are
shown to have roughly similar mean pitch levels. This is interesting for a number of reasons.
First, it demonstrates that the women vary how they speak in relation to the topic of their
talk. For this reason, we cannot speak of a characteristic ‘lesbian’ way of speaking, or even
of a characteristic ‘butch’ or ‘lipstick’ way of speaking, since how an individual speaks
clearly depends on other elements on the speech context (like topic). It is also potentially
meaningful that the pattern of interest with respect to mean pitch obtains when the women
are speaking about ‘gay’ topics. This could indicate that the variability in mean pitch that
we find is in some way related to women’s constructions and presentations of particular
sexual selves. Bow, for example, may strategically lower her pitch on gay topics to enact a
more ‘butch’ persona, whereas Lizzie may strategically raise hers in order to enact a more
‘lipstick’ one. While we would want further evidence to support this kind of claim, the quan-
titative pattern in Figure 3.2 is consistent with an analysis in which mean pitch is a symbolic
resource that the women can deploy to enact specific intersectional selves at particular inter-
actional moments (for more on variation as a symbolic resource for enacting a particular
interactional self, see, e.g. Bucholtz 2009; Eckert 2008; Levon 2011; Schilling-Estes 2004;
Sharma and Rampton 2015).

46
 Sexuality as non-binary

Figure 3.2 Average mean pitch level in Hz for speakers ordered from left to right by position on
the ‘butch’–‘lipstick’ continuum and divided by speech topic.

Conclusion
The foundational intersectional principle of moving beyond simple binaries was operational-
ised in Lawrence’s study by first looking at diversity of language use within a group of lesbians
(rather than assuming that there is a singular ‘lesbian’ way of speaking that can be contrasted
with the speech practices of others). Detailed quantitative examinations of mean pitch among
women in the group revealed a correlation between higher average mean pitch and more ‘lip-
stick’ embodiments of lesbian identity. This correlation allowed Lawrence (2014) to detail
the kind of intra-categorical complexity that exists within the group. Further analyses also
revealed this complexity was itself dynamic in nature, only emerging in certain interactional
contexts (i.e. when speaking on ‘gay’ topics in interviews). This more detailed result points to
an understanding of mean pitch as a strategic indexical resource that is recruited by the women
as a way of enacting the different types of lesbian genders with which they identify, and so
enriching the intersectional understanding of the women under investigation.
It is worthwhile noting that Lawrence’s (2014) analysis does not venture into an exploration
of how sexuality or ‘butch’/’lesbian’ personae are mutually constituted by their intersection
with other categories. If we wanted to explore this avenue of enquiry, we could, for example,
ask why it is that mean pitch is the tool that women use to enact differently gendered selves.
In addition to research on pitch and gender/sexuality, prior research has also shown that mean
pitch levels are also associated with perceived clarity, intelligence, sophistication, and even
height. It could therefore be the case that part of what it means to be a ‘lipstick’ lesbian is to
speak in a ‘clearer’ and more ‘sophisticated’ fashion and that one way of doing so is to use an
elevated mean pitch. This type of exploration would be one way of considering how categories
like ‘butch’ and ‘lipstick’ are mutually constituted by relevant others.

Future directions
The preceding discussion of a selection of findings from Lawrence (2014) provides a
brief, but nevertheless useful illustration of how we can approach one of the foundational

47
Erez Levon

principles of intersectionality – intra-categorical complexity – in variationist research

on sexuality. Yet it is the other foundational principle of intersectionality – the mutual
constitution of categories – that I believe is the next frontier in research on language,
gender, and sexuality research. To date, most research within variationist sociolinguistics
that has attempted to move beyond binary ways of thinking has tended to do so by adding
new subdivisions to the categories of identity we consider (e.g. woman > lesbian woman
> ‘butch’ lesbian woman) – a process that Eckert (2014: 530) describes as ‘nesting the
terms of the [gender] binary within each side of the binary’. While this work is crucial in
highlighting under-researched aspects of lived experience, I argue that its theoretical and
methodological potential for overcoming binary modes of thinking is ultimately limited.
What is needed is research that examines how binaries are themselves sustained, what
social forces conspire to scaffold and reproduce binaries in other domains, and, most
crucially, how individuals negotiate, resist, and transform these binaries in situated socio-
linguistic practice (see Davis, Zimman, and Raclaw 2014). Only once we have done so do
I believe that we will be able to provide a fuller analysis of the complex subjectivities of
the individuals we analyse, and of the crucial role of linguistic variation in bringing that
complexity to light.

Further reading
Collins, P. and Bilge, S. (2016) Intersectionality. Cambridge: Polity.
This book provides an up-to-date introduction to the concept of intersectionality and how it has
been used in a variety of different disciplines.
Levon, E. (2015) ‘Integrating intersectionality in language, gender, and sexuality research’, Language
and Linguistics Compass, 9(7), pp. 295–308.
This paper provides an overview of the concept of intersectionality, how it has been treated in
research on language, gender, and sexuality in the past, and suggestions for how to anchor it more
firmly in research in this area.
Levon, E. and Mendes, R. (eds) Language, sexuality and power: studies in intersectional
sociolinguistics. Oxford: Oxford University Press.
The edited collection showcases research on sexuality as it intersects with other social formations,
including religion, culture, and nation, focusing on both language use and perception.
Zimman, L., Davis, J., and Raclaw, J. (eds) Queer excursions: retheorizing binaries in language,
gender and sexuality. Oxford: Oxford University Press.
This edited collection highlights research in sociolinguistics that has attempted to move beyond
binary ways of thinking of gender and sexuality, and features studies utilising both quantitative and
qualitative approaches.

Related topics
Non-binary approaches to gender and sexuality; gender diversity and the voice; perception of
gender and sexuality; gender and sexuality normativities; an ethnographic approach to compulsory
heterosexuality.

References
Agha, A. (2007) Language and social relations. Cambridge: Cambridge University Press.
Anthias, F. and Yuval-Davis, N. (1983) ‘Contextualizing feminism: gender, ethnic and class divisions’,
Feminist Review, 15(1), pp. 62–75.

48
 Sexuality as non-binary

Barrett, R. (1995) ‘Supermodels of the world, unite!: political economy and the language of
performance among African American Drag Queens’, in Leap, W. (ed) Beyond the lavender
lexicon. Amsterdam: Gordon and Breach Science Publishers, pp. 207–226.
Barrett, R. (1997) ‘The ‘homo-genius’ speech community’, in Livia, A. and Hall, K. (eds) Queerly
phrased. Oxford: Oxford University Press, pp. 181–201.
Bucholtz, M. (2009) ‘From stance to style: gender, interaction and indexicality in Mexican immigrant
youth slang’, in Jaffe, A. (ed) Stance: sociolinguistic perspectives. Oxford: Oxford University
Press, pp. 146–170.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. London: Routledge.
Butler, J. (1993) Bodies that matter: on the discursive limits of sex. London: Routledge.
Cameron, D. and Kulick, D. (2003) Language and sexuality. Cambridge: Cambridge University Press.
Case, S. (1988) ‘Towards a butch-femme aesthetic’, Discourse, 11(1), pp. 55–73.
Cho, S., Crenshaw, K., and McCall, L. (2013) ‘Toward a field of intersectionality studies: theory,
applications, and praxis’, Signs: Journal of Women in Culture and Society, 38(4), pp. 785–810.
Choo, H. and Ferree, M. (2010) ‘Practicing intersectionality in sociological research: a critical analysis
of inclusions, interactions and institutions in the study of inequalities’, Sociological Theory, 28(2),
pp. 129–149.
Collins, P. and Bilge, S. (2016) Intersectionality. Cambridge: Polity.
Crenshaw, K. (1989) ‘Demarginalizing the intersection of race and sex: a black feminist critique of
antidiscrimination doctrine, feminist theory and antiracist politics’, University of Chicago Legal
Forum, 1989, pp. 139–168.
Crenshaw, K. (1991) ‘Mapping the margins: intersectionality, identity politics and violence against
women of color’, Stanford Law Review, 43, pp. 1241–1299.
Crenshaw, K. (2011) ‘Postscript’, in Lutz, H., Vivar, M., and Supik, L. (eds) Framing intersectionality:
debates on a multi-faceted concept in gender studies. London: Ashgate Publishing, pp. 221–233.
Davis, A. (1981) Women, race and class. New York: Random House.
Davis, J., Zimman, L., and Raclaw, J. (2014) ‘Opposites attract: retheorizing binaries in language,
gender and sexuality’, in Zimman, L., Davis, J., and Raclaw, J. (eds) Queer excursions: retheorizing
binaries in language, gender and sexuality. Oxford: Oxford University Press, pp. 1–12.
Davis, K. (2008) ‘Intersectionality as buzzword: a sociology of science perspective on what makes a
feminist theory successful’, Feminist Theory, 9(1), pp. 67–85.
Drager, K. (2011) Style and perceived sexuality. Paper presented at NWAV 40, Georgetown University.
Eckert, P. (2008) ‘Variation and the indexical field’, Journal of Sociolinguistics, 12(4), pp. 453–476.
Eckert, P. (2012) ‘Three waves of variation study: the emergence of meaning in the study of
sociolinguistic variation’, Annual Review of Anthropology, 41(1), pp. 87–100.
Eckert, P. (2014) ‘The problem with binaries: coding for gender and sexuality’, Language and
Linguistics Compass, 8 (11), pp. 529–535.
Eves, A. (2004) ‘Queer theory, butch/femme identities and lesbian space’, Sexualities, 7(4), pp.
480–496.
Gaudio, R. (1994) ‘Sounding gay: pitch properties in the speech of gay and straight men’, American
Speech, 69(1), pp. 30–57.
Henton, C. (1989) ‘Fact and fiction in the description of female and male pitch’, Language and
Communication, 9(4), pp. 299–311.
Henton, C. (1995) ‘Pitch dynamism in female and male speech’, Language and Communication,
15(1), pp. 43–61.
Hill, J. (2008) The everyday language of white racism. Oxford: Wiley-Blackwell.
hooks, b. (1981) Ain’t I woman? Black women and feminism. Boston: South End Press.
Hull, G., Scott, P., and Smith, B. (1982) All the women are white, all the blacks are men, but some of
us are brave: black women’s studies. New York: Feminist Press.
Ilbury, C. (2019) ‘“Sassy queens”: stylistic orthographic variation on twitter and the enregisterment of
AAVE’, Journal of Sociolinguistics. Available at: https://doi.org/10.1111/josl.12366.

49
Erez Levon

Jacobs, G. (1996) ‘Lesbian and gay male language use: a critical review of the literature’, American
Speech, 71(1), pp. 49–71.
Jones, L. (2011) ‘“The only dykey one”: constructions of (in)authenticity in a lesbian community of
practice’, Journal of Homosexuality, 58(6–7), pp. 719–741.
Jones, L. (2012) Dyke/girl: language and identities in a lesbian group. Basingstoke: Palgrave
Macmillan.
Jones, L. (2014) ‘“Dolls or teddies?”: constructing lesbian identity through community-specific
practice’, Journal of Language and Sexuality, 3(2), pp. 161–190.
Kulick, D. (2000) ‘Gay and lesbian language’, Annual Review of Anthropology, 29(1), pp. 243–285.
Lawrence, S. (2014) Lesbian language: the influence of ‘butch and lipstick’ identity on the linguistic
behaviour of a lesbian community. Unpublished MA dissertation in Linguistics, Queen Mary
University of London.
Leap, W. (1993) ‘Gay men’s English: cooperative discourse in a language of risk’, New York Folklore,
19, pp. 45–70.
Leap, W. (1996) Word’s out: gay men’s English. Minneapolis: University of Minnesota Press.
Levon, E. (2006) ‘Hearing “gay”: prosody, interpretation and the affective judgments of men’s
speech’, American Speech, 81(1), pp. 56–78.
Levon, E. (2007) ‘Sexuality in context: variation and the sociolinguistic perception of identity’,
Language in Society, 36(4), pp. 533–554.
Levon, E. (2009) ‘Dimensions of style: context, politics and motivation in gay Israeli speech’, Journal
of Sociolinguistics, 13(1), pp. 29–58.
Levon, E. (2010) Language and the politics of sexuality: lesbians and gays in Israel. Basingstoke:
Palgrave Macmillan.
Levon, E. (2011) ‘Teasing apart to bring together: gender and sexuality in variationist research’,
American Speech, 86(1), pp. 69–84.
Levon, E. (2015) ‘Integrating intersectionality in language, gender, and sexuality research’, Language
and Linguistics Compass, 9(7), pp. 295–308.
Levon, E. and Mendes, R. (2016) ‘Introduction: locating sexuality in language’, in Levon, E. and
Mendes, R. (eds) Language, sexuality and power: studies in intersectional sociolinguistics.
Oxford: Oxford University Press, pp. 1–18.
Livia, A. and Hall, K. (1997) ‘“It’s a girl!”: bringing performativity back to linguistics’, in Livia, A.
and Hall, K. (eds) Queerly phrased. Oxford: Oxford University Press, pp. 3–20.
Lutz, H., Vivar, M., and Supik, L. (eds) Framing intersectionality: debates on a multi-faceted concept
in gender studies. London: Ashgate Publishing.
Mack, S. (2010) ‘A sociophonetic analysis of perception of sexual orientation in Puerto Rican
Spanish’, Laboratory Phonology, 1(1), pp. 41–63.
Maegaard, M. and Pharao, N. (2016) ‘/s/ variation and perceptions of male sexuality in Denmark’,
in Levon, E. and Mendes, R. (eds) Language, sexuality and power: studies in intersectional
sociolinguistics. Oxford: Oxford University Press, 88–104.
Matsuda, M. (1991) ‘Beside my sister, facing the enemy: legal theory out of coalition’, Stanford Law
Review, 43(6), pp. 1183–1192.
McCall, L. (2005) ‘The complexity of intersectionality’, Signs, 30(3), pp. 1771–1800.
Mendes, R. (2016) ‘Non-standard plural noun phrase agreement as an index of masculinity’, in Levon,
E. and Mendes, R. (eds), Language, sexuality and power: studies in intersectional sociolinguistics.
Oxford: Oxford University Press, pp. 105–129.
Moonwomon, B. (1985) ‘Towards a study of lesbian language’, in Bremner, S., Caskey, N., and
Moonwomon, B. (eds) Proceedings of the first Berkeley women and language conference.
Berkeley: Berkeley Women and Language Group, pp. 96–107.
Morgan, M. (2004) ‘“I’m every woman”: black women’s (dis)placement in women’s language study’,
in Bucholtz, M. (ed) Language and women’s place: text and commentaries. Oxford: Oxford
University Press, pp. 252–259.

50
 Sexuality as non-binary

Munson, B. and Babel, M. (2007) ‘Loose lips and silver tongues, or, projecting sexual orientation
through speech’, Language and Linguistics Compass, 1(5), pp. 416–449.
Munt, S. and Smyth, C. (1998) Butch/femme: inside lesbian gender. London: Cassell.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context: language
as an interactive phenomenon. Cambridge: Cambridge University Press, pp. 336–358.
Pharao, N. and Maegaard, M. (2017) ‘On the influence of coronal sibilants and stops on the perception
of social meanings in Copenhagen Danish’, Linguistics, 55(5), pp. 1141–1167.
Pharao, N., Maegaard, M., Møller, J., and Kristiansen, T. (2014) ‘Indexical meanings of [s+] among
Copenhagen youth: social perception of a phonetic variant in different prosodic contexts’,
Langauge in Society, 43(1), pp. 1–31.
Pierrehumbert, J., Bent, T. Munson, B. Bradlow, A., and Bailey, J. M. (2004) ‘The influence of sexual
orientation on vowel production’, The Journal of the Acoustical Society of America, 116(4), pp.
1905–1908.
Podesva, R. (2007) ‘Phonation type as a stylistic variable: the use of falsetto in constructing a persona’,
Journal of Sociolinguistics, 11(4), pp. 478–504.
Podesva, R. (2011) ‘The California vowel shift and gay identity’, American Speech, 86(1), pp. 32–51.
Podesva, R. and Van Hofwegen, J. (2016) ‘/s/exuality in small-town California: gender normativity
and the acoustic realization of /s/’, in Levon, E. and Mendes, R. (eds) Language, sexuality and
power: studies in intersectional sociolinguistics. Oxford: Oxford University Press, pp. 168–188.
Queen, R. (1997) ‘I don’t speak spritch: locating lesbian language’, in Livia, A. and Hall, K. (eds)
Queerly phrased: language, gender and sexuality. Oxford: Oxford University Press, pp. 233–256.
Queen, R. (2007) ‘Sociolinguistic horizons: language and sexuality’, Language and Linguistics
Compass, 1(4), pp. 314–330.
Queen, R. (2013) ‘Gender, sex, sexuality and sexual identities’, in Chambers, J. and Schilling, N. (eds)
The handbook of language variation and change. Oxford: Wiley-Blackwell, pp. 368–387.
Queen, R. (2014) ‘Language and sexual identities’, in Ehrlich, S., Meyerhoff, M., and Holmes, J.
(eds), The handbook of language, gender and sexuality, 2nd edn. Oxford: Wiley-Blackwell, pp.
203–219.
Rácz, P. and Papp, V. (2016) ‘Percepts of Hungarian pitch-shifted male speech’, in Levon, E. and
Mendes, R. (eds), Language, sexuality and power: studies in intersectional sociolinguistics.
Oxford: Oxford University Press, pp. 151–167.
Schilling, N. (2013) Sociolinguistic fieldwork. Cambridge: Cambridge University Press.
Schilling-Estes, N. (2004) ‘Constructing ethnicity in interaction’, Journal of Sociolinguistics, 8(2),
pp. 163–195.
Sharma, D. and Rampton, B. (2015) ‘Lectal focusing in interaction: a new method for the study of
style variation’, Journal of English Linguistics, 43(1), pp. 3–35.
Shields, S. (2008) ‘Gender: an intersectionality perspective’, Sex Roles, 59(5–6), pp. 301–311.
Smyth, R., Jacobs, G., and Rogers, H. (2003) ‘Male voices and perceived sexual orientation: an
experimental and theoretical approach’, Language in Society, 32(3), pp. 329–350.
Van Borsel, J., Vandaele, J., and Corthais, P. (2013) ‘Pitch and pitch variation in lesbian women’,
Journal of Voice 27(5), pp. 656e.13–16.
Waksler, R. (2001) ‘Pitch range and women’s sexual orientation’, Word, 52(1), pp. 67–77.
Weldon, L. (2008) ‘Intersectionality’, in Goertz, G. and Mazur, A. (eds) Politics, gender and concepts:
theory and methodology. Cambridge: Cambridge University Press, 193–218.

51
 4
Perception of gender and sexuality
Kathryn Campbell-Kibler and deandre miles-hercules

Introduction/defnitions
Recent work in the third wave of sociolinguistic variation (Eckert 2012) has focused on how
speakers use variation to build social constructs, including gender and sexual identity. Speakers
can use language features like phonetic cues, lexical choices, or syntactic constructions to
invoke contextually constrained concepts like feminine, professional, or friendly (Ochs 1992;
Silverstein 1976, 2003). While production studies examine how and when features are used
by specific speakers in specific contexts, perception studies examine how language features
shape the reactions of observers. This chapter sketches the current literature on sociolinguis-
tic perceptions of gender and sexuality. The field has focused heavily on linking acoustic
characteristics to masculinity and sexual orientation in perceptions of the voices of English-
speaking cisgender men. Some work has extended from this core, exploring other languages,
other speakers, or asking more complex questions about the perceptual landscape of gender
and sexuality. Despite these attempts, by and large, the perceptual literature has not kept pace
with the production literature in developing a deeper and more nuanced approach to the multi-
dimensional landscape of gendered and sexual identities. In particular, perception work has
for the most part retained a basically binary view of gender and has neglected the turn towards
intersectional approaches to gender and race found elsewhere in the literature.
We present a small study of the perception of performative super-masculine types which
explores new methods for examining the landscape of gendered percepts. The results under-
line the role of listener gender ideology in the construction of any percept of a given voice
by a given listener. We suggest that focusing on the listener is one way for perceptual work
to move towards a more complex view of gender and sexuality. Two key concerns in this
respect are disrupting the hegemonic positioning of the gender binary in the analytical
approaches to perception and developing a greater emphasis on speakers of colour and the
intersection of race with gender and sexuality.

Core work in perception: the search for gay speech

The literature on the sociolinguistic perception of gender and sexuality (Munson and Babel
2007) has had a narrower focus than the larger field of language, gender, and sexuality (Gray

52
 Perception of gender and sexuality

2016; Jacobs 1996; Kulick 2000; Podesva and Kajino 2014). A majority of this work has
revolved around the question of ‘gay speech’ or ‘sounding gay’.
Early work in the area used a pathologising lens to examine what features were found
in presumed-to-be-problematic ‘effeminate’ speech (Terango 1966; Travis 1981) or which
features vocal coaching of trans women should concentrate on (Wolfe et al. 1990) to be
perceived by primarily cis audiences as feminine. In the 1990s, research turned to the social
percept called ‘gay speech’, ‘the gay accent’, or ‘the gay lisp’. This work focused first on the
accuracy with which listeners were able to identify men’s sexual orientation, suggesting that
they are able to perform above chance in at least some cases (Carahaly 2000; Gaudio 1994;
Munson 2007). Other work suggested that this perception was not always a reliable reflec-
tion of a speaker’s sexual identity (Jacobs et al. 2000; Lerman and Damsté 1969), but that
it possibly reflected a robust percept that was shared across listeners (Jacobs et al. 2000).
Based on these results, attention shifted to seeking specific acoustic correlates for ‘gay
speech’, asking what specific phonetic qualities prompted gay and/or feminine percepts of
men’s speech (Avery and Liss 1996; Linville 1998; Smyth et al. 2003). This work gathered
a number of potential acoustic cues, including characteristics of /s/ (Avery and Liss 1996;
Campbell-Kibler 2011; Crist 1997; Levon 2007; Linville 1998; Munson et al. 2005), /l/
(Crist 1997), vowels (Avery and Liss 1996; Mack 2010b; Munson 2007, Munson et al.
2005), and fundamental frequency, the primary acoustic correlate of pitch (Avery and Liss
1996; Levon 2007; Munson 2007). This literature has failed to find links between percep-
tions of sexuality and some characteristics of /s/ (Levon 2007; Munson 2007), fundamental
frequency (Avery and Liss 1996; Levon 2007; Munson 2007; Smyth et al. 2003), vowels
(Munson 2007), and speech rate (Avery and Liss 1996). A handful of work has explored
perceptions within specialised subpopulations, including Mormon missionaries (Borders
2015) or military and ex-military personnel (Kirtley 2011).
Women’s voices have gotten significantly less attention. During the heyday of the second
wave of feminism, there was a brief interest in the perception of women’s voices, particularly
with respect to their status as feminist/women’s liberation supporters or not (Giles et al. 1980;
Kramarae 1982; Kramer 1977, 1978). As with men’s voices, listeners showed mixed success
at identifying women’s sexual orientation by voices, with Moonwomon-Baird (1997) show-
ing listeners at chance while Munson et al. (2005) and Munson (2007) documented some
ability to distinguish straight women’s voices from those of lesbian and bisexual women.
Carahaly (2000) found that lesbian and gay listeners were better at identifying gay women’s
voices than straight listeners, an effect not seen in responses to men’s voices. As with men,
perceptions of women’s voices were influenced by some characteristics of vowels, funda-
mental frequency, and /s/, but not all (Munson 2007; Munson et al. 2005).
Trans and non-binary speakers have been examined in the production literature (e.g.
Becker et al. 2015; Gratton 2016; Hall and O’Donovan 1996; Nagar 2008) but not yet in the
perceptual literature, with the exception of Zimman (2013), discussed below.
The mixed results regarding acoustic cues has led to some doubt about the stability of
acoustic markers of masculinities. Zwicky (1997) suggested that men ‘sounding gay’ may
be less about them producing specific cues than not producing other cues heard as mascu-
line. In other words, there may be a relatively narrow window of ‘masculine’ speech, from
which speakers may diverge in a variety of ways, a point echoed by Rieger, Linsenmeier,
Gygax, Garcia, and Bailey (2010), who suggested that perception of men’s sexual orienta-
tion more broadly may function largely on the basis of perceived ‘sex (a)typicality’.
Some empirical results directly support this argument. Brown (2015), using a cue integra-
tion model, found what she referred to as an ‘androcentric template’ whereby the phonetic

53
K. Campbell-Kibler & d. miles-hercules

cues are perceptually understood in terms of male-centred norms. Similarly, Mack and
Munson (2012) suggest that a gay percept can be prompted by multiple types of divergence
from a neutral or normative /s/ articulation. Munson et al. (2005) found that male talkers
heard as more gay-sounding were also heard as shorter and having clearer speech, with the
opposite pattern found for straight vs. lesbian and bisexual women. This finding offers sup-
port for a style-based analysis along the lines of Ochs (1992), who argues that most semi-
otic resources used in gendered styles are indirectly, not directly, linked to gender. Rather,
linguistic resources (for example, expanded vowel space) may be linked to social qualities
(for example, precision). Gendered styles are then mutually constituted by/with these quali-
ties and activities, through cultural conventions which mark precision as a feminine trait in
particular contexts and for particular reasons. Support for this style-based model (Coupland
2007; Sharma 2018) may be seen also in the finding by Carahaly (2000) that ‘gaydar’ was
improved when the stimuli used (for all speakers of any sexual orientation) were taken
from a conversation with an unfamiliar same-sex gay/lesbian interlocutor. Carahaly’s talk-
ers self-reported code-switching based on the sexual orientation of their interlocutors, echo-
ing Podesva’s (2007) argument that individual personae combine personal identity with
situational goals and constraints. Carahaly’s perceptual results suggest that these situational
speech patterns draw on broader patterns of meanings which naive listeners can perceive.
Taken overall, the perceptual literature has documented a stable and cross-linguistically
widespread percept of ‘gay speech’, in which men’s voices are perceived as gay-sounding
or effeminate. Women’s speech, while comparatively understudied, has shown less of a
stable percept, suggesting a stronger influence of linguistic stereotyping in the perception of
gender in men’s speech.

Expanding beyond English ‘gay speech’

The focus on how English-speaking men’s voices are perceived has overwhelmingly
dominated the literature on the sociolinguistic perception of sex and gender. Some work,
however, explored other languages, other social evaluations, listener variability, and inter-
sections between multiple variables.
Gendered variation has been studied perceptually in Danish (Pharao et al. 2014, discussed
more below), Afrikaans, and South African English (Bekker and Levon 2017), Hungarian
(Rácz and Papp 2015), Spanish (Chappell 2016; Mack 2010b, 2010a, 2015; Walker et al.
2014), and Brazilian Portuguese (Mendes 2014; 2015b, 2015a, 2016). Expanding beyond
native listeners, Hardeman (2013) and Hardeman Guthrie (2017) examined L1 and L2
Mandarin listeners’ perceptions of L1 and L2 talkers’ use of sajiao, a marked ‘cute’ femi-
nine style. Boyd (2018) examined reactions of English-, French-, and German-speaking
listeners to talkers speaking English, French, German, and Estonian. While the details of
linguistic cues vary, by and large cross-linguistic work has found similar themes, with per-
ceptions of gay identities in men’s speech linked to cues associated with femininity, with a
particular emphasis on the acoustics of /s/.
Some work has examined how a listener’s own variety influences their perception of that
of others. Hardeman (2013) and Hardeman Guthrie (2017) found that L1 Mandarin speakers
were more sensitive to the ‘cuteness’ of sajiao than L2 speakers. Walker et al. (2014) found
that despite having less /s/ aspiration themselves, male Mexican listeners joined Puerto
Rican counterparts in rating aspirated tokens more masculine and less gay-sounding, while
female Mexican listeners gave the opposite assessment. Bekker and Levon (2017) showed

54
 Perception of gender and sexuality

that perception of /s/ fronting in South Africa is mediated by talker language and gender,
with men showing a shared effect of /s/-fronting whether speaking Afrikaans or South
African English while women’s evaluation effects differ across the two varieties.
Finally, some work exploring the interaction of multiple variables on each other has
examined gendered cues (Campbell-Kibler 2011; Levon 2007), illuminating the complex
interplay between features to build styles. Pharao et al. (2014) manipulated /s/ in ‘modern
Copenhagen speech’ vs. ‘street language’, styles embedded with ideas of ethnicity, citizen-
ship status, place of residence, among other things. The gendered meanings of fronted /s/ in
the ‘modern’ guise were radically reduced when juxtaposed with the ‘street’ guise (Pharao
et al. 2014; Pharao and Maegaard 2017).
These developments represent exciting new pathways for perception studies. More fun-
damentally, however, perceptual research on variation and gender has struggled with articu-
lating a clear and motivated understanding of gender itself. Two crucial theoretical turns
have been largely neglected in the field, namely the challenging of the gender binary and the
recognition of how the intersections of identity categories impact the experiences of those in
them, particularly in terms of how race shapes gender.

Challenging a binary view of gender

Recent studies of language, gender, and sexuality have begun to consider the role and flaws
of the gender binary in sociolinguistic research, but the binary remains a central and often
unquestioned construct in much perception work. Eckert (2014) considered the socially
constructed nature of identifying individuals as men or women, exclusively, while noting
‘gender correlates with variation on a large scale’. Zimman, Davis, and Raclaw (2014) prob-
lematised the extent to which binaries of gender and sexuality have misarticulated the posi-
tions of individuals that do not map onto normative gender and sexual identities. However,
most perception studies continue to assume that categories like man and woman (or male
and female) are universal and legible without explanation or exploration, despite longstand-
ing literature to the contrary (e.g. Butler 1990). Individuals that do not exist within this
binary are implicitly marginalised by this structure of research. Sociocultural anthropology
has marked the significance of gender identities such as hijra, berdache, and trans resist-
ing two-gender system classifications (Herdt 1994). Sociolinguistic perception research
has struggled to keep up with complexities introduced elsewhere, with only a handful of
exceptions.
Zimman (2013), one such exception, hypothesised that trans men might be predicted to
be particularly gay sounding, since childhood socialisation into feminine speech patterns
could contribute to being ‘gay-sounding’ as an adult. He noted, however, that the trans men
he has worked with have a range of stances towards the potential for being perceived as a
gay man, including those who are gay men, those for whom the percept is congruent with
a different queer identity, and those for whom it is a misidentification they would prefer to
avoid.
Relatedly, Zimman (2017) has shown that the deployment of individual features may not
only support the performance of a categorical identity such as ‘male’, but may be deployed
to mitigate or disrupt such a performance. He discusses the complex interplay of pitch and
/s/ production among transmasculine speakers, some of whom use /s/ tokens with com-
paratively higher-frequency centres of gravity to construct a queer, fem, or non-normative
masculinity, mitigating the masculinising effects of a lowered f0.

55
K. Campbell-Kibler & d. miles-hercules

Working along similar lines, Steele (2019) interviewed 20 non-binary gendered speak-
ers and linked their discussions of their gendered style to sociophonetic patterns. Steele
found that Black speakers showed distinct patterns of /s/ production, linked to their differing
responses on a self-assessment of masculinity visual analogue scale. Interviews with the
speakers revealed a heightened awareness of the threat perceived by others of perceptually
masculine Black people. This work points to the value of non-binary speakers for helping
illuminate gendered sociolinguistic patterning, but also underlines the crucial role of race in
gendered sociolinguistic presentation, which we address in the next section.

Intersections of race and gender

The framework of intersectionality is key for the perception of gender and sexuality.
Crenshaw (1991) pointed out that where oppressive structures intersect, their combined
effects on lived experience may differ profoundly from the effects of only one of those
structures, for example the differences in how Black women and white women experience
sexism and gendered violence. This focus forms a cornerstone of Black feminist theory’s
epistemological resistance to monolithic analyses of gender or race in isolation. From the
erasure (and resignification) of gender difference in the objectification of captive Black
women during the Middle Passage (Spillers 1987) to gendered and racialised public criti-
cism of Serena Williams contemporarily (Douglas 2002), hegemonic enactments of social
identity consistently depict racialised subjects as deviant vis-à-vis gender and sexuality, if
not irrelevant to the question altogether; studies of language and gender have generally fol-
lowed suit.
Scholars of African American Women’s Language (AAWL) have led the broaching of
linguistics’ silences in that regard. Responding to Robin Lakoff’s (1973) influential text,
Marsha Houston Stanback (1985) penned ‘Language and Black woman’s place’, explicat-
ing language researchers’ failure to attend to Black women’s particular subjectivities as they
depart from white women’s and Black men’s. Morgan (1996) analysed the gendered func-
tion of indirectness in conversational signifying amongst African American women, and
Jacobs-Huey (2006), focusing on the sociocultural integration of language, embodiment,
and identity, documented how the politics of race, gender, and hair care are woven into
quotidian interactions across contexts. Lanehart (2009a) comprises the to-date definitive
account of AAWL, including studies of its discourse markers, intonation, regionality, and
media representation.
Only a handful of variationists have drawn multiple axes of identity together explicitly.
Levon (2015) provided a programmatic discussion of how variationist approaches to lan-
guage, gender, and sexuality might effectively incorporate intersectionality. In addition to
applying Black feminist theory’s insights in a broader theoretical light, however, we also
hear its call as an impetus to see and account for race and its associated power structures in
our work. The first and most obvious need for the field in terms of race is simply the need
to engage with it, and to work to disrupt the treatment of whiteness as default that pervades
the majority of scholarship across all fields.
De (2017) is one of the few perceptual studies of gender and sexuality to explore race
and its intersections. De found that while fronted /u/ was rated both as more feminine and
more white-sounding across all talkers, it made white talkers sound more competent, but not
Black talkers, particularly men.
Outside of sociolinguistics, in Remedios, Chasteen, Rule, and Plaks (2011), predomi-
nantly white perceivers gave higher likeability ratings for faces of straight white men than

56
 Perception of gender and sexuality

gay white men. However, these responses were reversed for perceptions of the faces of
Black men, with gay men’s faces drawing higher ratings that straight men’s. This result
shows the intricate ways that hegemonic constructions of gender and sexuality are tied to
race and the racialised history of our current understandings of gender. Our understanding of
how language forms contribute to the construction of gendered styles is necessarily incom-
plete without a discussion of how the construct of race has shaped those styles.
A profound consequence of centring whiteness in sociolinguistics has been to shape the
questions asked, so that even when the speech of people of colour is explored in the litera-
ture, it has primarily been in counterpoint to discussions of whiteness. As Lanehart (2009b)
attests: ‘… though whiteness is seen as the norm in society and subsequently the stand-
ard for all judgements, comparisons, epistemologies, ontologies, etc., whiteness is not the
absence of bias—it is the presence of particular biases’. Broader inclusion of and support
for scholars of colour is one necessary precursor to studies of gender and sexuality which
emerge from organic questions about the sociolinguistic behaviour and experiences of peo-
ple of colour.

Current study
To conceptualise gendered sociolinguistic behaviour without relying on a binary structure
and without neglecting intersectional effects, we turn to two key third-wave theoreti-
cal tools, namely styles and personae. Styles are clusters of linguistic and non-linguistic
semiotic resources, such as phonetic cues, lexical items, clothing, and make-up practices,
or body hexis, which are be tied to each other and particular places and times and which
signify sets of social meanings such as group identities (Coupland 2007; Sharma 2018).
They also include personae, which are similarly clustered sets of resources through which
individuals create situationally appropriate selves, so that a single person might present
their ‘professional doctor’ in one context and ‘prissy diva’ in another (Podesva 2007).
Personae are necessarily developed in communication with broadly circulating charac-
terological figures of personhood, such as the Valley Girl, football star, or distinguished
professor (Agha 2007: 177). These figures provide marker points in multi-dimensional
social space which speakers can use to orient themselves for their own identities and for
understanding those of others.
The current study explores perceptions of intentional performances of hyper-masculine
styles. Talkers, all native speakers of American English, were asked to read a list of sen-
tences in a manner ‘as masculine as possible, whatever that mean[t] to [them]’. While we
conduct an acoustic analysis of the performances and examine the perceptual contributions
of those cues, our primary goal is to better understand the terrain of highly masculine arche-
types that the speakers and listeners (all students at the same university) work with. In
so doing, we explore alternative methods of sociolinguistic perception research, to help
address the constraints noted above.

Methods
The stimuli for this study were taken from the OhioSpeaks corpus (Wanjema et al. 2013),
from a task designed for students in language and gender courses. Students were given a
word list, sentences, and passages to read in three successive guises: in their own voice,
‘as feminine as possible’, and ‘as masculine as possible’. The latter two guises were
described additionally with the phrase ‘whatever that means to you’. These recordings were

57
K. Campbell-Kibler & d. miles-hercules

subsequently used as part of a class presentation and discussion. Wanjema et al. (2013)
describes the system and tasks in more detail.
For the current study, sentence performances of self-identified men performing the
masculine guise were selected. Because the speakers recorded themselves in a variety of
situations and with a variety of equipment, the performances were judged for clarity of
recording and those with significant problems were excluded. This resulted in 68 record-
ings, each consisting of the same set of short sentences such as ‘Cute shoes!’, ‘That’s
my mom, hold on’, and ‘I thought so’. Fifty talkers self-identified as white, seven as
Asian, four as Black, four as multiracial, two as another racial or ethnic category, and
one declined to describe their race/ethnicity. An acoustic analysis was conducted on both
the masculine performances used as stimuli and the corresponding ‘regular voice’ sen-
tence recordings from the same speakers. These values were not ultimately predictive
of responses, however, and our discussion focuses on the patterns within the evaluations
across the full set of stimuli.
Perceptual responses to the stimuli were collected in and around the Columbus main
campus of the Ohio State University. Potential participants visually identified as young
adults were approached in public places and asked if they had ten minutes available to par-
ticipate in a research study. Those that agreed were taken through a verbal consent process,
then asked for their age, regional history, and their racial and/or ethnic identification. They
were then asked to listen to recordings through noise-cancelling headphones.
Participants were told that the voices they heard would be of speakers auditioning for an
acting role. They were asked to listen and form a picture of the character the speaker was
attempting to portray. Each participant was asked to listen and respond to three stimuli,
presented in a consistent order, with stimuli balanced across listeners. In the open-ended
study, each listener was asked to offer three words or phrases that describe the character
being portrayed. In the second, they were asked to provide ratings on a scale of 1 to 10 of the
intended character with respect to the pairs: not educated/educated, kind/mean, thoughtless/
thoughtful, outgoing/shy, lazy/energetic, straight/gay, and feminine/masculine. Participants
reported their ratings verbally, but were shown a piece of paper listing the scales throughout
the task. After offering these character ratings, listeners were asked to rate the skill of the
speaker’s performance, also on a 1 to 10 scale. All verbal responses were entered into a
spreadsheet on the same tablet used to play the stimuli.

Open-ended responses
Open-ended responses were used to understand the range of types and traits prompted for
listeners by the stimuli. They were also used as the basis for creating the ratings for the next
task. Responses were examined for common themes, shown in Table 4.1.
The most common response from listeners was to note the low affect of the speakers (e.g.
boring; chill; disinterested; easy-going; indifferent; mellow; no emotion; not enthusiastic),
a result either of the portrayal of super-masculine speech as noticeably reserved, or a result
of the speakers themselves bringing decreased energy to their online homework recording
task. It is also possible that the low-affect recordings leave less to say, creating a larger
consistent block of comments, while the more energetic performances prompt a wider range
of comments. Other common observations include noting positive (nice; friendly; relatable;
likeable) or negative traits in the characters portrayed (aggressive; sarcastic; brusque; harsh
and mean). The terms offered in the open-ended responses were used to develop the scales
for the next task, the rating task.

58
 Perception of gender and sexuality

Table 4.1 Common comment types

normal 3.46
south/country 3.46
monotone 3.90
other 6.06
young 6.06
high affect 7.36
masculine 8.23
bad affect 11.26
positive affect 12.12
speech comment 12.55
low affect 25.54

Ratings
Our first goal in analysing the ratings was to note key differences in how listeners approach
the task, and to better understand how these differences illuminate the gendered ideologies
which shape the process of person perception. Our focus was on the relationship between
‘gay’ and ‘masculine’ ratings, since the literature has struggled with these response tasks
and their connection. On the one hand, both previous literature and non-academic discus-
sion have stressed the importance of disentangling gendered self-presentation and sexual
orientation or patterns of preference. In the simplest terms, it is not only possible, but com-
mon, for gay men to self-present with cues coded as masculine and for straight men to self-
present with cues coded as feminine, whether linguistic (e.g. /s/ production) or nonlinguistic
(e.g. body hexis). Perceptual work, however, has found that a majority of evaluations show
correlations between these conceptually distinct axes.
Smyth et al. (2003) observed that their results showed a strong correlation between the
two scales, but that the preferred ranges used by listeners differed. They suggested on that
basis that the two responses reflect distinct perceptual constructs despite their close cor-
relation. More intriguingly, they found that while neither masculine/feminine scores nor
straight/gay scores were predicted by pitch, the difference between the two scales was.
Higher-pitched voices showed a closer correlation between the two ratings, while responses
to lower-pitched voices were more likely to allow for higher ‘gay’ ratings alongside more
‘masculine’ ratings. Munson (2007) also differentiated between the two constructs, suggest-
ing that they may be tied to different acoustic factors, although he notes that the evidence in
his study for a dissociation is stronger for male than female talkers.
As a result, we turn first to the responses to these two scales, whose distributions are
shown in Figures 4.1 and 4.2.
In these graphs, we can see several general trends. First, a clear bias towards the high end
of the ‘masculine’ scale shows that the listeners are responding to the efforts of the talkers to
‘sound as masculine as possible’. Consistent with previous research on the negative associa-
tions between the scales, ‘gay’ ratings are concentrated on the low end of the scale. In addition
to these overall distributions, we see a few marked peaks, which disrupt the otherwise rela-
tively smooth distributions. These appear to consist of listeners ‘opting out’ of the rating task
for these scales. We followed the first author’s established practice of using an even-numbered
scale to discourage use of the middle of the scale as a pseudo-neutral answer. However, the

59
K. Campbell-Kibler & d. miles-hercules

Figure 4.1 Distribution of gay ratings.

Figure 4.2 Distribution of masculine ratings.

60
 Perception of gender and sexuality

use of a verbal response and the highly familiar 1 to 10 scale appears to have defeated that
goal, with five perceived by listeners as a midpoint. Of respondents, 16% (38) answered 5 for
a ‘gay’ rating of all three speakers, suggesting that, despite not being the mathematical middle
of the scale offered, they perceived 5 as a neutral response to this question.
A similar pattern emerges in the ‘masculine’ scale, with a number of respondents pro-
viding exactly the same ‘masculine’ rating for all speakers. Interestingly, however, for this
scale 5 is not perceived to be a neutral response, perhaps due to its being markedly low on
the range used by listeners overall. Instead, 9% (21) of the listeners chose ten for all three
‘masculine’ ratings while 7% (17) chose 8. Only five of these listeners also responded with
5 for all three ‘gay’ ratings.
For all participants with variability in both scales, the correlation between the items ‘gay’
and ‘masculine’ was calculated based on the three ratings each listener completed. The his-
togram of this metric is shown in Figure 4.3.
Based on this distribution, these listeners were divided into the categories ‘strongly nega-
tive’, ‘moderately negative’, ‘flat’, and ‘positive’. with the boundaries indicated on the graph.
This measure was used to capture the diversity of listeners’ pre-existing understandings of
gender and sexuality and how those understandings interact with the performances they hear.
The majority of listeners display the commonly-seen negative association, where high ‘gay’
ratings correspond to low ‘masculine’ ratings. However, listeners vary along this dimension,
with a substantial minority reversing the trend. Gender had no effect on this distribution but
Black listeners showed significantly less of a connection between ‘gay’ ratings and lowered
‘masculine’ ratings compared to white listeners (p = 0.042), with listeners of other racial iden-
tities falling between these two groups, not differing significantly from either.

Ratings analysis: clusters

A central goal of the study is to map out identifiable archetypes or characterological figures
that our speakers might be drawing on and our listeners perceiving. To that end, we turned to

Figure 4.3 Distribution of listener slopes between gay and masculine.

61
K. Campbell-Kibler & d. miles-hercules

cluster analysis of the responses to identify areas of the response space that might offer clues
to such types. Both a clustergram (Schonlau 2002) and an elbow plot were used to determine
the ideal number of clusters, which was set at four. The clusters were created using k means
clustering in R, using the kmeans() function.
Although calculated in seven dimensions, for the seven rating items, the clusters can
easily be summarised into two dimensions, as shown in Figure 4.4. This is due to the simi-
larity of profile of feminine/masculine and gay/straight, and the similarity of all five of the
other items. In this essentially two-dimensional space, the four clusters capture two ‘high
masculine’ and two ‘moderately masculine’ percepts, where each pair consists of one more
positive (more educated, kinder, more thoughtful, more outgoing, and more energetic) and
one more negative characterisation.
The different listener patterns contribute significantly differently to the four clusters, as
shown in Table 4.2.
Table 4.2 shows that responses with the hegemonic negative masculine/gay correlation
are split relatively evenly across the four clusters. In contrast, nearly half the positively
correlated responses fall into the cluster of negative, moderately masculine performances.
In other words, given the performatively masculine stimuli, a positive correlation between

Figure 4.4 Perceptual clusters.

Table 4.2 Perceptual clusters, by listener gay/masculine pattern. Bold indicates largest cluster in the
subset, italic the smallest

Strong Neg. Mod. Neg. Flat Positive Gay 5s Masc 8s Masc 10s

negative Masc 0.29 0.23 0.22 0.24 0.12 0.33 0.36

negative Mid 0.24 0.33 0.29 0.44 0.43 0.10 0.08
positive Masc 0.26 0.26 0.37 0.19 0.14 0.44 0.42
positive Mid 0.21 0.18 0.12 0.13 0.31 0.12 0.14

62
 Perception of gender and sexuality

these traits is driven more by moderate responses to perceptually moderately masculine

performances rather than an active embracing of gay, highly masculine identities.

Discussion
This study offers a few points for the perception literature. First, it shows that perceptions of
performative hyper-masculinity show a very simplistic perceptual structure, focused around
two dimensions: degree of hyper-masculinity and a general positive–negative assessment.
Second, it shows that individual listeners do differ in the ideological structure underlying
their person perception, particularly with respect to the link between gender performance
and sexuality. While the majority of listeners display the hegemonic pattern linking gay men
to less masculine gendered performances, this is not the case for all listeners. Black partici-
pants’ differing patterned assessments of ‘gay’-ness and ‘masculine’-ness in our samples
suggest that racial identity – and likely other axes of social identification – can influence the
perception of gender and sexuality in meaningful ways. Put another way: as intersections of
identities structure linguistic production, so too do they permeate perception.
Finally, these data show little support for any specific acoustic cues used by listeners.
This suggests either that listeners are drawing on cues not covered in the previous literature
or that different listeners are responding to different cues, leaving each individual cue with-
out strong support.

Future directions
Building on the discussions above, we suggest that perceptual work can contribute sig-
nificantly to probing the gender binary. We want to encourage continued exploration of the
dynamic relationship between gender and sexuality, while acknowledging the challenges
this presents in experimental work. One step is, of course, not to assume that listeners will
consistently conflate the two. Additionally, however, investigating a broader range of gen-
dered styles will broaden listener responses. Finally, we would like to support Lanehart’s
(2009b) and Levon’s (2015) calls for attention to the intersections. Perceptual work on lan-
guage and gender has been slow to engage with race in a serious way. More attention needs
to be paid to the intersections of race, gender, and sexuality.
Perceptual work needs more nuanced methodological frameworks for interpreting gen-
der’s relation to language variation and perception, including attention to non-binary indi-
viduals, who are currently understudied in linguistics. One of the challenges for perceptual
research is designing studies which can capture ideological differences across participants.
For example, theoretical discussions of gender differentiate between ‘masculine’ and ‘femi-
nine’ as distinct constructs and not ends of the same continuum (e.g. Bem 1977; Connell
1995). The sociolinguistic perception work has, however, consistently shown strong cor-
relations between the two.
Without continued interrogation, research which conflates gay identity in men with femi-
ninity risks maintaining heterosexist ideology rather than illuminating it. Zimman (2013)
argues that ‘there is no single gay-sounding style, but rather a multiplicity of styles that
can be interpreted as indexing sexuality by virtue of their departure from normative mas-
culinity’. A style-based model creates greater flexibility in capturing the various possible
connections between gendered performances and sexual practices, as distinct styles relate
to gender, sexuality, and other constructs differently. One initial question that arises along
this line of inquiry is whether the construct ‘femininity’ is the same when discussed in the

63
K. Campbell-Kibler & d. miles-hercules

context of (gay) men as when it is discussed in the context of women (see, for example
Barrett 1999). Critiquing this conflation has long been a concern of the field, with Gaudio
(1994) positioning his work as ‘critiquing the use of the terms feminine and especially
effeminate in characterising gay male speech’. Nonetheless, these constructs have not been
explored as fully as is needed, and we must inspect more closely how the categories used to
study gender and sexuality are structured by speakers and listeners.
We note that one of the challenges that has prevented this move in the subfield is the
difficulty of exploring complex variability through experimental paradigms. New methods
are needed to capture listener behaviour and specific subpopulations need to be sought out,
to allow for a broader range of listeners informing our models. One specific example is a
greater interest in exploring the language and perspectives of people of colour, who have
been radically understudied in the perceptual domain of language and gender, as in many
subfields of linguistics. Avoidance of widespread engagement with race and racism in lin-
guistics has also contributed to the continued marginalisation of voices of colour. To date,
the majority of linguistic scholarship focusing on race has been concerned largely with
production. While the study of language upon these axes has been invaluable to the field, it
also has had a way of othering what are already marked and stigmatised varieties. In other
words, by showing that race is only relevant when it comes to analysing the speech of peo-
ple of colour (e.g. Black Englishes, Spanglish, etc.) a lack of serious engagement with race
across the subfields of linguistics further normalises whiteness. Fully addressing this prob-
lem, of course, requires attention beyond individual research choices to the broader politics
of research, in considering who is supported, hired, and published in the field.

Acknowledgements
Many thanks go to Alec Buchner, Jordan Maier, Rebecca Wiley, Carter Taylor, and Ashley
Rambacher for their assistance.

Further reading
Lanehart, S. (ed) (2009) African American women’s language: discourse, education, and identity.
Cambridge: Cambridge Scholars Press.
The contributions comprising this essential volume form the only sustained simultaneous treatment
of Blackness and womanness by linguists to-date.
Levon, E. (2015) ‘Integrating intersectionality in language, gender, and sexuality research’, Language
and Linguistics Compass, 9(7), pp. 295–308.
This programmatic article offers a cogent summary of intersectionality aimed at a linguistics
audience.
Marsilli-Vargas, X. (2014) ‘Listening genres: the emergence of relevance structures through the
reception of sound’, Journal of Pragmatics, 69, pp. 42–51.
What might be called an anthropological view on perception is outlined here through the concept
of listening genres. As the field develops novel approaches to perception, insights from related
disciplines will be invaluable.
Pharao, N., Maegaard, M., Møller, J. S., and Kristiansen, T. (2014) ‘Indexical meanings of [s+]
among Copenhagen youth: social perception of a phonetic variant in different prosodic contexts’,
Language in Society, 43(1), pp. 1–31.
This paper is one of the key contributions showing that gendered meanings of variation are
dependent on stylistic context.

64
 Perception of gender and sexuality

Zimman, L. (2013) ‘Hegemonic masculinity and the variability of gay-sounding speech: the perceived
sexuality of transgender men’, Journal of Language and Sexuality, 2(1), pp. 1–39.
Zimman explores listeners’ gendered perceptions of cis gay men, cis straight men and transmasculine
speakers, helping prompt the field beyond traditional binaries.

Related topics
Sexuality as non-binary: a variationist perspective; non-binary approaches to gender and sexuality;
gender diversity and the voice; gender and sexuality normativities; an ethnographic approach to
compulsory heterosexuality.

References
Agha, A. (2007) Language and social relations. Cambridge: Cambridge University Press.
Avery, J. D. and Liss, J. M. (1996) ‘Acoustic characteristics of less-masculine-sounding male speech’,
Journal of the Acoustical Society of America, 99, pp. 3738–3748.
Barrett, R. (1999) ‘Indexing polyphonous identity in the speech of African American drag queens’,
in Bucholtz, M., Sutton, L. A., and Liang, A. C. (eds) Reinventing identities: the gendered self in
discourse. New York: Oxford University Press, pp. 313–331.
Becker, K., Kahn, S., and Zimman, L. (2015) Creaky voice in a diverse gender sample: challenging
ideologies about sex, gender and creak in American English. Paper presented at New Ways of
Analyzing Variation 44.
Bekker, I. and Levon, E. (2017) ‘The embedded indexical value of /s/-fronting in Afrikaans and South
African English’, Linguistics, 55(5), pp. 1109–1139.
Bem, S. L. (1977) ‘On the utility of alternative procedures for assessing psychological androgyny’,
Journal of Consulting and Clinical Psychology, 45(2), p. 196.
Borders, D. S. (2015) The role of gender socialization and sibilants in the perception of gay-and
straight-sounding voices: a study of returned latter-day saint missionaries in Utah. PhD
dissertation, The University of Utah.
Boyd, Z. (2018) Cross-linguistic variation of /s/ as an index of non-normative sexual orientation and
masculinity in French and German men. PhD dissertation, University of Edinburgh, Edinburgh.
Brown, L. (2015) Phonetic cues and the perception of gender and sexual orientation. PhD dissertation,
University of Toronto (Canada).
Butler, J. (1990) Gender trouble. Abingdon: Routledge.
Campbell-Kibler, K. (2011) ‘The sociolinguistic variant as a carrier of social meaning’, Language
Variation and Change, 22(3), pp. 423–441.
Carahaly, L. (2000) Listener accuracy in identifying the sexual orientation of male and female
speakers. Master’s thesis, Department of Speech and Hearing Science, the Ohio State University,
Columbus, OH.
Chappell, W. (2016) ‘On the social perception of intervocalic /s/ voicing in Costa Rican Spanish’,
Language Variation and Change, 28(3), pp. 357–378.
Connell, R. W. (1995) Masculinities. Berkeley: University of California Press.
Coupland, N. (2007) Style: language variation and identity. Cambridge: Cambridge University Press.
Crenshaw, K. (1991) ‘Mapping the margins: intersectionality, identity politics, and violence against
women of color’, Stanford law review, 43, pp. 1241–1299.
Crist, S. (1997) ‘Duration of onset consonants in gay male stereotyped speech’, University of
Pennsylvania Working Papers in Linguistics, 4, p. 3.
De, Z. (2017) Race, gender, and /u/: social perceptions of a non-stereotype feature. Master’s thesis,
The Ohio State University.
Douglas, D. (2002) ‘To be young, gifted, black and female: a meditation on the cultural politics at
play in representations of Venus and Serena Williams’, Sociology of Sport Online, 5(2), pp. 1–16.

65
K. Campbell-Kibler & d. miles-hercules

Eckert, P. (2012) ‘Three waves of variation study: the emergence of meaning in the study of variation’,
Annual Review of Anthropology, 41, pp. 87–100.
Eckert, P. (2014) ‘The problem with binaries: coding for gender and sexuality’, Language and
Linguistics Compass, 8(11), pp. 529–535.
Gaudio, R. (1994) ‘Sounding gay: pitch properties of gay and straight men’, American Speech, 69,
pp. 30–57.
Giles, H., Smith, P., Browne, C., Whiteman, S., and Williams, J. (1980) ‘Women’s speech: the voice
of feminism’, in McConnell-Ginet, S., Borker, R., and Furman, N. (eds) Women and language in
literature and society, vol. 10. New York: Praeger, pp. 150–156.
Gratton, C. (2016) ‘Resisting the gender binary: the use of (ING) in the construction of non-binary
transgender identities’, University of Pennsylvania Working Papers in Linguistics, 22(2), p. 7.
Gray, J. (2016) ‘Language and non-normative sexual identities’, in Preece, S. (ed) The Routledge
handbook of language and identity. Abingdon: Routledge, pp. 225–240.
Hall, K. and O’Donovan, V. (1996) ‘Shifting gender positions among Hindi-speaking hijras’, In
Bergvall, V. L., Bing, J. M., and Freed, A. F. (eds) Rethinking language and gender research:
theory practice. London: Longman Press, pp. 228–266.
Hardeman, G. K. (2013) Gender and second language style: American learner perceptions and use of
Mandarin sajiao. PhD dissertation, University of Hawai’i at Mānoa.
Hardeman, G. K. (2017) ‘Gender and second language style’, Asia-Pacific Language Variation, 2(2),
pp. 157–187.
Herdt, G. (ed) (1994) Third sex, third gender: beyond sexual dimorphism in culture and history.
Cambridge, MA: Zone Books.
Jacobs, G. (1996) ‘Lesbian and gay male language use: a critical review of the literature’, American
speech, 71(1), pp. 49–71.
Jacobs, G., Smyth, R., and Rogers, H. (2000) ‘Language and sexuality: searching for the phonetic
correlates of gay- and straight-sounding male voices’, Toronto Working Papers in Linguistics, 18,
pp. 46–61.
Jacobs-Huey, L. (2006) From the kitchen to the parlor: language and becoming in African American
women’s hair care. Oxford: Oxford University Press.
Kirtley, M. J. (2011) Speech in the US military: a sociophonetic perception approach to identity and
meaning. Master’s thesis, University of Hawai’i at Mānoa.
Kramarae, C. (1982) ‘Gender: how she speaks’, in Ryan, E. B. and Giles, H. (eds) Attitudes towards
language variation: social and applied contexts volume 1 of the social psychology of language.
London: Edward Arnold, pp. 84–98.
Kramer, C. (1977) ‘Perceptions of female and male speech’, Language and Speech, 20(2), pp. 151–161.
Kramer, C. (1978) ‘Women’s and men’s ratings of their own and ideal speech’, Communication
Quarterly, 26(2), pp. 2–11.
Kulick, D. (2000) ‘Gay and lesbian language’, Annual Review of Anthropology, 29, pp. 243–285.
Lakoff, R. (1973) ‘Language and woman’s place’, Language in Society, 2(1), pp. 45–79.
Lanehart, S. (ed) (2009a) African American women’s language: discourse, education, and identity.
Cambridge: Cambridge Scholars Press.
Lanehart, S. (2009b) ‘Diversity and intersectionality’, in Proceedings of the seventeenth annual
symposium about language and society, Austin. Austin: University of Texas, pp. 1–7.
Lerman, J. W. and Damsté, P. H. (1969) ‘Voice pitch of homosexuals’, Folia Phoniatrica et
Logopaedica, 21(5), pp. 340–346.
Levon, E. (2007) ‘Sexuality in context: variation and the sociolinguistic perception of identity’,
Language in Society, 36(4), pp. 533–554.
Levon, E. (2015) ‘Integrating intersectionality in language, gender, and sexuality research’, Language
and Linguistics Compass, 9(7), pp. 295–308.
Linville, S. E. (1998) ‘Acoustic correlates of perceived versus actual sexual orientation in men’s
speech’, Folia Phoniatrica et Logopaedica, 50(1), pp. 35–48.

66
 Perception of gender and sexuality

Mack, S. (2010a) ‘Perception and identity: stereotypes of speech and sexual orientation in Puerto
Rican Spanish’, in Selected proceedings of the 12th hispanic linguistics symposium. Somerville,
MA: Cascadilla Proceedings Project, pp. 136–147.
Mack, S. (2010b) ‘A sociophonetic analysis of perception of sexual orientation in Puerto Rican
Spanish’, Laboratory phonology, 1(1), pp. 41–63.
Mack, S. (2015) ‘Phonetic variation and perception of sexual orientation in Caribbean Spanish’,
in Levon, E. and Mendes, R. B. (eds) Language, sexuality, and power: studies in intersectional
sociolinguistics, vol. 130. Oxford: Oxford University Press.
Mack, S. and Munson, B. (2012) ‘The influence of /s/ quality on ratings of men’s sexual orientation:
explicit and implicit measures of the ‘gay lisp’ stereotype’, Journal of Phonetics, 40(1), pp. 198–212.
Mendes, R. B. (2014) ‘Gendered perceptions of noun agreement in Brazilian Portuguese’, Revista
Internacional de Lingüística Iberoamericana, 12(1), pp. 93–108.
Mendes, R. B. (2015a) ‘Diminutives and masculinity in Brazilian portuguese’, in Milani, T. (ed)
Language and masculinities: performances, intersections, dislocations, vol. 7. Abingdon:
Routledge, p. 117.
Mendes, R. B. (2015b) ‘Nonstandard plural noun phrase agreement as an index of masculinity’, in
Levon, E. and Mendes, R. B. (eds) Language, sexuality, and power: studies in intersectional
sociolinguistics. Oxford: Oxford University Press, pp. 105–130.
Mendes, R. B. (2016) ‘Diphthongized (en) and the indexation of femininity and paulistanity’,
Cadernos de Estudos Linguísticos, 58(3), pp. 425–444.
Moonwomon-Baird, B. (1997) ‘Toward a study of lesbian speech’, in Hall, K. and Livia, A. (eds)
Queerly phrased. Oxford: Oxford University Press, pp. 202–213.
Morgan, M. (1996) ‘Conversational signifying: grammar and indirectness among African American
women’, in Ochs, E., Schegloff, E., and Thompson, S. (eds) Interaction and grammar. Cambridge:
Cambridge University Press, pp. 405–434.
Munson, B. (2007) ‘The acoustic correlates of perceived masculinity, perceived femininity, and
perceived sexual orientation’, Language and Speech, 50(1), pp. 125–142.
Munson, B. and Babel, M. (2007) ‘Loose lips and silver tongues, or, projecting sexual orientation
through speech’, Language and Linguistics Compass, 1(5), pp. 416–449.
Munson, B., McDonald, E. C., DeBoe, N. L., and White, A. R. (2005) ‘The acoustic and perceptual
bases of judgments of women and men’s sexual orientation from read speech’, Journal of Phonetics,
34, pp. 202–240.
Nagar, I. (2008) Language, gender and identity: the case of kotis in Lucknow-India. PhD dissertation,
The Ohio State University.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context: language
as an interactive phenomenon. Cambridge: Cambridge University Press, pp. 335–358.
Pharao, N. and Maegaard, M. (2017) ‘On the influence of coronal sibilants and stops on the perception
of social meanings in Copenhagen Danish’, Linguistics, 55(5), pp. 1141–1167.
Pharao, N., Maegaard, M., Møller, J. S., and Kristiansen, T. (2014) ‘Indexical meanings of [s+]
among Copenhagen youth: social perception of a phonetic variant in different prosodic contexts’,
Language in Society, 43(1), pp. 1–31.
Podesva, R. J. (2007) ‘Phonation type as a stylistic variable: the use of falsetto in constructing a
persona’, Journal of Sociolinguistics, 11(4), pp. 478–504.
Podesva, R. J. and Kajino, S. (2014) ‘Sociophonetics, gender, and sexuality’, in Ehrlich, S., Meyerhoff,
M., and Holmes, J. (eds) The handbook of language, gender and sexuality. New Jersey: Wiley
Online Library, pp. 103–122.
Rácz, P. and Papp, V. (2015) ‘Percepts of hungarian pitch-shifted male speech’, in Levon, E. and
Mendes, R. B. (eds) Language, sexuality, and power: studies in intersectional sociolinguistics, vol.
151. Oxford: Oxford University Press.
Remedios, J. D., Chasteen, A. L., Rule, N. O., and Plaks, J. E. (2011) ‘Impressions at the intersection
of ambiguous and obvious social categories: does gay+ black= likable?’, Journal of Experimental
Social Psychology, 47(6), pp. 1312–1315.

67
K. Campbell-Kibler & d. miles-hercules

Rieger, G., Linsenmeier, J. A., Gygax, L., Garcia, S., and Bailey, J. M. (2010) ‘Dissecting ‘gaydar’:
accuracy and the role of masculinity-femininity’, Archives of Sexual Behavior, 39(1), pp. 124–140.
Schonlau, M. (2002) ‘The clustergram: a graph for visualizing hierarchical and non-hierarchical
cluster analyses’, STATA Journal, 3, pp. 316–327.
Sharma, D. (2018) ‘Style dominance: attention, audience, and the “real me”’, Language in Society,
47(1), pp. 1–31.
Silverstein, M. (1976) ‘Shifters, linguistic categories, and cultural description’, in Basso, K. and
Selby, H. (eds) Meaning in anthropology. Albuquerque, NM: University of New Mexico Press,
pp. 11–55.
Silverstein, M. (2003) ‘Indexical order and the dialectics of sociolinguistic life’, Language and
Communication, 23, pp. 193–229.
Smyth, R., Jacobs, G., and Rogers, H. (2003) ‘Male voices and perceived sexual orientation: an
experimental and theoretical approach’, Language in Society, 32, pp. 329–350.
Spillers, H. (1987) ‘Mama’s baby, papa’s maybe: an american grammar book’, Diacritics, 12(2), pp.
64–81.
Stanback, M. H. (1985) ‘Language and black woman’s place: evidence from the black middle class’,
in Treichler, P. A., Kramarae, C., and Stafford, B. (eds) For alma mater: theory and practice in
feminist scholarship. Chicago: University of Chicago, pp. 177–193.
Steele, A. (2019) Sociophonetic correlates of nonbinary gender speech styles. First Qualifying Paper,
Ohio State University Department of Linguistics.
Terango, L. (1966) ‘Pitch and duration characteristics of the oral reading of males on a masculinity-
femininity dimension’, Journal of Speech and Hearing Research, 9, pp 590–595.
Travis, N. J. (1981) A study of the relationship of certain variables to sex characteristic identification
from the speech of heterosexual and homosexual individuals. PhD dissertation, Louisiana State
University.
Walker, A., García, C., Cortés, Y., and Campbell-Kibler, K. (2014) ‘Comparing social meanings across
listener and speaker groups: the indexical field of Spanish /s/’, Language Variation and Change,
26, pp. 169–189.
Wanjema, S., Carmichael, K., Walker, A., and Campbell-Kibler, K. (2013) ‘The ohiospeaks project:
engaging undergraduates in sociolinguistic research’, American Speech, 88(2), 223–235.
Wolfe, V. I., Ratusnik, D. L., Smith, F. H., and Northrop, G. (1990) ‘Intonation and fundamental
frequency in male-to-female transsexuals’, Journal of Speech Hear Disord, 55(1), pp. 43–50.
Zimman, L. (2013) ‘Hegemonic masculinity and the variability of gay-sounding speech: the perceived
sexuality of transgender men’, Journal of Language and Sexuality, 2(1), pp. 1–39.
Zimman, L. (2017) ‘Gender as stylistic bricolage: transmasculine voices and the relationship between
fundamental frequency and /s/’, Language in Society, 46(3), pp. 339–370.
Zimman, L., Davis, J., and Raclaw, J. (eds) (2014) Queer excursions: retheorizing binaries in
language, gender, and sexuality. Oxford: Oxford University Press.
Zwicky, A. (1997) ‘Two lavender issues for linguists’, in Hall, K. and Livia, A. (eds) Queerly phrased.
Oxford: Oxford University Press, pp. 21–32.

68
 5
Gender diversity and the voice
Lal Zimman

Introduction
Recent years have seen unprecedented visibility for transgender people and issues, and
scholars of language are increasingly interested in exploring the ways gender diversity is
manifested on a linguistic level. Attention to the linguistic consequences of gender diver-
sity demands not only the inclusion of transgender and gender non-conforming speakers in
sociolinguistic research, but also a shift in how the field thinks about gender and sex – espe-
cially when it comes to the voice.
‘Gender diversity’ is a concept that recognises a range of gendered identities, expres-
sions, and bodies, including those that diverge from, transcend, or exceed normative
ideas about what it means to be a woman or man, female or male, feminine or masculine.
Transgender individuals – i.e. those who self-identify with a gender other than the one
assigned to them at birth – are prototypical examples of gender diversity. The category
of ‘transgender’, as it is currently conceived in the United States and many other parts
of the English-speaking world, includes transgender women and men and, in many cases,
non-binary individuals who do not identify as exclusively female or male.1 Recognition of
non-binary identities has proliferated in recent years, resulting in labels like ‘agender’ (no
gender), ‘bigender’ (both binary genders), ‘genderfluid’ (movement between gender identi-
ties or expressions), ‘genderqueer’ (having a non-normative or distinctively queer gender),
and ‘demi’-[category] (belonging only partially to a category, e.g. ‘demi-girl’), among oth-
ers. In addition to Western frameworks of trans identity, gender diversity can include gender
non-normativity and ‘third/fourth gender’ categories throughout the world, e.g. in India
(Hall 2005), Brazil (Borba 2015), Thailand (Jackson 2004), and indigenous North America
(Davis 2014). Finally, gender diversity is broad enough to encompass variation in the femi-
ninities and masculinities of ‘cisgender’ (i.e. non-transgender) women and men. Sensitivity
to gender diversity thus works in concert with an intersectional perspective on gender as
inseparable from race, ethnicity, class, age, sexuality, (dis)ability, and other forms of social
experience (Crenshaw 1989).
Variationist sociolinguistic research typically rests on a binary model of gender, in
which all speakers are categorised as either female or male, often based on the researcher’s

69
Lal Zimman

perceptions of speakers’ bodies rather than participants’ self-description. This binary cat-
egorisation is then implemented in both acoustic and statistical analysis, limiting the pos-
sibility of discovering a more complex gender system at work. If sociolinguists hope to
learn from the gender diversity that surrounds us, a new approach to gender and the voice
is necessary. This chapter begins by outlining a multi-dimensional model of sex and gender
that is grounded in ethnographic research in US-based transgender communities, which
enables more nuanced accounts of the relationship between gender and the voice. Rather
than imposing a simplified system that uses demographic categories defined by linguists, an
ethnographic approach encourages researchers to prioritise the worldviews of the communi-
ties under study, to learn about those worldviews through long-term participant-observation
in community members’ lives, and to be cautious about the potential for misreadings that
result from the researchers’ own subjectivity and cultural lens (Clifford and Marcus 1986;
Geertz 1973). This approach, I argue, is crucial for the study of gender diversity, precisely
because non-normatively gendered speakers reveal the inadequacies of mainstream ways of
thinking about gender.
Following a brief literature review, this multidimensional approach is put to work in an
analysis of /s/ among 32 English-speaking transgender people, including trans women, trans
men, and non-binary individuals. By delving into the complexity of speakers’ gendered sub-
jectivities, the analysis presented here undermines several normative assumptions about the
relationship between gender and the voice that circulate in variationist sociolinguistics: that
sex/gender is a single, binary characteristic; that speakers’ sex/gender can be conflated and
categorised based on researchers’ perceptions of participants’ bodies or identities; and that
all voices can be unproblematically categorised as either female or male on the basis of ‘bio-
logical sex’. Crucially, these implications are not only relevant for the study of transgender,
non-conforming, or third gender communities. A more complex model of gender also pro-
motes deeper understanding of gender and the voice among normatively gendered speakers.
(Cisgender) women and men are often treated as internally homogenous categories, but our
knowledge that gender intersects meaningfully with other domains of identity should make
us sceptical of claims about ‘sex differences’ in the voice that rely largely or exclusively
on white, middle-class, (presumed) heterosexual, gender normative, able-bodied speak-
ers. The analysis presented in this chapter, however, demonstrates that even before turning
to intersectional analysis, the gendered elements of our accounts must be complexified.
Specifically, it shows that the robust models of sex and gender developed in trans communi-
ties provide better explanations for variation in the gendered voice than does the normative
binary. As sociocultural linguists increasingly recognise the complexities of sex and gender,
our analytic tools must also be refined so they can account for the full range of gendered
voices that speakers produce.

Sex, gender, and sexuality as multidimensional scales

Sex
The distinction between ‘sex’, as a system of classifying bodies, and ‘gender’, as a sys-
tem for classifying the social practices, dispositions, and experiences that are associated
with one sex or another, is well established in sociolinguistics. Yet phonetically oriented
research, including much sociophonetic work, tends to conflate sex and gender and refer to
all aspects of difference between women’s and men’s voices as a function of sex. This ter-
minological choice is driven by the recognition that the size of the larynx, vocal folds, and

70
 Gender diversity and the voice

vocal tract constrains the kinds of speech a person can produce. The practice of emphasis-
ing ‘sex’, however, reinforces the naturalisation of differences between women’s and men’s
voices and constitutes a stance that the observed differences are biological in nature. Other
sociophoneticians are informed by the classic feminist distinction between sex (the biologi-
cal) and gender (the social), treating differences in women’s and men’s voices as shaped
largely by physiology, with supplementation from learned gender differences (e.g. Fuchs
and Toda 2010). However, this perspective misses a key contribution of third-wave femi-
nist/queer theory: that sex itself is a social construct (Butler 1993; Nicholson 1994). Butler’s
theory of gender famously draws on practices of gender crossing and blurring as an example
of ‘performativity’. Butler’s performativity mirrors Austin’s (1962) notion of performative
speech acts, in which utterances not only describe, but actually create, reality; for example,
the pronouncement of marriage made by a proper authority over a culturally appropriate
context brings a marriage into being that did not previously exist. Gender performativity,
by extension, refers to the idea that a person’s gender is not a pre-existing fact, but rather
something that must be continuously brought into being through the enactment of social
practices. As Butler (1990) famously opined, non-normative genders are particularly salient
as examples of gender performativity, but they illustrate a universal process of identity crea-
tion. Importantly, Butler does not stop at the notion of the long-recognised feminist precept
that gender identity is socially constructed; she also argues that the gendered meanings
attributed to the body are enacted through social practice – particularly discourse – and that
we can never fully separate the ‘natural’ body from the social and cultural contexts that pro-
duce the actual bodies which we animate and those we come into contact with. The notion
that sex, like gender, is constructed is well supported by scholars of language, gender, and
sexuality, who have examined the ways bodies and body parts are invested with gendered
meaning (Braun and Kitzinger 2001; Motschenbacher 2009; Zimman 2014). ‘Sex’ may be a
useful concept to highlight the embodied aspects of gender, but it is important to maintain a
critical perspective on both sex and gender and to recognise their situatedness within a par-
ticular historical and cultural context. If we take seriously Butler’s argument that the social
and the material are fully bound up with one another, the attempt to isolate purely biological
differences in the voice from those that are culturally learned may be a fool’s errand (see
Zimman 2014 for a thorough exploration of this stance).
Though we might separate sex and gender analytically, gender diversity includes vari-
ability in both bodily forms and the ways bodies act and interact in the world. The sex binary
is a cultural imposition that erases intersex bodies, whose chromosomal, hormonal, and/or
anatomical features are not contained by normative definitions of female or male embodi-
ment. With an inclusive definition, the prevalence of intersex bodies is placed as high as 1
in 100 births (ISNA 2008), but linguists have yet to incorporate the challenges this popula-
tion poses to the field’s understanding of sex (see King 2016 for more on intersex bodies
and discourse). What intersex bodies make clear is that sex is not a single characteristic, but
rather a bundle of traits that are often, but not always, arranged in predictable constellations.
As the discussion below will demonstrate, the voice itself operates in a similar manner: not
as a single attribute but as clusters of features that may or may not be combined in norma-
tive ways. In addition to showing variability within, as well as between, sex categories, the
characteristics that constitute sex – and, by extension, the sexed voice – can be altered in
any number of ways. This is true not only for trans and intersex people, but also for cis,
non-intersex bodies which are the product of normative body-changing processes (puberty,
pregnancy, hormonal birth control and hormone replacement therapy, menopause and age-
ing, exercise, diet, cosmetic surgery, etc.) and which show variation across populations as

71
Lal Zimman

well as across sexes (in height, breast size, muscle mass, hair, fat distribution, etc.). What
remains to be investigated is how this form of intra-gender variation plays out in the voice.
With none of these elements of sex forming a simple binary, and many of them fluid to
some degree or another, a classification system that recognises only two types of bodies
loses the gradations that might help us understand how sex shapes the gendered voice.

Gender
Part of how sex and gender are naturalised is through the appearance of total continuity
between an individual’s body, identity, style of presentation, and how they are perceived by
others. However, the study of non-normative genders requires a distinction between differ-
ent elements of gender and sex that goes beyond what is typically found in sociolinguistic
research. In addition to the complexity of bodily sex, people also have an ‘assigned sex/
gender’, which represents the category they were placed in at (or before) birth.2 People often
refer to trans people’s gender assignment as their ‘biological sex’; however, such usage
naturalises the process of binary gender assignment, implies that sex is immutable and fixed
at birth, and ignores cases in which a person is assigned to a binary gender despite having
an intersex body. Gender assignment is designed to determine ‘gender role’, which refers to
the culturally and institutionally recognised social roles a person can occupy. Gender role
is typically what changes for trans people in their transition, regardless of whether medical
technology is employed. Even non-binary individuals may shift from one binary gender
role to another without necessarily identifying with either binary gender. Gender role is dif-
ficult to define exhaustively, but it can involve self-presentation (e.g. the choice of a name),
institutional recognition (e.g. identification documents), the use of gendered public spaces
(e.g. restrooms), the language used by others (e.g. third-person pronouns), and other ways
people are sorted into one gender category or another. Gender role, then, is based in part
on the recognition of others, and it is for this reason that the primary struggle non-binary
people encounter with respect to this aspect of gender is whether more than two roles can
or should be recognised/able. ‘Gender identity’, on the other hand, refers to the category/ies
with which a person self-identifies; as female, male, trans, non-binary, and so on. Gender
assignment, role, and identity can be combined in any number of ways – for instance, a
person might have been assigned female at birth, identify as non-binary, and live in a male
gender role in cases where their non-binary identity is not recognised. A final dimension of
gender that must be distinguished from the others identified here is ‘gender presentation’ or
‘expression’. Gender expression captures the semiotic resources an individual deploys such
as clothing, hairstyle, and makeup choices, gendered linguistic practices, and bodily hexis
(gesture, posture, gait, etc.). Gender presentation is often taken as an index of gender iden-
tity or sexual orientation, but clearly not all (straight) women express normative femininity
and not all (straight) men are normatively masculine, whether they are cis or trans. Nor can
non-binary people be anticipated to have any particular gender presentation. These distinc-
tions between different elements of gender and sex take on critical importance when gender
diversity is in focus, but they also provide greater explanatory power in our theorisation of
how and why normative phonetic gender differences arise. For these reasons, the commu-
nity-specific model I introduce based on ethnographic fieldwork in trans communities may
prove to be useful in the study of cisgender voices as well.
This multidimensional model of sex and gender is sketched out in Table 5.1, which
summarises each of the elements I have discussed. It is worth noting that other aspects of
identity can be subjected to similar divisions – for instance, a multidimensional model of

72
 Gender diversity and the voice

Table 5.1 Dimensions of sex and gender

Gender Assignment Categorisation (generally binary) made at birth based on

recognised social categories.
Role Socially and/or institutionally recognised gender categories
within a given culture.
Identity Self-identification as female, male, non-binary, trans, etc.
Presentation/expression The collection of semiotic resources a person displays that
together index gender.
Sex Chromosomes Categorical but non-binary combinations of X and Y
chromosomes (XX, XY, XXY, XYY, XØ, etc.).
Hormones The collective levels of an array of hormones including (but not
limited to) testosterone, oestrogen, and progesterone.
Anatomy Internal and external anatomy, including reproductive organs
and genitals.
Other? Other/more fine grained embodied gender difference (e.g. in
the larynx).

sexuality might distinguish erotic desire, romantic desire, self-identified labels, and com-
munity membership.
By teasing apart different elements of sex and gender, a model such as this one demands
more precision from analysts, which pushes us to make explicit our assumptions and
hypotheses about how sex and gender, along with other aspects of identity, shape the voice.

Research on gender and the voice

The gendered voice – like the sex of the body – is often treated as a single dimension despite
the fact that it is composed of a number of distinguishable features. Most phonetic research
on gender and the voice assumes that differences between women’s and men’s voices ema-
nate directly from the anatomical consequences of membership in one of two sex categories
(see Zimman 2018 for a more in-depth review). Scholars of language, gender, and sexuality,
however, have long pushed back against the idea of gender as an innate, natural, or biologi-
cal phenomenon and advocated a constructivist understanding of the gendered voice. This
began in early research in the field, which called attention to androcentric assumptions and
the importance of childhood socialisation in developing a normatively gendered voice (e.g.
Henton 1989; McConnell-Ginet 1983; Sachs 1975). More recently, studies of gay, lesbian,
bisexual, and queer voices have made clear that not all men embody norms for ‘male voices’
and not all women exhibit the features attributed to ‘female speakers’ (e.g. Munson 2007;
Podesva 2007). A constructivist model treats phonetic gender differences as culture-specific,
non-binary, flexible, and intersectional. Importantly, constructivist approaches to gender do
not maintain that the body is irrelevant. Instead, ‘biology’ is seen as always filtered through
some sort of cultural lens and never completely separable from social practices.
A good example is fundamental frequency (or F0), which represents the speed of vocal
fold vibration and roughly corresponds to the concept of pitch. As the most salient and
intuitive gender difference in the voice, pitch tends to differ substantially between cisgender
women and men, with much of this difference apparently arising in puberty (Hollien et al.
1994). Changing testosterone levels drive this differentiation in normative populations; the
hormone also has a marked effect on trans speakers who use it for physical masculinisation
(Azul et al. 2016 for a review; Zimman 2017a). Among speakers of American English,

73
Lal Zimman

average speaking fundamental frequency is generally placed at around 100–120 Hz for

men and 200–220 Hz for women (e.g. Simpson 2009). However, it is clear that sociocul-
tural factors also exert a significant influence on the way women and men use pitch. The
size of the gap between women’s and men’s average pitch differs across languages, with
Japanese-speaking women famously maintaining a higher pitch than their English-speaking
counterparts even as Japanese-speaking men use lower pitch than English-speaking men
(e.g. Loveday 1981; Yuasa 2008). Furthermore, these differences tend to arise in childhood,
before sex differences in the vocal anatomy appear (Ferrand and Bloom 1995; Hasek et al.
1980). While physiology constrains the types of voices speakers can produce, they have
access to a far larger range of voice types than they typically employ, at least when voicing
themselves. This research demonstrates that a purely biological account of gender differ-
ences in the voice is untenable.
While sex differences in the vocal anatomy clearly impact pitch to some degree or
another, other gender differences in the voice seem to be driven more strictly by articula-
tory practice. /s/ is the most widely-studied segment in the sociophonetics of gender and
sexuality in both a variety of English dialects (Bekker and Levon 2017; Campbell-Kibler
2011; Heffernan 2004; Holmes-Elliott and Levon 2017; Podesva and van Hofwegen 2016;
Stuart-Smith 2007; Zimman 2017b) and other languages (e.g. Bekker and Levon 2017 on
Afrikaans; Chappell 2016, Pharao et al. 2014 on Danish; Walker et al. 2014; and Zimman
2017b on Spanish). Acoustic measures for /s/ generally identify the most prominent fre-
quencies, often in the form of a weighted mean frequency, with higher values associated
with femininity. Differences in the frequencies present in /s/ depend in large part on where
the tongue is placed within the mouth (Fuchs and Toda 2010). When the tongue is placed
further forward in the mouth, it typically results in a higher frequency turbulent sound that
might be described by non-linguists as ‘lispy’. Several studies have suggested or argued that
gender differences in /s/ are at least partly physiological (Flipsen et al. 1999, Fuchs and Toda
2010), but Zimman’s (2017b) analysis of trans men with a particularly wide range of /s/ pro-
duction challenges the viability of this supposition by showing that members of the ‘same
sex’ can produce /s/ in ways that cover the full range reported for either women or men.
Studies of /s/ among presumed cisgender speakers also offer substantial evidence that
gender differences in this sound have little, if anything, to do with anatomical difference.
Heffernan (2004) shows that gender differences in /s/ are more robust among speakers of
Canadian English than speakers of Japanese, which is notably a reversal of the pattern for
pitch. Fuchs and Toda (2010), though focused on gender, found larger differences between
Britons and Germans than between women and men of either group. Flipsen and colleagues
(1999) report evidence of gender differences in /s/ among prepubescent children, and Stuart-
Smith’s (2007) analysis of /s/ in Glasgow shows that class predicted whether preteen girls
produced the sibilant in ways more similar to adult women or adult men (see also Holmes-
Elliott and Levon 2017). Other studies have identified connections to dialect or geogra-
phy (Campbell-Kibler 2011; Podesva and van Hofwegen 2016; Stuart-Smith 2007), race
or racialised citizenship status (Bekker and Levon 2017; Pharao et al. 2014), and sexuality
(Munson 2007; Zimman 2013). When differences between /s/ among ‘women’ and ‘men’
are discussed, however, it is usually under the unspoken assumption that these categories
represent socially unmarked – i.e. white, middle-class, straight, cisgender, gender-norma-
tive – women and men.
Disputes over the origins of gender differences in /s/ provide an opportunity to illus-
trate how trans speakers contribute something to the study of the voice that cisgender
speakers alone cannot. For this reason, /s/ has attracted the attention of several researchers

74
 Gender diversity and the voice

investigating transgender speech. Zimman (2015, 2017b) addresses the variability in the
production of /s/ among trans masculine people in the San Francisco Bay Area and argues
for a finer-grained theory of gender, including some – but not all – of the aspects of gender
presented in Table 5.1. Podesva and van Hofwegen’s (2016) analysis of /s/ in rural northern
California places both trans women and trans men in relation to cis speakers to illustrate a
continuum of gender normativity. In contrast to Zimman’s focus on variability and queer-
ness among trans masculine speakers, Podesva and van Hofwegen find that trans speakers
produced /s/ in ways that were mostly similar to cis people of the same gender (i.e. trans
women were similar to cis women and trans men to cis men), but ultimately both groups fell
in-between the two cisgender identities. Hazenberg’s (2016) analysis of /s/ among trans peo-
ple in Canada similarly places binary-identified trans people in-between cisgender women
and men. Together, these studies demonstrate that gender identity is not the only social
factor driving gender-based variation in /s/ and that a finer-grained model of gender may
be needed to understand why groups of trans women and trans men would fall somewhere
between cisgender women and men without resorting to biological determinism or invali-
dating trans identities.
The analysis presented below builds on previous studies in a few ways. First, it includes a
wider range of trans identities, including trans women, trans men, and non-binary individu-
als (both female-assigned and male-assigned). Second, while previous studies that included
both women and men have generally focused on inter-group differences, the present analy-
sis also looks at differences within categories in order to understand what might drive vari-
ation on an intra-gender basis. Finally, while Zimman (2015) sketched out different aspects
of gender that are relevant for trans speakers, this chapter adds to that model and tests how
a multidimensional model of gender might be used in a quantitative analysis.

From theory to analysis

As an illustration of how a multidimensional model of gender might be employed to exam-
ine a diverse set of gender identities, the remainder of this chapter focuses on an analysis of
/s/ in a group of English-speaking transgender people recorded between 2010 and 2014 in
and around San Francisco, CA and Portland, OR.

Methods
This analysis builds on previous work presented by Zimman (2015, 2017b), in which /s/ was
compared among 15 trans masculine speakers, including transgender men and non-binary
individuals who were assigned female at birth (AFAB). These speakers were recorded as
part of a two-year ethnographic study in 2010–2012 in the San Francisco Bay Area focused
on changes in the voices of trans masculine people starting testosterone therapy. These data
are supplemented by the addition of 17 trans feminine speakers, including trans women and
non-binary individuals who were assigned male at birth. These speakers were interviewed
in Portland, Oregon in 2014.
This study employs a range of methods, including ethnographic fieldwork, sociolinguis-
tic interviews, and acoustic analysis of read speech. Triangulating these approaches provides
a range of benefits. Ethnography involves the establishment of long-term relationships with
research subjects through participant-observation in their everyday activities. Ethnography
is particularly useful for providing a culturally grounded account of quantitative analysis.
It can also shed light on individual differences within a population, which will become

75
Lal Zimman

particularly important in the discussion below. At the same time, the analysis of controlled
speech genres, like read passages, can provide another set of insights. First, read speech
largely controls for the content of what is said, the kinds of stances speakers are able to take,
and the phonetic context of each token of a sound. This allows for a direct, equivalent com-
parison of /s/ produced by different speakers. Second, read speech is thought to elicit greater
linguistic self-monitoring than extemporaneous speech (e.g. Labov 1972). As a result, read-
ing a passage serves as an opportunity for a speaker to project a desirable sociolinguistic
image. This idea has historically been invoked in research on standardised and stigmatised
linguistic variants, but the same principle may shape the gendered styles speakers produce.
The comparison of gender differences across discourse contexts will be an important area of
future investigation. For now, the analysis of read speech offers a view of what speakers are
able and willing to do when it comes to enacting a desired gendered style.
The acoustic and statistical methods followed here are the same as those used in Zimman
(2017b), in which tokens of word-initial /s/ were extracted from productions of the Rainbow
Passage that had been filtered to remove frequencies below 1,000 Hz and above 13,000 Hz
(see also Stuart-Smith 2007). Centre of gravity (COG) – a weighted mean frequency – was
calculated as an average across the duration of each token of /s/ using Praat’s automated
moments analysis function. Linear mixed effects regression models were constructed to
examine the relationship between /s/, gender assignment, and gender identity. In these mod-
els, gender identity (F, M, or NB) and gender assignment (AFAB or AMAB) were treated as
fixed effects while speaker and word were random effects. ANOVAs (analysis of variance)
were also carried out using Tukey HSD (honestly significant difference) post-hoc analyses
to explore group-based differences. Further distinctions based on sexuality and gender role
experience are introduced in the discussion of results.

Speakers
This analysis presents results from 32 transgender speakers, 15 of whom were assigned
female at birth (AFAB) and 17 of whom were assigned male (AMAB). The speakers’
self-described gender identities, assignments, and sexualities are represented in Table 5.2.
Among the AMAB speakers, 12 identified as women and 5 with a non-binary category.
Among the AFAB speakers, nine identified as men and six as non-binary. Table 5.2 also
reports the self-identified sexual orientations of speakers. However, the number of speakers
per cell were too small and imbalanced to include sexuality in the statistical analysis of this
sample. Notably, some combinations of sexualities and genders are unlikely to pair together,
e.g. straight and non-binary.

Table 5.2 Number of speakers by gender assignment, identity, and sexuality

AMAB AFAB

Women Non-binary Men Non-Binary

Straight 1 0 3 0
Queer3 5 5 4 6
Both straight and queer 0 0 2 0
Lesbian 6 0 0 0
Total 12 5 9 6

76
 Gender diversity and the voice

In terms of other identities, most speakers were between the ages of 20 and 35, though a
few were in their 40s (three trans men), 50s (one trans woman), and 60s (one trans woman).
The non-binary speakers were all under the age of 33. Almost all of the participants described
themselves as white, with two identifying as Latinx4 (both AMAB non-binary individuals),
one as Native American (the straight trans woman), and one as Filipino (one of the straight/
queer trans men). All were native speakers of American English, save one AFAB non-binary
speaker who grew up in Spain and acquired a standardised Southern British English variety
from his mother. Self-reported socioeconomic class is varied in this sample, but its assess-
ment is complicated by the rampant employment discrimination trans people experience
(Grant et al. 2011). Based on the literature review above, identities other than gender and
sexuality are undoubtedly important for understanding these speakers. However, the analy-
sis presented below underscores the fact that the gendered elements of our accounts alone
must be complexified.

Results and discussion: interspeaker analysis

The results of the linear mixed effects regression analysis indicate that centre of gravity
for /s/ differs according to both gender identity and gender assignment in this sample. As
Table 5.3 shows, the intercept estimate of 8,459 Hz for the trans women was estimated as
being 1,827 Hz lower for the trans men (b = −1827, SE = 680, t = -2.69) and 199 Hz lower for
non-binary speakers when all of the non-binary speakers are lumped together without regard
to assigned sex (b = −199, SE = 474, t = −0.42). Speakers who were assigned male at birth
were also estimated as a whole to have a lower COG than the female-assigned speakers (b =
−1736, SE = 563, t = − 3.08); however, as the more detailed analysis to follow demonstrates,
this difference was driven largely by the non-binary speakers. Notably, given the intercept
of 8,459 Hz for the women and 6,632 Hz for the men, both AFAB and AMAB speakers
display a relatively high frequency /s/ relative to reference values reported for ‘women’
(6,400–8,500 Hz) and ‘men’ (4,000–7,000 Hz) in studies of American English using com-
parable methods (Zimman 2017b).
The group-based (ANOVA) comparisons indicate that the trans women, the trans men,
and the male-assigned non-binary speakers all had lower centres of gravity than the non-
binary individuals who were assigned female at birth (p < 0.001 in each case). There was
no statistically significant difference between the trans women and the trans men, the trans
women and the AMAB non-binary speakers, or the trans men and the AMAB non-binary
speakers. Tables 5.4 and 5.5 contain results from the ANOVA model, which uses a col-
lapsed variable for gender identity and assignment in order to facilitate group-by-group
comparisons.
Figure 5.1 shows the results for centre of gravity according to gender grouping. The
figures, like the ANOVAs, sort speakers into four categories based on the intersection of

Table 5.3 Results of linear regression Model #1

Estimate Standard error t value

Intercept 8459.3 629.6 13.436

Gender identity (male) −1826.7 680.1 −2.686
Gender identity (non-binary) −198.5 473.9 −0.419
Gender assignment (AMAB) −1735.8 563.2 −3.082

77
Lal Zimman

Table 5.4 Results of ANOVA Model #1

Degrees of freedom Sum of squares Mean squares F value Pr(>F)

Combined gender identity 3 155168252 51722751 44.68 p < 0.001 ***

and gender assignment
Residuals 447 517436749 1157577

Table 5.5 Results of Tukey HSD post-hoc testing on ANOVA Model #1

Group Versus Difference Lower Upper Adjusted p

NB (AFAB) Men 1628 1222 2034 0.000 ***

NB (AMAB) Men −108 −514 298 0.903
Women Men 85 −231 401 0.9
NB (AMAB) NB (AFAB) −1736 −2205 −1267 0.000 ***
Women NB (AFAB) −1543 −1937 −1150 0.000 ***
Women NB (AMAB) 193 −201 586 0.588

Figure 5.1 Mean centre of gravity for /s/ by gender grouping (by group). For more legible,
color plots, visit http://www.lalzimman.com/GenderDiversityPlots.

gender assignment and identity: trans male, AFAB non-binary, AMAB non-binary, and trans
female. Most of the findings are well represented in this image: the AFAB non-binary speak-
ers tend to be higher than the trans men and the AMAB speakers tend to be lower than the
AFAB speakers. The AMAB non-binary speakers also have a narrower distribution than the
trans women speakers, perhaps due to the smaller number of speakers in the former group.

78
 Gender diversity and the voice

These results indicate that both gender assignment and gender identity matter. When it
comes to gender assignment, it seems that the socialisation experiences trans people have
early in life may exert an influence on their sociolinguistic styles in adulthood. That said,
it is worth highlighting that this difference is driven in large part by the non-binary speak-
ers, given that there is no statistically significant difference between the trans women and
the trans men. The fact that the trans women and men fall into the same range and that the
trans men were significantly different from the non-binary AFAB speakers, demonstrates
that gender identity is important as well. Speakers do not mechanistically enact the gender
norms ascribed to them; there is some room for individuals to push back against or reject
the socialisation messages they receive. After all, trans people are defined as those who do
not identify with the gender that their socialisation experiences were, as a whole, designed
to produce. As Butler (1990) and other poststructuralist feminists have argued, agency and
resistance against the gender order can be observed in the endless iterations of gender that
one must perform, creating the possibility for those iterations to shift gender in new direc-
tions. In this sense, the potential for individual agency can be observed through the analysis
of the gendered practices in which gender diverse speakers engage.
Although the trans women speakers tend to have relatively high centres of gravity for
/s/, there is a cluster of tokens in the data from trans women that fall on the low end of
these data, far lower than any of the speakers from the other groups, including the AMAB
non-binary individuals. There is similarly a grouping of tokens produced by trans men
that are much higher than the majority of their productions, on a par with tokens from the
high end of the AFAB non-binary individuals. The distribution of data within the gen-
der categories – particularly among the binary-identified trans people – warrants further
investigation.

Results and discussion: intragroup analysis

The discussion so far has accounted for two aspects of the gender model delineated above
– gender assignment and gender identity – but we have not yet discussed gender expres-
sion, gender role, or the issue of sexual orientation. These factors become important for the
analysis of variation within each gender group. Since gender assignment and identity have
already been demonstrated to be important, speakers are split here according to assignment,
which best facilitates an intragroup comparison according to the other dimensions of gen-
der, each of which influences speakers’ relationships to the norms assigned to them at birth.
Figure 5.2 shows the centre of gravity for /s/ among the 15 AFAB speakers. Gender iden-
tity is indicated by grey for non-binary speakers and black for trans men. Sexual orientations
represented in this group are indicated by shape, with circles for participants who identified
as queer, squares for participants who identified as straight, and triangles for speakers who
described themselves as both straight and queer. Speakers are ordered from lowest to high-
est mean COG.
Both sexual orientation and gender identity are important for explaining the variation in
Figure 5.2. The three strictly straight-identified AFAB participants in this study also have
the three lowest mean centres of gravity. The fourth and fifth lowest COGs belong to the
two men who identified as both straight (by virtue of their attraction to women) and queer
(by virtue of their gender and history). Figure 5.2 also illustrates the intersection of gender
identity and sexuality, as all the straight speakers were – unsurprisingly – male-identified.
The three straight men, along with one of the straight/queer men, were the only AFAB
speakers to have means below the range typically reported for English speaking women (i.e.

79
Lal Zimman

Figure 5.2 Mean centre of gravity for /s/ (AFAB speakers). See http://www.lalzimman.com/
GenderDiversityPlots.

6,400–8,500 Hz), with the other straight/queer speaker following close behind with a mean
of 6,522 Hz. The non-binary speakers, by contrast, all had means at or above 8,000 Hz.
Gender expression also becomes key in explaining the one male-identified speaker with
a particularly high frequency /s/. This speaker, Dave, is a fem man; that is, while he identi-
fies as a man rather than non-binary, both self-description and ethnographic observation
make clear that his gender expression is not normatively masculine by the standards of his
surrounding communities. Over the year we spent together, I heard Dave refer to himself as
‘faggy’, ‘queeny’, and ‘flamboyant’, an image he conveys linguistically through his use of
a broad pitch range, plentiful falsetto and creak (Podesva 2007), and a generally expressive
and engaging interactional style. His fem presentation is also constructed through his style
of clothing (brightly coloured, form-fitting, body-revealing), adornment (make-up, jewel-
lery, fingernails), and, in the context of Dave’s participation in local BDSM/fetish commu-
nities, gear that is typically used or worn by women (corsets, high-heeled boots).
The trans women and male-assigned non-binary speakers, who are plotted in Figure 5.3,
provide another set of insights. Like the trans masculine participants in this study, the trans
feminine speakers are divided into two gender identities – female (black) and non-binary
(grey) – and three sexual orientations – in this case, queer (triangles), straight (squares), and
lesbian (circles).
While the trans masculine speakers showed a fairly clean division between binary and
non-binary participants, the pattern is more variable in the trans feminine case. The AMAB
non-binary speakers do tend to have lower centres of gravity than the trans women, but two
non-binary speakers are positioned in the higher ranges along with trans women. Similarly,
the two lowest centres of gravity come from trans women rather than non-binary individu-
als. Sexuality, too, is hard to read from this particular sample because there was only one
straight-identified trans woman in the study, though she did turn out to have among the

80
 Gender diversity and the voice

Figure 5.3 Mean centre of gravity for /s/ (AMAB speakers).

highest centres of gravity among the AMAB speakers. However, gender role turns out to be
salient as an explanatory factor for the only two trans women whose mean centre of gravity
falls below the normative female range of 6,400 to 8,100 Hz. There is a large gap in Figure
5.3 between these two speakers’ means of 5,176 Hz and 5,905 Hz, and those of the three
next lowest frequency trans women, all of whom have means of approximately 6,500 Hz.
However, these two speakers also differ from the others in a particularly significant way:
their relationship with their gender role.
The lowest COG in this study belongs to Erika, who identifies as a trans women but
who lives in a male social role and is only out as trans to a select group of intimates. In our
interview, Erika spoke at length about the barriers that kept her from transitioning, one of
which was her concern about whether she could ‘pass’ as a woman because of her large
body frame, shape, and other physical markers that might undermine her ability to be
read as a woman. She also specifically mentioned her voice as a potential obstacle. Even
though this speaker identifies as a woman, her day-to-day experiences – including the
dynamics in her relationships with loved ones – still pressure her to present and comport
herself in a masculine way. The interactional goal of being perceived as a woman is not
one Erika is attending to the way other trans women in this study are, shifting the kinds
of motivations that might shape the gendered styles she adopts. The fact that Erika sees
her voice as ‘too masculine’ to transition could even introduce an ironic motivation not to
feminise her voice, since her perception of her voice as masculine affirms the choice she
has made not to transition, thereby potentially mitigating anxiety or depression about such
a difficult – if not heartbreaking – choice. Although we cannot know with certainty what
motivates any individual, it is clear that the context of living as a man despite identifying
as a woman provides a different set of limitations and affordances than pursuing a gender
role transition.

81
Lal Zimman

The importance of gender role and its relationship to gender expression is further high-
lighted by Katherine, the other trans woman with a particularly low centre of gravity. Unlike
Erika, Katherine has transitioned. However, at 63 years old at the time of our interview,
Katherine is the oldest participant in this study by 10 years. Katherine also transitioned
fairly late in life, before which she had lived and presented herself as a fairly conventional
straight man. In this sense, her experience with her gender role was, until recently, similar
to Erika’s: both occupied masculine gender roles for the majority of their lives, both formed
ostensibly heterosexual relationships with women, and both appeared to be gender norma-
tive men from the outside. Erika and Katherine’s experience with their assigned gender role
differs from that of Ethan, a trans man in his late 40s who was the oldest female-assigned
participant in this sample, and Adam, who was the second oldest AFAB participant at 40.
Although Ethan and Adam did not begin their formal transitions until well into adulthood,
they had always had predominantly masculine gender presentations, formed ostensibly les-
bian relationships with women, and before their transitions had been perceived as visibly
queer, gender non-conforming females. This helps to explain why Ethan and Adam have the
lowest (most gender normative) /s/ among the AFAB speakers while Erika and Katherine
have the lowest (least gender normative) /s/ among the AMAB speakers.
Given that Erika and Katherine are the only trans women to have had these kinds of
experiences, and that equivalent experiences are not represented among the trans men in
this study, a second statistical run was performed in order to consider whether group-based
differences might be altered by the inclusion or exclusion of trans women who had spent so
long occupying a normative male gender role. This should not be taken as a replacement for
the results discussed above, but rather as another perspective on how these data might look
if the sample were better controlled for gender role experiences.
In the second statistical run, both gender identity and assignment still exert a large influ-
ence on centre of gravity. As we would expect, the effect size for gender identity is larger
both in the comparison between the trans women and the trans men (b = −1736 in Model
1, b = −2144 in Model 2) and between the trans women and the non-binary AMAB speak-
ers (b = −199 in Model 1, b = −516 in Model 2). The effect of gender assignment is largely
unchanged (Table 5.6).
More dramatically, the second ANOVA model indicates statistically significant differ-
ences between almost all of the gender groups considered here; the only pairing not to
show a significant difference was the trans men and the male-assigned non-binary speakers.
Tables 5.7 and 5.8 present the results of the second model and post-hoc testing in the same
format as Tables 5.4 and 5.5. The bolded rows indicate a change from the previous model.
The group-based comparison performed by the Tukey HSD tests indicate that the previ-
ously significant differences between the AFAB non-binary speakers and each of the other
three groups remain. Additionally, the trans women’s COGs are now significantly higher

Table 5.6 Results of linear regression Model #2

Estimate Standard error t value

Intercept 8776 569.9 13.399

Gender identity (male) −2144 610.7 −3.511
Gender identity (non-binary) −516 431.8 −1.194
Gender assignment (AMAB) −1736 498.6 −3.481

82
 Gender diversity and the voice

Table 5.7 Results of ANOVA Model #2

Degrees of Sum of squares Mean squares F value Pr(>F)

freedom

Combined gender identity 3 145940077 48646692 47.3 p < 0.001 ***

and gender assignment
Residuals 419 430944134 1028506

Table 5.8 Results of Tukey HSD post-hoc testing on ANOVA Model #2

Group Versus Difference Lower Upper Adjusted p

NB (AFAB) Men 1628 1222 2034 0.000 ***

NB (AMAB) Men −108 −514 298 0.903
Women Men 395 84 706 0.006 ***
NB (AMAB) NB (AFAB) −1736 −2205 −1267 0.000 ***
Women NB (AFAB) −1234 −1615 −852 0.000 ***
Women NB (AMAB) 502 121 884 0.004 ***

Bold text indicates statistically significant comparisons present in Model 2 but not Model 1.

than those of either the trans men (diff = 395, p < 0.006) or the male-assigned non-binary
individuals (diff = 502, p = 0.004), though the magnitudes of difference were relatively small
in both cases, at least relative to the distance between the AFAB non-binary speakers and
everyone else.
Of course, simply removing speakers from a sample on the basis of their difference is not
an ideal approach to understanding variation within the trans population. In this case, the
speakers removed from the second model represent a very real subset of the trans commu-
nity who adopt an outward appearance of gender normative cisgender identity, often even
while internally identifying differently. These are not marginal subjects who fail to embody
transness – for all we know, there are more trans people living in their assigned gender role
than publicly transitioning. Instead, what these speakers make clear is that gender role and
age of transition are important variables that should be accounted for when studying and
modelling variation within trans communities. Their inclusion or exclusion has significant
effects on the results of a study like this one. An important goal for future research will be
to include large enough numbers of speakers to fully operationalise gender identity, assign-
ment, presentation, and role. For now, a combination of quantitative and qualitative analysis
demonstrates how each of these elements contributes to an account of variation in /s/ in a
sample of transgender women, men, and non-binary people.
To summarise, this analysis has demonstrated that both gender identity and gender
expression are significant predictors of how this sample of trans speakers produce /s/. The
male-assigned speakers tended to have lower centres of gravity for /s/ than the female-
assigned speakers, suggesting a socialisation effect. However, there were also differences
between the binary and non-binary speakers, indicating that gender identity differentiates
speakers of the same assigned gender. Gender role was necessary to explain outliers among
the AMAB speakers – two trans women who had spent most or all of their adult lives living

83
Lal Zimman

in a normatively masculine role. Once gender role experiences were balanced by the exclu-
sion of those two speakers, there were significant differences between almost every group
considered here, with the sole exception of the comparison between trans men and male-
assigned non-binary speakers. Notably, though, all of these speaker groups had relatively
high centres of gravity for /s/ compared to reference values for straight, gender-normative
cisgender men.
The use of both qualitative and quantitative methods has been crucial to address these
questions about gender diversity and the voice. Quantitative acoustic measures are impor-
tant to support claims about phonetic differences across groups of speakers. Acoustic meas-
urements and statistical analysis provide solid grounding for socially informed observations
about the voice, which is particularly important given that our auditory perceptions are
shaped by our social expectations (e.g. Strand 1999). At the same time, the accumulation of
ethnographic knowledge has been necessary for the construction of the models of sex and
gender taken up in this analysis, and to account for exceptional cases. A purely quantitative
approach would have missed the individual differences that led to a revised statistical model
and ultimately a different set of results in which gender role was taken into account. The
use of mixed methods may therefore be of particular usefulness for the study of voices at
the margins of social categories, which can cause the breakdown of traditional models for
identities like sex, gender, and sexuality.

Conclusion
This chapter has laid out a theory of sex and gender that complicates widely used methods
and commonly held theoretical assumptions in sociophonetics by highlighting the multi-
dimensionality of these aspects of the self. By distinguishing multiple elements of gender in
this analysis – including the gender assigned to speakers at birth, the gender(s) with which
speakers self-identify, the social role(s) in which speakers have or currently live(d), and
the way speakers present their gender semiotically – it is possible to consider patterns in
both inter- and intra-gender variation. In other words, this theory enables us to see broad
gender category-based differences, but it also helps explain why not all members of those
categories follow the overall trend. As we, as sociolinguists, increasingly include transgen-
der speakers in our research, it is important to articulate more explicit, and more explicitly
trans-inclusive, theories of how sex and gender operate on – and are produced through – the
voice. To illustrate the usefulness of teasing apart different elements of gender, the analysis
presented here focused on /s/ in a group of 32 transgender speakers with a varied set of
gender identities, presentations, roles, and assignments. While previous analyses have dis-
tinguished gender identity and presentation, a wider variety of speakers included here illus-
trates further delineation based on gender assignment and gender role. Attending to gender
diversity is particularly useful for investigating questions of how and why gender might
influence a particular aspect of the voice. In this case, the results suggest that each element
of gender – assignment, identity, presentation, and role – has a role to play in explaining the
quite variable productions of /s/ among the trans speakers considered here.
When gender diversity is considered, even the categories of cisgender woman and man
become complex, since gender presentation can vary considerably within those populations,
as can the way people understand and relate to identities like ‘women’ or ‘men’. Even as our
field opens up to include trans people of various sorts, it is important to find a place for gen-
der non-conforming cisgender speakers as well. For instance, this model can help us achieve
a better understanding of why some cisgender voices are perceived as ‘gay sounding’, and

84
 Gender diversity and the voice

why this perception does not perfectly align with speakers’ self-identification as gay (as
Zimman 2013 explores). Given the way categories are defined in part through what they are
not, a fuller picture of the boundaries and limits of gender categories is a crucial component
of our understanding of gender.
The challenges posed by trans speakers run deep and push the field to question our most
foundational assumptions about how and why sex/gender influences the voice. Trans speak-
ers problematise our discipline’s a priori division of voices into two categories on the basis
of perceived sex and demand that we specify what we mean when we describe a voice as
‘female’ or ‘male’. While the details of our models of gender may differ over time and across
(social) space, what non-normatively gendered speakers underscore is that the female/male
binary proves entirely too blunt an instrument for the study of gender in all its diverse forms.

Future directions
The study of gender diversity in the voice remains young, and there are a number of speaker
populations, theoretical questions, and methodological changes that remain to be identified.
One particularly promising theoretical issue is the role of individual agency over the gen-
dered voice. Given that phonetic gender differences tend to be attributed to physiological
sex, relatively little sociolinguistic research has considered the extent to which speakers can
consciously shift the gendered qualities of the voice. In addition, more attention is needed
to the way speakers construct their own agency and how those constructions might produce
different patterns of language usage.
For example, Zimman (2016) discusses the way trans men on testosterone may reject the
idea of intentionally masculinising their styles of speaking, instead depending on hormones
to transform their vocal physiology. While these speakers may have the cognitive and/or
vocal ability to change their speaking style, they see such practices as having problematic
implications for the authenticity of their gender and/or their political stance towards gender
non-normativity. Another, related, area for future investigation would focus on the degree
of variation in features like vocal pitch across different interactional contexts. With analysis
of read speech as a first step, it will be important to ask whether speakers maintain the same
kinds of usage in interviews, conversations, and other kinds of interactions. Importantly,
these questions must be approached cautiously because of their implications for trans popu-
lations and the ease with which trans identities have been delegitimated. The sensitivity of
these issues makes the use of ethnographic methods particularly useful, since that approach
necessarily involves a level of scepticism about and reflection on the researchers’ own
perspectives, assumptions, and positionalities. Other ethical safeguards that prioritise the
empowerment of the communities we study, such as identifying opportunities for collabora-
tion with non-academic community members, are also of critical importance.
Although the speakers included in this chapter represent a range of gender identities and
expressions, there are other ways in which they fail to represent the diversity of transgen-
der communities. Most notably, the majority participants included here were white young
adults from urban and suburban parts of the United States. The most important direction for
future work on transgender voices, then, is towards explorations of sex/gender in intersec-
tion with other dimensions of identity such as race/ethnicity, age, disability, class, politics,
and speakers’ native linguistic variety/ies. In my ongoing fieldwork in trans communities of
colour, the way trans people are gendered is often very overtly caught up in the ways they
are racialised; for instance, several African American trans men I have worked with describe
sometimes being perceived as female by other Black people even while they are universally

85
Lal Zimman

perceived as male by non–African Americans. This observation, which must be contextu-

alised by racial ideologies about Black hyper-masculinity, also highlights the paucity of
research on the production of gender difference in the voices of people of colour in general.
Another important area for future research is perception. As we learn more about the
diverse ways gender is produced, we need parallel investigations into the way gender is
perceived. In addition to considering how gender is perceived when speakers are trans or
gender non-conforming, we need to learn about how trans and non-conforming people
themselves perceive gender. Informal attempts to conduct gender perception experiments
with trans listeners has at times engendered refusals to participate on the grounds that a
person’s gender is not determinable based on the voice. Investigating the effects of listeners’
political orientation on the task of gender categorisation will help us understand the flex-
ibility of the process of hearing gender and how social changes might produce different kind
of linguistic and/or cognitive perceptions.
Finally, there is a need to connect linguistic research on trans speakers with the needs of
trans communities. Much of the published research on trans people’s voices is carried out
by speech pathologists, who have an interest in the topic because some trans people – par-
ticularly those who wish to feminise their voices – seek out speech therapy as part of their
transitions. However, much of this research offers problematic perspectives on the nature
of the gendered voice while also pathologising and denaturalising trans people’s identi-
ties. Culturally sensitive sociophonetic analyses can inform new approaches towards vocal
change that might be more empowering for transgender communities. Far from a niche
theoretical concern, the way the gendered voice is conceptualised – both in scholarly and
public domains, both for trans and cis speakers – has real implications for the liveability of
trans lives. We must consider how the gender of the voice acts as an arena for transphobia
and how our field may inadvertently contribute towards the exercise of linguistic violence.
If sociolinguists hope to benefit from the insight of trans speakers, to make linguistics a field
that welcomes trans scholars, and to fulfil our ethical commitments to the communities we
study, now is the time to reconsider our assumptions and discover new ways to advocate for
sociolinguistic justice in a gender diverse world.

Notes
1 Not all non-binary individuals identify as ‘trans’ and some trans-identified people use the term to
refer only to trans women or men. As the discussion in the following sections makes clear, ‘transgen-
der’ is a word with a number of potential definitions. For the purposes of this chapter, I have adopted
the more expansive option, which includes people who are non-binary as well as those who are
binary-identified.
2 ‘Assigned sex’ and ‘assigned gender’ are typically used interchangeably since the assignment process
assumes that both sex and gender are identified at birth.
3 This category includes participants who identified as bisexual or pansexual (i.e. attracted to all gen-
ders), who also typically described themselves as queer.
4 ‘Latinx’ is a gender-neutral replacement for Latina/o or Latin@ that includes non-binary as well as
female and male identities.

Further reading
Simpson, A. P. (2009) ‘Phonetic differences between male and female speech’, Language and
Linguistics Compass, 3, pp. 621–640.
Simpson provides an overview of research on gender differences in the voice that draws on
constructivist approaches to language and gender. It also challenges some often-repeated claims about

86
 Gender diversity and the voice

women’s and men’s voices, such as the notion that gender differences in formants are caused by
differences in body size.
Stryker, S. and Whittle, S. (eds) (2006) The transgender studies reader. New York: Routledge.
Stryker, S. and Aizura, A. (eds) (2013) The transgender studies reader, vol. 2. New York: Routledge.
The two volumes of this reader are central texts in trans studies, making them an excellent starting
point for linguists who are beginning to learn about or work with trans and otherwise non-normatively
gendered speakers.
Zimman, L. (2018) ‘Transgender voices: insights on identity, embodiment, and the gender of the
voice’, Language and Linguistic Compass, 12, p. e12284.
This article presents a brief overview of the kinds of questions transgender voices can help answer
and the sociophonetic lessons to be gleaned from the existing research on trans people’s speech.
Zimman, L. (forthcoming) ‘Transgender’, in Hall, K. and Barrett, R. (eds) The Oxford handbook of
language and sexuality. Oxford: Oxford University Press.
This paper synthesises three decades of research on transgender people’s linguistic practices. It
also sketches a framework, trans linguistics, that can be used for the empowerment and ethical study
of language in trans communities.

Related topics
Non-binary approaches to gender and sexuality; perception of gender and sexuality; sexuality as non-
binary: a variationist perspective; gender and sexuality normativities; an ethnographic approach to
compulsory heterosexuality.

References
Austin, J. L. (1962) How to do things with words. Cambridge, MA: Harvard University Press.
Azul, D., Nygren, U., Södersten, M., and Neuschaefer-Rube, C. (2016) ‘Transmasculine people’s
voice function: a review of the currently available evidence’, Journal of Voice, 31, pp. 261.e9–261.
e23.
Bekker, I. and Levon, E. (2017) ‘The embedded indexical value of /s/-fronting in Afrikaans and South
African English’, Linguistics, 55, pp. 1109–139.
Borba, R. (2015) ‘How an individual becomes a subject: discourse, interaction & subjectification at a
Brazilian gender identity clinic’, Working Papers in Urban Language & Literacies, 163, pp. 1–23.
Braun, V. and Kitzinger, C. (2001) ‘Telling it straight? Dictionary definitions of women’s genitals’,
Journal of Sociolinguistics, 5, pp. 214–232.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Butler, J. (1993) Bodies that matter: on the discursive limits of “sex”. New York: Routledge.
Campbell-Kibler, K. (2011) ‘Intersecting variables and perceived sexual orientation in men’, American
Speech, 86, pp. 52–68.
Chappell, W. (2016) ‘On the social perception of intervocalic /s/ voicing in Costa Rican Spanish’,
Language Variation and Change, 28, pp. 357–378.
Clifford, J. and Marcus, G. E. (eds) (1986) Writing culture: the poetics and politics of ethnography.
Berkeley: University of California Press.
Crenshaw, K. (1989) ‘Demarginalizing the intersection of race and sex: a black feminist critique of
antidiscrimination doctrine, feminist theory, and antiracist politics’, University of Chicago, 1989,
pp. 314–343.
Davis, J. L. (2014) ‘More than just “gay Indians”: intersecting articulations of two-spirit gender,
sexuality, and indigenousness’, in Zimman, L., Davis, J. L., and Raclaw, J. (eds) Queer excursions:
retheorizing binaries in language, gender, and sexuality. New York: Oxford University Press.
Flipsen, P. Jr., Shrilberg, L., Weismer, G., Karlsson, H., and McSweeny, J. (1999) ‘Acoustic
characteristics of /s/ in adolescents’, Journal of Speech, Language, and Hearing Research, 42, pp.
663–677.

87
Lal Zimman

Fuchs, S. and Toda, M. (2010) ‘Do differences in male versus female /s/ reflect biological or
sociophonetic factors?’, in Fuchs, S., Toda, M., and Zygis, M. (eds) An Interdisciplinary guide to
turbulent sounds. Berlin: Mouton de Gruyter, pp. 281–302.
Geertz, C. (1973) ‘Thick description: toward an interpretive theory of culture’, in Geertz, C. (ed) The
interpretation of cultures: selected essays. New York: Basic Books, pp. 3–32.
Hall, K. (2005) ‘Intertextual sexuality: parodies of class, identity, and desire in liminal Delhi’, Journal
of Linguistic Anthropology, 15, pp. 125–144.
Hazenberg, E. (2016) ‘Walking the straight and narrow: linguistic choice and gendered presentation’,
Gender & Language, 10, pp. 270–294.
Heffernan, K. (2004) ‘Evidence from HNR that/s/is a social marker of gender’, Toronto Working
Papers in Linguistics, 23. Available at: http://twpl.library.utoronto.ca/index.php/twpl/article/view/
6208 (Accessed: 10th September 2018).
Henton, C. G. (1989) ‘Fact and fiction in the description of female and male pitch’, Language &
Communication, 9, pp. 299–311.
Hollien, H., Green, R., and Massey, K. (1994) ‘Longitudinal research on adolescent voice change in
males’, Journal of the Acoustical Society of America, 96, pp. 2646–2654.
Holmes-Elliott, S. and Levon, E. (2017) ‘The substance of style: gender, social class and interactional
stance in /s/-fronting in southeast England’, Linguistics, 55, pp. 1045–1072.
Intersex Society of North America (2008) How common is intersex? Available at: http://www.isna.org
/faq/frequency 6208 (Accessed 10th September 2018).
Jackson, P. A. (2004) ‘Gay adaption, tom-dee resistance, and kathoey indifference: Thailand’s gender/
sex minorities and the episodic allure of queer English’, in Leap, W. L. and Boellstorff, T. (eds)
Speaking in queer tongues: globalization and gay language. Urbana: University of Illinois Press.
King, B. W. (2016) ‘Becoming the intelligible other: speaking intersex bodies against the grain’,
Critical Discourse Studies, 13, pp. 359–378.
Labov, W. (1972) ‘Some principles of linguistic methodology’, Language in Society, 1, pp. 97–120.
Loveday, L. (1981) ‘Pitch, politeness and sexual role: an exploratory investigation’, Language and
Speech, 24, pp. 71–88.
McConnell-Ginet, S. (1983) ‘Intonation in a man’s world’, in Thorne, B., Kramarae, C., and Henley,
N. (eds) Language, gender and society. Rowley, MA: Newbury House Publishers, pp. 69–88.
Motschenbacher, H. (2009) ‘Speaking the gendered body: the performative construction of commercial
femininities and masculinities via body-part vocabulary’, Language in Society, 38, pp. 1–22.
Munson, B. (2007) ‘The acoustic correlates of perceived masculinity, perceived femininity, and
perceived sexual orientation’, Language and Speech, 50, pp. 125–142.
Nicholson, L. (1994) ‘Interpreting gender’, Signs: Journal of Women in Culture and Society, 20, pp.
79–105.
Pharao, N., Maegaard, M., Møller, J. S., and Kristiansen, T. (2014) ‘Indexical meanings of [s+]
among Copenhagen youth: social perception of a phonetic variant in different prosodic contexts’,
Language in Society, 43, pp. 1–31.
Podesva, R. J. (2007) ‘Phonation type as a stylistic variable: the use of falsetto in constructing a
persona’, Journal of Sociolinguistics, 11, pp. 478–504.
Podesva, R. J. and Van Hofwegen, J. (2016) ‘/s/exuality in small-town California: gender normativity
and the acoustic realization of /s/’, in Levon, E. and Mendes, R. B. (eds) Language, sexuality, and
power: studies in intersectional sociolinguistics. New York: Oxford University Press, pp. 168–188.
Sachs, J. (1975) ‘Cues to the identification of sex in children’s speech’, in Thorne, B., and Henley, N.
(eds) Language and sex: difference and dominance. Newbury, MA: Newbury House Publishers,
pp. 172–151.
Simpson, A. P. (2009) ‘Phonetic differences between male and female speech’, Language and
Linguistics Compass, 3, pp. 621–640.
Strand, E. A. (1999) ‘Uncovering the role of gender stereotypes in speech perception’, Journal of
Language and Social Psychology, 18, pp. 86–99.

88
 Gender diversity and the voice

Stuart-Smith, J. (2007) ‘Empirical evidence for gendered speech production: /s/ in Glaswegian’, in
Cole, J. and Hualde, J. I. (eds) Laboratory phonology, vol. 9. New York: Mouton de Gruyter, pp.
65–86.
Walker, A., García, C., Cortés, Y., and Campbell-Kibler, K. (2014) ‘Comparing social meanings across
listener and speaker groups: the indexical field of Spanish /s/’, Language Variation and Change,
26, pp. 169–189.
Yuasa, I. P. (2008) Culture and gender of voice pitch: a sociophonetic comparison of the Japanese and
Americans. London: Equinox Publishing.
Zimman, L. (2013) ‘Hegemonic masculinity and the variability of gay-sounding speech: the perceived
sexuality of transgender men’, Journal of Language and Sexuality, 2, pp. 1–39.
Zimman, L. (2014) ‘The discursive construction of sex: remaking and reclaiming the gendered body
in talk about genitals among trans men’, in Zimman, L., Davis, J. L., and Raclaw, J. (eds) Queer
excursions: retheorizing binaries in language, gender and sexuality. New York: Oxford University
Press, pp. 13–34.
Zimman, L. (2015) ‘Transmasculinity and the voice: gender assignment, identity, and presentation’, in
Milani, T. (ed) Language and masculinities: performances, intersections, dislocations. New York:
Routledge, pp. 197–219.
Zimman, L. (2016) ‘Sociolinguistic agency and the gendered voice: metalinguistic negotiations of
vocal masculinization among female-to-male transgender speakers’, in Babel, A. (ed) Awareness
and control in sociolinguistic research. New York: Cambridge University Press, pp. 253–277.
Zimman, L. (2017a) ‘Gender as stylistic bricolage: gransmasculine voices and the relationship
between fundamental frequency and /s/’, Language in Society, 46, pp. 339–370.
Zimman, L. (2017b) ‘Variability in /s/ among transgender speakers: evidence for a socially grounded
account of gender and sibilants’, Linguistics, 55, pp. 993–1019.
Zimman, L. (2018) ‘Transgender voices: insights on identity, embodiment, and the gender of the
voice’, Language and Linguistics Compass, 12, p. e12284.

89
 Part II
Anthropological and
ethnographic approaches
 6
Ethnography and the shifting
semiotics of gender and sexuality
Kira Hall and Jenny L. Davis (Part II leads)

Introduction
This chapter focuses on ethnography as a multi-method research approach in the study of lan-
guage, gender, and sexuality. Based on the practice of long-term participatory fieldwork, the
approach primarily originated within cultural and linguistic anthropology, where it remains
the central anchor of research today. Yet, as seen in the following chapters in Part II, ethnog-
raphy has also been taken up by scholars in diverse fields across the humanities and social
sciences and shaped to fit the particularities of each discipline. Several recent collections
have addressed the use of ethnography within language-oriented fields such as linguistic
anthropology (Perrino and Pritzker forthcoming), linguistic ethnography (Snell, Shaw, and
Copland 2015), and ethnography of communication (Kaplan-Weinger and Ullman 2015).
Our overview focuses on the use of ethnography within the now robust tradition of research
in the field of language, gender, and sexuality.
Our discussion highlights the ways that ethnography enables the analysis of semiosis –
here defined as sign processes that produce social meaning – as embedded in social context.
This approach is uniquely appropriate to the field’s long-held understanding of gender and
sexuality as intertwined social systems that are brought into being through everyday dis-
cursive practice. Although the development of this understanding is often traced to Butler’s
(1990) philosophical work on performativity, it is also evident in early language and gen-
der scholarship informed by ethnography, including research on indexicality (Ochs 1992),
language ideology (Gal 1989), and communities of practice (Eckert and McConnell-Ginet
1992). The ethnographers who advanced these formative concepts, each in different ways,
enabled the field’s later uptake of Butler’s work by countering sex-based generalisations
with a dynamic vision of gender as produced in everyday discourse (for a review, see Hall,
Borba, and Hiramoto 2021). Each stressed the crucial role played by social context in this
production, establishing ethnography as a necessary partner to the analysis of discourse. In
one of the chapters appearing in Part II, Philips, a leading ethnographer of language and
social life, calls this partnership ‘anthropological discourse analysis’. If gender is cultivated
through community-based practice as a ‘dynamic verb’, as Eckert and McConnell-Ginet
(1992: 462) argue in their early influential review, then ethnography is the approach for

93
Kira Hall & Jenny L. Davis

tracing those dynamics in discourse (on ethnography as a central approach in language,

gender, and sexuality research, see Besnier and Philips 2014; Gaudio 2019; Hall 2009).
We have come together to co-author this discussion as two linguistic anthropologists
who are deeply committed to what ethnography can bring to the social analysis of language,
even as we acknowledge the important critiques made of the method by each new genera-
tion of scholars. In fact, ethnography is one of the ‘most critiqued’ methods in the social
sciences, in part because it asks for a kind of reflexivity from the researcher that methods
aspiring to a dominant model of scientific objectivity do not share. Today’s ethnographers
are trained to be suspicious of claims to objectivity, holding that all research – even research
based purely in quantitative methods – is in some sense influenced by the position of the
researcher. Certainly, our own positions as a ‘native ethnographer’ writing from the inside
about Indigenous communities in the United States (e.g. Davis 2014, 2018, 2019) and a
‘foreign ethnographer’ writing from the outside about Hindi-speaking communities in India
(e.g. Hall 2005, 2009, 2019) affects the kind of data we collect and the type of analysis
we pursue. This reflexive awareness arises in the very act of doing fieldwork, whether in
a village, at school, in front of the television, or online. In ethnography, researchers do not
do objective observation, collecting specific pre-determined information from a detached
vantage point; rather, they do participant observation, taking part in the everyday practices
that are formative to the social, cultural, and linguistic behaviours they analyse. In the broad
interdisciplinary study of language and society subsumed under sociocultural linguistics
(Bucholtz and Hall 2008), these practices include the face-to-face interactions that are the
focus of more traditional fieldwork alongside the digital interactions that pervade twenty-
first-century social life. We are all participant observers of the media systems that surround
us; an ethnographic sensibility makes this participation the subject of analysis.
This chapter advances an understanding of social context as situated in a specific time
and place yet complexly informed by what came before and what exists elsewhere. This
deep contextualisation is the hallmark of ethnography and, as we argue in the pages that fol-
low, undergirds all phases of ethnographic activity, from collection and analysis to writing
and dissemination. The term ‘ethnography’, in our view, comprises much more than simply
‘describing a social group’, as its Greek etymology (ethno ‘social group’ + graphia ‘descrip-
tion’, ‘writing’) may suggest. As a kind of describing that is based on the author’s participa-
tion in the practices of others, ethnography refers to the process of research as well as its
product, involving much more than narrowly defined tools for data collection. As research-
ers of language in social life, we have found it challenging to represent the dynamism of
gender and sexuality in published work: How can we write about a specific time and place
in a way that acknowledges the ongoing processual nature of that particularity? We suggest
that ethnography offers an answer through its attention to the conceptual triad of practice,
ideology, and theory. We draw from our own work and the excellent work featured in Part
II to illustrate how ethnography, designed anew to encompass the heavily mediatised nature
of contemporary sociality, enables researchers to assess how gender and sexuality come to
matter in the semiotic exchange of everyday life.

Practice
The concept of practice runs deep in language, gender, and sexuality scholarship. Prominent
lines of research assume Bourdieu’s (1977) influential understanding of language as a prac-
tice that shapes, through repetition, a social actor’s habitus, or way of being in the world.

94
 Ethnography and shifting semiotics

Where the concept of practice has perhaps surfaced most robustly in the field is in research
focused on ‘communities of practice’ – a term initially advanced by Lave and Wenger (1991)
in their exposition of learning as a process of becoming a member of a sustained community.
Eckert and McConnell-Ginet’s (1992) introduction of this model into language and gender
scholarship countered broad-scale generalisations about women and men’s language pat-
terns found in early research in the field. In a community of practice view, links between
language and gender are not merely a binary product of childhood language socialisation, as
scholarship advocating a two-cultures understanding of gender often implied. Rather, these
links are ‘learned’ throughout the life course as social actors become members of diverse
communities that cultivate the relationship between language and gender differently.
Ethnographic research inspired by the community of practice tradition has convincingly
shown that indexical knowledge – that is, knowledge of how linguistic forms are connected
to social meanings – arises from sustained participation with others. In her research on uses
of ethnic jokes by lesbian and transgender youth in Delhi, Hall (2019) identifies this kind of
knowledge as ‘indexical competence’ to emphasise the exclusionary semiotic mastery that
is required for localised forms of identity work (see also Parish and Hall 2021). The impor-
tance of this form of competence is amply illustrated by research on organisations of gender
in educational youth environments such as high schools, institutions that are recognised
in social scientific scholarship as vital sites of identity formation. Ethnographers entering
these sites have explored the ways that competing youth communities ascribe social mean-
ing to constellations of language, apparel, embodiment, and space as a means of achieving
stylistic distinction (e.g. Bucholtz 2011; Eckert 2000; Mendoza-Denton 2008; Pichler 2009;
Shankar 2008; Smalls 2018). Consider, for example, Bucholtz’s (1999) influential account
within language and gender studies of a community of female nerds at a Northern California
high school. As the girls in her study engage with one another across multiple interactions,
they learn to use and interpret hyper-standardised uses of the English language as indexical
of a female nerd identity that opposes the perceived superficiality of more popular peers.
Community members display nerd identity by demonstrating knowledge of these indexical
relations and the ideologies that inform them.
The link between knowledge and practice is what makes ethnography, with its key com-
ponent of participant observation, such an important approach for understanding the social
analytics of language. A primary strength of ethnographically informed analysis is its atten-
tion to the ways that indexical relations are situated within time and space. Through longi-
tudinal participation in situated communities, researchers can come to know the multiple,
ever-shifting, and often competing indexical relations that give meaning to gender and sexu-
ality. Scholars often comment on the paradoxical nature of participant observation: How
can one be both a participant and an observer? But the term is paradoxical only within a
perspective holding that observational knowledge must be detached from participation to
escape bias. This perspective, still dominant across the social and natural sciences, is built
on the premise that ‘knowing’ must exist independently from ‘being’ – that we can only
know about the world when we refrain from participating in it (see discussion in Ingold
2014). However, community-based research in sociocultural linguistics has demonstrated
that our understanding of how to use and interpret language (‘knowing’) is in fact culti-
vated through our everyday interactions with others (‘being’). When cultural anthropologist
McGranahan characterises ethnography as a ‘unique way of knowing’, she is speaking to
the sensibility that derives from this cultivation: ‘The ethnographic consists of the rhythms
and logics through which we, in sociocultural groups, collectively make, and make sense

95
Kira Hall & Jenny L. Davis

of, the world’ (2018: 2). In this respect, we are all participant observers, acquiring indexical
knowledge as we engage with others through our bodies, minds, and senses. As a research
method, participant observation is designed to approximate the learning process that takes
place in everyday life, as lived experience.
Nevertheless, this approximation is always partial, given the ethnographer’s peculiar
investment in the learning process. Feminist anthropologists have argued for decades that
the asymmetry between ethnographer and subject has consequences and requires care.
Scholars in language, gender, and sexuality do not often display the self-reflection seen
in certain genres of anthropological writing, yet the field’s ongoing concern with power
relations requires researchers to be attentive to biases that unavoidably pervade all stages
of the research process, whether personal, cultural, or institutional. For instance, how
might our own social backgrounds affect the kinds of things we notice in the field? How
might our previous histories of knowing and being influence the way we analyse the
data we collect? This attentiveness is precisely what is captured by the term ‘reflexivity’.
Ethnographers of language, as a special category of ethnographers, must also consider
the semiotic biases that inform our entry into worlds of practice different from our own.
How might our interpretations of language be influenced by life experiences in commu-
nities that view the relationship between linguistic form and social meaning differently?
As Briggs (1986) argued over three decades ago when reflecting on his research among
Spanish speakers in northern New Mexico, the assumptions academics may hold about
communicative events as seemingly ubiquitous as the interview can lead us to ask the
wrong kinds of questions and to draw interpretations that may inaccurately reflect the
perspectives of those we write about.
It is for this reason that the feminist concept of intersectionality figures so prominently in
ethnographically based research on language, gender, and sexuality (cf. Chun and Walters
forthcoming; Cornelius 2020; Levon and Mendes 2016). Because identity is multiply con-
stituted by engagement in diverse communities of practice, there can never be seamless
congruence between a researcher’s subjectivity and the subjectivity of the individuals under
focus. Rather, as language-and-gender scholar Jacobs-Huey has pointed out, ‘ethnographic
fieldwork is an intersubjective process that entails an interaction of various subjectivities’
(2002: 791). It is this acknowledgement of intersubjectivity that transforms ethnography
into a feminist method, compelling us to see our interlocutors not as objects of study but
rather partners in discovery. This brings us to the second concept we see as integral to eth-
nographic analysis, ‘ideology’.

Ideology
In the course of our respective careers, we have each encountered colleagues in linguistics
who view ethnographic work as ‘narrow’. A recent event in one of our departments comes
to mind, when a sociolinguistic presentation analysing over 14,000 tokens of the sound
/s/ as used by a gender variant community was characterised as based on ‘small data’. We
counter with the following response: ethnography is big data. It is the kind of data that can
be collected only by immersive participation over an extended period of time, often involv-
ing observation of hundreds or even thousands of hours of interaction. In fact, as Radin
(2017) and Lemov (2017) point out, Big Data owes much to ethnography and the associated
methods outlined in this chapter. Consider, for example, the Pima Indian Diabetes Dataset
(PIDD) that now forms an integral part of the UC Irvine Machine Learning Repository

96
 Ethnography and shifting semiotics

responsible for testing data-mining algorithms (Radin 2017: 53) or the more than 300 hours
of anthropological interviews with Hopi consultant Don C. Talayesva that are foundational
to the web-based full-text database eHRAF World Cultures (Human Relations Area Files).
The difference between big data and small data, then, is often more a matter of how the
‘local’ is acknowledged:

What makes data “big” is not so much its size – though that is relevant too – but its abil-
ity to radically transcend the circumstances and locality of its production. Computers
and algorithms make that possible, but understanding the politics of Big Data also
requires attention to the creation and processing of the data itself, including the recog-
nition that it often comes from living, breathing people.
(Radin 2017: 45–46)

Because ethnographers of language and social life investigate the ways that linguistic forms
and social meanings emerge within an array of practices that include consumption, cultural
traditions, education, kinship relations, media, politics, and religion, there is nothing ‘nar-
row’ or ‘small’ about ethnography. On the contrary, ethnographers examine a situated aspect
of semiotic practice in comprehensive detail as a means of discovering the historical, cul-
tural, political, and interactional processes that invest language with social meaning in the
(living, breathing) lives of those who use it.
To recall a well-cited phrase from Silverstein (1985), ethnographers seek to uncover
the ‘total linguistic fact’ – that is, the dialectic interaction between linguistic form (struc-
ture), social use (practice), and human reflection on the meaning of those forms in use
(ideology) (see also Woolard 2008). This totality makes ethnography time-intensive with
respect to both data collection and analysis, so much so that it often disadvantages schol-
ars in departments expecting rapid publication. And yet for an ethnographer, a focus on
only one or two of these elements instead of three would betray the fundamental anthro-
pological insight that relations between form and meaning are forged in ‘situations of
interested human use mediated by the fact of cultural ideology’ (Silverstein 1985: 220).
For ethnographers of language and social life, ideology is the glue that holds form and
meaning together. When community members ascribe gendered meanings to a certain
sign form – whether a phonetic variable, a taboo term, an intonational contour, or a move-
ment of the body – they do so through appeal to local and broader ideologies that give
sense to everyday life, that bring a logic to its messiness. Language ideologies are never
really just about language; rather, they reflect the prejudices and privileges of the social
systems in which they are situated.
Participant observation is often held up as the investigative practice that makes ethnog-
raphy unique, but ethnography is inherently a mixed methodology. It involves a methodo-
logical complexity that is in many senses iconic of the complexity of social life (Blommaert
2007). While the field-based method of participant observation is the bedrock of ethnogra-
phy, it is always used together with a variety of other methods (some specific to sociocul-
tural linguistics; others associated with cultural anthropology or other fields), among them
sociolinguistic interviews, archival research, media analysis, collaborations with field-based
research partners, recording, transcription, translation, discourse analysis, and fieldnotes.
The multifaceted methodologies that result from these combinations are designed to make
the ideological bond between micro and macro discoverable. A central tenet of ethnography
is that more information is always good information, particularly when taken from data

97
Kira Hall & Jenny L. Davis

sources that illuminate the focus of investigation from different spacetimes. For ethnog-
raphers of language, one of the most challenging aspects of this tenet is that the methods
associated with this diversity may lead to contradictory findings regarding language use.
For example, the method of sociolinguistic interviews may uncover ideas about language
use that are not borne out in an analysis of actual language practice. It is in this disconnect
that ideology is found.
A case in point comes from Hall’s (1995) early dissertation research among Hindi-
speaking hijras in northern India (see also Hall and O’Donovan 1996), a group whose
members identify as na mard na aurat, ‘neither man nor woman’. When conducting socio-
linguistic interviews with members of the community, Hall repeatedly heard the refrain ‘We
never speak like men! We always address each other as women!’. Yet longitudinal partici-
pant observation of hijras’ actual speech practices, coupled with discourse analysis, revealed
that they did in fact often use masculine reference for each other and even for themselves.
Why this disconnect between saying and doing?
For ethnographers of language, methods such as sociolinguistic interviews highlight
the ideologies of language and society that background speakers’ discursive behaviours.
Further interviews revealed that hijras, most of whom were raised as boys, wished to dis-
tance themselves from the masculine representations of their youth. This stance was made
stronger by society’s unwillingness to address them in the feminine, which to them indi-
cated a lack of respect. However, in actual language practice, a pattern emerged whereby
these same hijras would use masculine self-reference among themselves when establish-
ing relations of hierarchy. The disconnect between saying and doing is thus explained
as the difference between a public-facing communal identity that distances itself from
masculinity (‘indirect indexicality’, in Ochs’s 1992 terminology) and an in-group prac-
tice-based identity that deploys masculine self-reference for certain conversational ends
(‘direct indexicality’, in Ochs’s 1992 terminology). Should Hall have stopped at the soci-
olinguistic interview she would not have seen the complexity of identification practices
within the community, where hijras exploit broader indexical links between language
and gender to take stances of hierarchy and solidarity. In fact, it was these shifting uses
of gender morphology that enabled Hall to understand hijra positionality as non-binary.
Davis’s work (2014, 2019) in a Native American Two-Spirit community in the western
United States additionally illustrates how ethnography can explain a contradictory use of
identity labels, in this case through a consideration of local vs. regional social contexts.
In her fieldwork, Davis encountered several instances in which individuals identifying as
Two-Spirit (Native Americans who are spiritually both male and female) simultaneously
used and contested a variety of terms for their identity, among them ‘gay’, ‘trans’, and
‘queer’; ‘Two-Spirit’; and tribally specific terms such as nadlé (taken from Diné/Navajo).
Multi-sited discourse analysis revealed seemingly contradictory transcripts, both within
single events and across multiple discourse events, in which terms used as synonymous
in some instances were used with different meanings in others. Davis argues that these
terms are contextually polysemous: their meanings change based on factors that include
the audience’s presumed knowledge (or lack thereof) of Indigenous cultures in North
America as well as discourse uses of micro- and macro-categories with which Two-Spirit
identity might be compared.
As in Hall’s research, the disconnect between ideology and practice becomes most vis-
ible in moments when language use appears to contradict community members’ previous
statements. Consider, for example, Brent’s discussion of his use of these different terms,
which he shared with peers in a regional Two-Spirit group (Excerpt 1):

98
 Ethnography and shifting semiotics

Excerpt 1

1 Brent: that is actually one of the biggest misconceptions.

2 on on the reservations
3 (.3)
4 all these tribes actually had names for for Two-Spirit
5 people.
6 but how people see them as
7 just like ‘oh they just mean gay’ but there is a deeper
8 root
9 James: ((cough))
10 Brent: um that um for nadhle.
11 I’m sorry I say nadhle more than I say Two-Spirit cause
12 I (hhh)’m just stubborn that way
13 TS Group: ((laughter))

Brent’s justification for using ‘nadhle’ (‘cause I’m just stubborn that way’; Lines 11–12)
indexes a belief that community-specific terms are more automatic or even more ‘natural’
for Native Americans than the term ‘Two-Spirit’. His reluctance to use the more generalised
term echoes Epple’s critique of the broad academic use of terms such as ‘berdache’, ‘gay’,
and even ‘Two-Spirit’, which in her view lack cultural and temporal grounding: ‘current
analytical concepts simply do not accommodate the simultaneous distinctness (identity as
nádleehí [plural]) and fluidity (identity as context-dependent) of nádleehí’s self-descrip-
tions’ (1998: 268). It is perhaps for this reason that when group members offered accounts
in a formal presentation of specific historical figures now included under the Two-Spirit
umbrella, they referred to such figures as ‘Two-Spirit’ even as they used the term specific
to that individual’s tribal affiliation: for instance, winkte (Lakhota), nadhle (Dine), and lha-
mana (Zuni). Individuals in the group were thus very attentive to using the appropriate local
designation for historical figures as well as for themselves and other group members.
However, it is important to note that these tribally specific terms were asserted in a
regional, multi-tribal Two-Spirit group, not in a local organisation comprised of individuals
from a single Nation. Participant observation combined with the analysis of discourse in
varied settings revealed that these same group members strongly identified as Two-Spirit in
ways that were relevant to their daily lives. The importance placed on local Indigenous iden-
tity labels in the above example in no way contradicts the appropriateness of the Two-Spirit
label as another facet of these speakers’ identities. In fact, the mutual dependence of local
and multi-tribal terms could be observed in their formal presentations precisely because the
presenters were recognised as holding multiple forms of identification that crossed local and
regional lines.
The above examples taken from our respective fieldwork sites illustrate what can be
gained by combining participant observation with more specifically linguistic methods
such as sociolinguistic interviews and multi-sited discourse analysis. Identity claims are
never simple; like all features of language, they emerge from complex social processes that
inevitably bring semiotic instability. The digital recordings that constitute the gold standard
of sociocultural linguistic data collection are important, but as static snapshots of a much
longer discursive history, they are never enough. At the same time, they are sometimes not
even necessary, as demonstrated by the rich linguistic insights offered by ethnographers who
are asked to refrain from using this method due to a community’s marginalisation, as seen

99
Kira Hall & Jenny L. Davis

in Borba’s (2018) work on Brazilian sex workers and Gaudio’s (2019) work on Nigerian
‘yan daudu. Although rarely highlighted in language, gender, and sexuality scholarship,
the anthropological method of writing fieldnotes is a powerful tool for tracking the shifting
meanings of language across time and space. As qualitative data ideally recorded immedi-
ately after a research encounter, fieldnotes can provide important descriptive evidence (both
factual and reflective) about the discourse context under investigation. Indeed, as Goldstein
(2017) shows in her analysis of FBI director James Comey’s scrupulously detailed memos
of his interactions with President Trump, fieldnotes, when done well, may even bring to life
the behind-the-scenes manipulations of a corrupt leader. The importance of Comey’s note-
taking after his meetings with the President was not lost on the media; in fact, his fieldnotes
offered credibility to his testimony before the Senate Intelligence Committee and helped
break the usual ‘he said, he said’ stasis. This reminds us of the importance of the -graphia
in ethnography’s etymology. In comparison to other approaches, ethnography is especially
concerned with the descriptive techniques of writing that will best display the complexity
of the people under focus, which for sociocultural linguists, also includes their language
practices.
This brings us again to the topic of reflexivity. We suggest that all of ethnography’s
methods, when adopted and adapted for the needs of a study, require the reflexivity that we
often associate with participant observation. Consider, for example, transcription, the work-
horse method used by discourse analysts to represent language practice in written form.
As Bucholtz reminds us, ‘transcription is not solely a research methodology for under-
standing discourse but also, and just as importantly, a sociocultural practice of representing
discourse’ (2007: 785). Sociocultural linguists have hundreds of transcription systems to
choose from, each with their own set of conventions. Decisions about which conventions
to use in a given transcript may be driven by research needs, but they also have ‘potentially
significant analytical and political consequences’ (2007: 786). In this sense, methods such
as transcription are inherently theoretical (cf. Ochs 1979), a point that leads us to the final
element in our conceptual triad, ‘theory’.

Theory
In an article entitled ‘Ethnography as theory’, Nader reflects on key ethnographic texts
across 100 years of cultural anthropology and asserts the following: ‘Ethnography, what-
ever it is, has never been mere description. It is also theoretical in its mode of description.
Indeed, ethnography is a theory of description’ (2011: 211, emphasis in the original). As we
conclude this chapter, we want to reflect on the ways that this claim is also relevant to the
history of ethnographic research on language, gender, and sexuality.
The first observation to make in this regard is that the field’s use of ethnography has
evolved in tandem with shifts in theoretical understandings of gender and sexuality. For
instance, early ethnographies of non-Indo-European ‘women’s languages’ and ‘men’s lan-
guages’ in the first half of the twentieth century (e.g. Chamberlain 1912; Jespersen 1922)
are often characterised as descriptive, but their emphasis on the rigidity of linguistic gen-
der in non-European languages affirmed colonialist readings of these languages as primi-
tive (Hall 2003). In the second half of the century, ethnography was deployed by a new
generation of scholars to challenge broad generalisations made about women’s speech in
so-called difference models of language and gender. Research in sites such as Madagascar
(Keenan Ochs 1974), Hungary (Gal 1978), southern Mexico (Brown 1980), a US high
school (Eckert 1989), and a Philadelphia African American community (Goodwin 1990)

100
 Ethnography and shifting semiotics

brought complexity to the field’s unmarked focus on middle-class white speakers. In the
1990s, Butler’s (1990) theory of gender performativity inspired the application of ethno-
graphic work to non-normative organisations of language, gender, and sexuality in varied
locations (see, e.g. articles in Leap 1995; Livia and Hall 1997). Many of the ethnographies
emerging in this period were positioned as overtly political in their commitment to ‘queer-
ing’ a largely heterosexual and cisgender canon, hence the field’s name ‘queer linguistics’.
Similarly, the rise of multicultural feminism and its emphasis on intersectionality inspired
a deeper ethnographic consideration of the relationship between gender and race, as seen in
turn-of-the-century work by Jacobs-Huey (2006), Mendoza-Denton (2008), Morgan (2002),
and Zentella (1997).
In our current era of research on language, gender, and sexuality, ethnography contin-
ues to assist this decisively critical turn towards political advocacy for marginalised per-
spectives. Its diversity of method is now dedicated to the task of uncovering how gender
and sexuality articulate with systemic hierarchies of race, class, age, disability, colonial-
ism, imperialism, and geopolitics, among other topics. LGBTQ scholars are now using
ethnography to retheorise binaries (Zimman, Davis, and Raclaw 2014), counter cisgender
assumptions found in previous language and gender scholarship (Zimman 2020), estab-
lish the centrality of the body to sociolinguistic investigation (Calder 2019; King 2019;
Peck and Stroud 2015; Zimman and Hall 2010), and revise the queer theoretical concept
of normativity (Barrett 2017; Cashman 2019; Hall 2019; Hall, Levon, and Milani 2019).
Scholars of race are using ethnography to explore connections between language, sexuality,
and Blackness in ways that challenge the whiteness of previous work on gay male speech
(Cornelius 2020; Cornelius and Barrett 2020) and draw attention to everyday political pres-
sures confronting Black Queer Women (Lane 2019). Finally, scholars of the Global South
are using ethnographic methods to rethink organisations of gender and sexuality in Southern
contexts and thereby contest the dominance of Northern-originating forms of knowledge
production (Borba 2017; Deumert and Mabandla 2017; King 2017; Lazar 2017; Ostermann
2017; Shaikjee and Stroud 2017). This latter body of scholarship is particularly relevant
to our discussion, as it turns the reflexivity that is ethnography’s strength onto geopoliti-
cal exclusions in the language, gender, and sexuality canon (for programmatic statements,
see Hall, Borba, and Hiramoto 2021; Milani and Lazar 2017). To return to Nader’s point,
ethnography is never merely descriptive. Like all scientific methodologies, even those held
up by their practitioners as pillars of objectivity, ethnography is embedded in the broader
theoretical questions that motivate its application.
Our second and final observation concerns the importance of social theory to ethnogra-
phy more generally. Those of us who identify as ethnographers frequently characterise our
research approach as ‘bottom-up’, much like our colleagues in conversation analysis, a field
that shares our disciplinary roots in ethnomethodology. Our work is focused on the micro-
details of everyday discourse, collected across space and time in our capacity as participant
observers. At the same time, we share with our colleagues in the field of critical discourse
analysis an interest in top-down questions of power: How are broader social hierarchies
constituted through interaction? As researchers situated between the micro and the macro,
we would never characterise our method as ‘atheoretical’, as conversation analysts often do.
Rather, we make sense of the patterns we find across diverse sources of data by consulting
the ideas of those who have dedicated their careers to understanding social life, otherwise
known as social theorists. The importance of social theory is perhaps obvious to researchers
in language, gender, and sexuality, particularly given the field’s long-term intimacy with
evolving traditions of feminist and queer theory (Bucholtz 2014; Kramer 2016; McElhinny

101
Kira Hall & Jenny L. Davis

2003). And of course sociocultural linguists have also developed their own social theo-
ries, among them the social semiotic concepts of indexical order, enregisterment, style, and
language ideology. Together, such theoretical perspectives carry the potential to illuminate
patterns in our data, to make our claims regarding the workings of language in society more
robust. In addition, by engaging with broader social theoretical perspectives, we may be
able to persuade our colleagues in other socially oriented fields that language, no matter how
small, matters to societal organisations of gender and sexuality.
The authors of the five chapters that appear in Part II each make use of the triad of con-
cepts we have outlined in this essay: practice, ideology, and theory. Locating their analy-
sis in the practices and ideologies they have observed as participants, they build on social
theory in ways that highlight the importance of language to the constitution of gender and
sexuality. In keeping with the spirit of this Handbook, the authors are reflexive about their
methodologies, offering a wealth of perspectives on how ethnography may enable us to see
the workings of language, gender, and sexuality more clearly.
Shaw situates her ‘elite ethnography’ of UK parliamentary contexts within an emer-
gent tradition of scholarship known as ‘linguistic ethnography’ (Snell, Shaw, and Copland
2015), an approach that arose primarily in Europe. Shaw relies on a triangulation of meth-
ods that include participant observation, formal interviews, field notes, and archival work.
Longitudinal fieldwork in a variety of debating chambers revealed the ways that parliaments
share a habitus that prioritises some speakers and not others. For instance, Shaw details how
a ‘full view’ experience from the public galleries enabled a more robust understanding of
the gendered hierarchies that inform interaction in the chamber: Why is the female minister
sitting in a row of male ministers asked to fetch the First Minister a glass of water? This
same habitus appears to explain Assembly Member reactions to a female colleague who per-
formed ‘illegal sustained interventions’ when Cheryl Gillan, the British Secretary of State
for Wales, visited the chamber. Backstage interviews revealed that Assembly Members were
uncomfortable with their female colleague’s contrary behaviour, even as they disapproved
of Gillan’s elite ‘queen-mother like’ demeanour. On the surface, the Assembly Members’
disapproval of the behaviours of both women appears to affirm Lakoff’s (1975) early read-
ing of women’s language as ‘damned if you do, damned if you don’t’. But Shaw’s ethno-
graphic insights offer more complexity, exposing how the responses of the Welsh Assembly
Members (and indeed the female Assembly Member’s own behaviour) are also guided by
a history of tension between Wales and Westminster. Her analysis of how micro-details of
interaction connect to ‘macro questions of exclusion and power’ comes to life through her
reflexivity at every stage of the ethnographic process, an approach she views as distinctly
feminist.
Clark’s chapter also engages the theoretical work of Bourdieu when she asserts the use-
fulness of ethnography for exposing practices of symbolic violence – the unmarked forms of
non-physical violence that manifest in organisations of social hierarchy. Her focus is on the
compulsory heterosexuality found in everyday discourse: How do we analyse something so
systemic in conversation that even participants themselves may not recognise it? For Clark,
the answer to this question lies in ethnography’s ‘sustained engagement’ with the people we
study. Her analysis of interactional data collected over time through her participation on a
women’s hockey team brings to light the everyday grammar of compulsory heterosexual-
ity and its adverse effects on LGBTQ individuals. As seen in other queer linguistic work,
Clark’s use of ethnography in this chapter is overtly political: in her words, ‘a priority for
Queer linguistic ethnography is to reveal instances in which those unwritten rules [of inter-
action] require participants to adhere to a heteronormative framework’.

102
 Ethnography and shifting semiotics

Philips returns the focus to gender ideology as she reflects on her ethnographic work in
Tongan courtrooms in Polynesia. In a careful discussion of the ways ‘ideas about women vary
systematically across social domains’, Philips distinguishes uses of ethnography in linguistic
anthropology (her professional field) as requiring discourse analysis across time and space, a
method she identifies as ‘anthropological discourse analysis’. She compares discourses about
‘bad words’ in two different domains – the public domain of the Magistrate’s Courts and the
private domain of a women’s work group – to uncover the ways that gender ideology differs
across organisational contexts. Her work thus emphasises the importance of understanding
the ‘larger system’ in which ideologies about language and gender circulate, as the nature
of the activity may shape the way gender ideology emerges in the data. Importantly, Philips
also outlines how her analytic observations arose from collaborations with co-researchers in
the field who offered key insights as they recorded, transcribed, and translated discourse data
from different domains. When these backstage forms of linguistic labour go unrecognised,
ethnography retains its colonialist roots, extending the power relations inherent to fieldwork
to practices of description, authorship, and citation. However, when this linguistic labour is
the site of recognised collaboration, it can produce theoretical and methodological models
that better align researcher and community positions. In sum, ethnographic research is made
better – more honest, more feminist, even more insightful – when we are transparent about
the ways these backstage forms of collaboration unfold to shape our findings.
With Nagar’s chapter, we move to one of our field’s signature frameworks for the eth-
nographic investigation of language, gender, and sexuality: the community of practice (see
Holmes and Meyerhoff 1999). Nagar reflects on her use of this framework for understanding
meaning-making among jananas, a non-normative gender identity in India. Not all com-
munities have clearly defined boundaries; members of the janana community, for instance,
cannot be located in a ‘common space, profession, or cause’. Rather, they come together
around mutually defined practices, which ethnography, as a method ‘based in practice and
learning’, enabled her to discover. The excerpts she analyses from her conversations with
jananas in 2004 and 2006 suggest different and even contradictory views of janana identity,
yet through a diversity of ethnographic methods applied over time, she was able to ‘find’ the
shared practices that gave these divergent views meaning.
Finally, Varis reflects on her research in a discourse environment that has only recently
captured the attention of ethnographers of language: digital media. Her focus on an early
twenty-first-century online genre known for rapid semiotic shifts in gender and sexuality
– YouTube ‘camgirl’ broadcasts – provides a fitting conclusion to our discussion. We men-
tioned at the outset of this chapter that ethnography must be revised to reflect the highly
mediatised nature of current social life. Drawing on her immersion in social media cul-
ture, Varis constructs a compelling analysis of why the broadcasts of one female YouTuber,
Hannah Witton, have attracted almost 400,000 subscribers. She seeks to understand why
Witton’s videos of her encounters with normative reproductive practices such as menstrua-
tion and birth control are so tantalising for her viewers. The answer is found not just in the
agentive intimacy Witton displays in these videos, but also in the way her broadcasts are
mediated by ever-shifting online environments.
Varis suggests that digital ethnography is novel in its assumption of a ‘changing media and
communication landscape’, but as we see in the chapters that follow, all ethnographers grapple
with the challenge of following people and their social practices over time (see also methodo-
logical discussions in Goldstein 2020; Hall 2009; Wortham 2006; Wortham and Reyes 2015).
In fact, Philips asserts in her chapter on Tongan gender ideology that we can truly understand
larger systems of social organisation only by examining the ways ‘talk at different points in

103
Kira Hall & Jenny L. Davis

time are related’. Taken together, the work featured in this section reminds us that the strength
of ethnography lies in its ability to consider forms of discourse across time as well as space.
Like all social scientific research approaches, ethnography is imperfect, but its reflexivity,
diversity, and staying power are well suited to the elusive nature of social life.

References
Barrett, R. (2017) From drag queens to leathermen: language, gender, and gay male subcultures. New
York: Oxford University Press.
Besnier, N. and Philips, S. U. (2014) ‘Ethnographic methods for language and gender research’, in
Ehrlich, S, Meyerhoff, M., and Holmes, J. (eds) The handbook of language, gender and sexuality.
West Sussex: John Wiley & Sons, pp. 123–140.
Blommaert, J. (2007) ‘On scope and depth in linguistic anthropology’, Journal of Sociolinguistics,
11(5), pp. 682–688.
Borba, R. (2017) ‘Ex‐centric textualities and rehearsed narratives at a gender identity clinic in Brazil:
challenging discursive colonization’, Journal of Sociolinguistics, 21(3), pp. 320–347.
Borba, R. (2018) ‘Doing tricks: affordances and challenges for a sociolinguistics of sex work’, in
Hall, K. and Barrett, R. (eds) The Oxford handbook of language and sexuality. New York: Oxford
University Press. (Published at Oxford Handbooks Online.)
Bourdieu, P. (1977) Outline of a theory of practice. New York: Cambridge University Press.
Briggs, C. L. (1986) Learning how to ask. New York: Cambridge University Press.
Brown, P. (1980) ‘How and why are women more polite: some evidence from a Mayan community’,
in McConnell-Ginet, M., Borker, R., and Furman N. (eds) Women and language in literature and
society. New York: Praeger Publishers, pp. 111–136.
Bucholtz, M. (1999) ‘“Why be normal?”: language and identity practices in a community of nerd
girls’, Language in Society, 28(2), pp. 203–223.
Bucholtz, M. (2007) ‘Variation in transcription’, Discourse Studies, 9(6), pp. 784–808.
Bucholtz, M. (2011) White kids: language, race, and styles of youth identity. Cambridge: Cambridge
University Press.
Bucholtz, M. (2014) ‘The feminist foundations of language, gender, and sexuality research’, in
Ehrlich, S, Meyerhoff, M., and Holmes, J. (eds) The handbook of language, gender and sexuality.
West Sussex: John Wiley & Sons, pp. 21–47.
Bucholtz, M. and Hall, K. (2008) ‘All of the above: new coalitions in sociocultural linguistics’,
Journal of Sociolinguistics, 12(4), pp. 401–431.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Calder, J. (2019) ‘The fierceness of fronted /s/: linguistic rhematization through visual transformation’,
Language in Society, 48(1), pp. 31–64.
Cashman, H. R. (2019) ‘What Phoenix’s jotería is saying: identity, normativity, resistance’, Language
in Society, 48(4), pp. 519–539.
Chamberlain, A. F. (1912) ‘Women’s languages’, American Anthropologist, 14, pp. 579–581.
Chun, E. and Walters, K. (2021). ‘At the crossroads of language, race, and sexuality’, in Hall, K. and
Barrett, R. (eds) The Oxford handbook of language and sexuality. New York: Oxford University
Press. (Published at Oxford Handbooks Online.)
Cornelius, B. R. (2020). ‘Gay (white) male speech and the e(race)sure of identity,’ in Hall, K. and
Barrett, R. (eds) The Oxford handbook of language and sexuality. New York: Oxford University
Press. (Published at Oxford Handbooks Online.)
Cornelius, B. R. (2020) ‘“You met my ambassador”: language and self-monitoring at the intersection
of race and sexuality’, in Alim, H.S., Reyes, A., and Kroskrity, P.V. (eds) The Oxford handbook of
language and race. New York: Oxford University Press, pp. 315–341.
Davis, J. L. (2014) ‘More than just “gay Indians”: intersecting articulations of two-spirit gender,
sexuality, and indigenousness’, in Zimman, L., Davis, J. L., and Raclaw, J. (eds) Queer excursions:
retheorizing binaries in language, gender, and sexuality. New York: Oxford University Press.

104
 Ethnography and shifting semiotics

Davis, J. L. (2018) Talking Indian: identity and language revitalization in the Chickasaw Renaissance.
Tucson: University of Arizona Press.
Davis, J. L. (2019) ‘Refusing (mis)recognition: navigating multiple marginalization in the U.S. two-
spirit movement’, Review of International American Studies, 12(1), pp. 65–86.
Deumert, A. and Mabandla, N. (2017) ‘A luta continua: black queer visibilities and philosophies of
hospitality in a South African rural town’, Journal of Sociolinguistics, 21(3), pp. 397–419.
Eckert, P. (1989) Jocks and burnouts: social categories and identity in the high school. New York:
Teachers College Press.
Eckert, P. (2000) Linguistic variation as social practice: the linguistic construction of identity in
Belten High. Malden, MA: Blackwell.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Think practically and look locally: language and gender as
community-based practice’, Annual Review of Anthropology, 21, pp. 461–490.
Epple, C. (1998) ‘Coming to terms with Navajo “nádleehí”: a critique of “berdache”, “gay”, “alternate
gender”, and “two-spirit”’, American Ethnologist, 25(2), pp. 267–290.
Gal, S. (1978) ‘Peasant men can’t get wives: language change and sex roles in a bilingual community’,
Language in Society, 7(1), pp. 1–16.
Gal, S. (1989) ‘Between speech and silence: the problematics of research on language and gender’,
IPrA Papers in Pragmatics, 3(1), pp. 1–38.
Gaudio, R. P. (2019) ‘Talk about intimate subjects: ethnographic approaches to language, gender, and
sexuality’, in Hall, K. and Barrett, R. (eds) The Oxford handbook of language and sexuality. New
York: Oxford University Press. (Published at Oxford Handbooks Online.)
Goldstein, D. M. (2017) ‘Fieldnote as political weapon: James Comey’s ethnographic turn?’
Member Voices, Fieldsights, June 22. https://culanth.org/fieldsights/fieldnote-as-political-
weapon-james-comeys-ethnographic-turn.
Goldstein, D. M. (2020) ‘Beyond thin description: biography, theory, ethnographic writing’, in
McGranahan, C. (ed) Writing anthropology: essays on craft and commitment. Durham, NC: Duke
University Press, pp. 78–82.
Goodwin, M. H. (1990) He-said-she-said: talk as social organization among black children.
Bloomington: Indiana University Press.
Hall, K. (1995) Hijra/hijrin: language and gender identity. Unpublished dissertation, University of
California, Berkeley.
Hall, K. (2003) ‘Exceptional speakers: contested and problematized gender identities’, in Meyerhoff,
M. and Holmes, J. (eds) Handbook of language and gender. Oxford: Basil Blackwell, pp. 353–380.
Hall, K. (2005) ‘Intertextual sexuality: parodies of class, identity, and desire in liminal Delhi’, Journal
of Linguistic Anthropology, 15(1), pp 125–144.
Hall, K. (2009) ‘“Boys’ talk”: Hindi, moustaches, and masculinity in New Delhi’, in Pichler, P. and
Eppler, E. (eds) Gender and spoken interaction. Houndmills: Palgrave Macmillan, pp. 139–162.
Hall, K. (2019) ‘Middle class timelines: ethnic jokes and sexual modernity in Delhi’, Language in
Society, 48(4), pp. 491–517.
Hall, K., Borba, R., and Hiramoto, M. (2021) ‘Language and gender’, in Stanlaw, J. M. (ed), The
international encyclopedia of linguistic anthropology. Hoboken, NJ: John Wiley & Sons, pp. 892–912.
Hall, K., Levon, E., and Milani, T. M. (2019) ‘Navigating normativities: gender and sexuality in text
and talk’, Language in Society, 48(4), pp. 481–489.
Hall, K. and O’Donovan, V. (1996) ‘Shifting gender positions among Hindi-speaking hijras’, in
Bergvall, V., Bing, J., and Freed, A. (eds) Rethinking language and gender research: theory and
practice. London: Longman, pp. 228–266.
Holmes, J. and Meyerhoff, M. (1999) ‘The community of practice: theories and methodologies in
language and gender research’, Language in Society, 28(2), pp. 173–183.
Ingold, T. (2014) ‘That’s enough about ethnography!’, Journal of Ethnographic Theory, 4(1), pp.
383–395.
Jacobs-Huey, L. (2002) ‘The natives are gazing and talking back: reviewing the problematics of
positionality, voice, and accountability among “native” anthropologists’, American Anthropologist,
104(3), pp. 791–804.

105
Kira Hall & Jenny L. Davis

Jacobs-Huey, L. (2006) From the kitchen to the parlor: language and becoming in African American
women’s hair care. New York: Oxford University Press.
Jespersen, O. (1922) Language: its nature, development, and origin. London: George Allen & Unwin.
Kaplan-Weinger, J. and Ullman, C. (2015) Methods for the ethnography of communication: language
in use in schools and communities. New York: Routledge.
Keenan (Ochs), E. (1974) ‘Norm-makers, norm-breakers: uses of speech by men and women in a
Malagasy community’, in Bauman, R. and Sherzer, J. (eds) Explorations in the ethnography of
speaking. London: Cambridge University Press, pp. 125–143.
King, B. W. (2017) ‘Querying heteronormativity among transnational Pasifika teenagers in New
Zealand: an oceanic approach to language and masculinity’, Journal of Sociolinguistics, 21(3),
pp. 442–464.
King, B. W. (2019) Language and embodied sexuality. In Hall, K. and Barrett, R. (eds) The Oxford
handbook of language and sexuality. New York: Oxford University Press. (Published at Oxford
Handbooks Online.)
Kramer, E. (2016) ‘Feminist linguistics and linguistic feminisms’, in Lewin, E. and Silverstein, L.
(eds) Mapping feminist anthropology in the twenty-first century. New Brunswick: Rutgers, pp.
65–83.
Lakoff, R. (1975) Language and woman’s place. New York: Harper & Row.
Lane, N. (2019). The Black Queer work of ratchet: race, gender, sexuality, and the (anti)politics of
respectability. Palgrave Macmillan.
Lave, J., and Wenger, E. (1991) Situated learning: legitimate peripheral participation. Cambridge:
Cambridge University Press.
Lazar, M. M. (2017) ‘Homonationalist discourse as a politics of pragmatic resistance in Singapore’s
Pink Dot movement: towards a southern praxis’, Journal of Sociolinguistics, 21(3), pp. 420–441.
Leap, W. L. (ed) (1995) Beyond the lavender lexicon. Amsterdam: Gordon & Breach.
Lemov, R. (2017) ‘Anthropology’s most documented man, ca. 1947: a prefiguration of big data from
the big social science era,’ Osiris, 32(1), pp. 21–42.
Levon, E. and Mendes, R. B. (2016) Language, sexuality, and power. New York: Oxford University
Press.
Livia, A. and Hall, K. (eds) (1997) Queerly phrased: language, gender, and sexuality. New York:
Oxford University Press.
McElhinny, B. (2003) ‘Theorizing gender in sociolinguistics and linguistic anthropology’, in
Meyerhoff, M. and Holmes, J. (eds) The handbook of language and gender. Oxford: Basil
Blackwell, pp. 21–42.
McGranahan, C. (2018) ‘Ethnography beyond method: the importance of an ethnographic sensibility’,
Sites: A Journal of Social Anthropology and Cultural Studies, 15(1), p. 2.
Mendoza-Denton, N. (2008) Homegirls: language and cultural practice among Latina youth gangs.
Hoboken: John Wiley & Sons.
Milani, T. M. and Lazar, M. M. (2017) ‘Seeing from the South: discourse, gender, and sexuality from
Southern perspectives’, Journal of Sociolinguistics, 21(3), pp. 307–319.
Morgan, M. (2002) Language, discourse, and power in African American culture. Cambridge:
Cambridge University Press.
Nader, L. (2011) ‘Ethnography as theory’, HAU: Journal of Ethnographic Theory, 1(1), pp. 211–219.
Ochs, E. (1979) ‘Transcription as theory’, Developmental Pragmatics, 10(1), pp. 43–72.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context: language
as an interactive phenomenon. Cambridge: Cambridge University Press, pp. 335–358.
Ostermann, A. C. (2017) ‘“No mam. You are heterosexual”: whose language? whose sexuality?’,
Journal of Sociolinguistics, 21(3), pp. 348–370.
Parish, A. and Hall, K. (2021) ‘Agency’, in Stanlaw, J. M. (ed) The international encyclopedia of
linguistic anthropology. Hoboken, NJ: John Wiley & Sons, 16–23.
Peck, A. and Stroud, C. (2015) ‘Skinscapes’, Linguistic Landscape, 1(1–2), pp. 133–151.

106
 Ethnography and shifting semiotics

Perrino, S. and Pritzker, S. (eds) (forthcoming) Research methods in linguistic anthropology. London:
Bloomsbury Academic.
Pichler, P. (2009) Talking young femininities. Basingstoke: Palgrave Macmillan.
Radin, J. (2017) ‘Digital natives: how medical and indigenous histories matter for big data’, Osiris,
32(1), pp. 43–64.
Shaikjee, M. and Stroud, C. (2017) ‘Fanon in drag: decoloniality in sociolinguistics?’, Journal of
Sociolinguistics, 21(3), pp. 371–396.
Shankar, S. (2008) ‘Speaking like a model minority: “FOB” styles, gender, and racial meanings among
Desi teens in Silicon Valley’, Journal of Linguistic Anthropology, 18(2), pp. 268–289.
Silverstein, M. (1985) ‘Language and the culture of gender: at the intersection of structure, usage, and
ideology’, in Mertz, E. and Parmentier, R. J. (eds) Semiotic mediation. Orlando, FL: Academic
Press, pp. 219–259.
Smalls, K. A. (2018) ‘Fighting words: antiblackness and discursive violence in an American high
school’, Journal of Linguistic Anthropology, 28(3), pp. 356–383.
Snell, J., Shaw, S., and Copland, F. (eds) (2015) Linguistic ethnography: interdisciplinary explorations.
Houndmills: Palgrave.
Woolard, K. (2008) ‘Why dat now?: linguistic-anthropological contributions to the explanation of
sociolinguistic icons and change’, Journal of Sociolinguistics, 12(4), pp. 432–452.
Wortham, S. (2006) Learning identity: the joint emergence of social identity and academic learning.
New York: Cambridge University Press.
Wortham, S. and Reyes, A. (2015) Discourse analysis beyond the speech event. London: Routledge.
Zentella, A. C. (1997) Growing up bilingual: Puerto Rican children in New York. Oxford: Blackwell
Basil.
Zimman, L. (2020) ‘Transgender language, transgender moment: toward a trans linguistics’, in Hall,
K. and Barrett, R. (eds) The Oxford handbook of language and sexuality. New York: Oxford
University Press. (Published at Oxford Handbooks Online.)
Zimman, L., Davis, J. L., and Raclaw, J. (eds) (2014) Queer excursions: retheorizing binaries in
language, gender, and sexuality. New York: Oxford University Press.
Zimman, L. and Hall, K. (2010) ‘Language, embodiment, and the “third sex”’, in Llamas, C. and
Watt, D. (eds) Language and identities. Edinburgh: Edinburgh University Press, pp. 166–178.

107
 7
Gender, language, and
elite ethnographies in UK
political institutions
Sylvia Shaw

Introduction
In this chapter I critically evaluate how linguistic research using mixed ethnographic and
discourse analytic methods can contribute to an understanding of the complex workings of
gender and power in parliamentary institutions. This feminist research has at its core the
problem of the continued and persistent underrepresentation of women in politics. Tackling
this problem involves recognising that both formal and informal institutional rules are not
gender-neutral and ‘have gendered consequences that can reinforce or challenge a status
quo in which men enjoy greater access to and power within the political arena’ (Franceschet
2017: 115).
My use of the term ‘elite ethnography’ in the title of this chapter reflects the interdiscipli-
nary orientation of the research primarily in sociolinguistics but also in political science and
its focus on political elites. Elites can be thought of as groups of people who have the most
influence, power, or skill in a given field, and in the case of politics this is a closed group of
elected members. As Rhodes et al. (2007: 2–5) note, the adoption of ethnography for study-
ing powerful political actors such as Members of Parliament (MPs) opens up the ‘black
boxes of elite behaviour’ and allows ‘studies of elite attitudes, and studies of elite decision-
making in context’ that have not traditionally been the preserve of quantitative methods
within political science (Gains 2011; Jourde 2009). My use of the term does not rule out the
existence of other types of elites (from areas as diverse as sporting, intellectual, and military
elites) but is adopted here to encapsulate the characteristics of the parliamentary contexts in
which the linguistic ethnography is undertaken.
The strength of qualitative, ethnographic work in political science aiming to uncover
such informal rules and traditional gender norms lies in its ability to identify their impact
on wider political innovations such as gender quotas or women-friendly policies (Gains
and Lowndes 2014: 530). Here I explain how a specific focus on linguistic rules, prac-
tices, and norms can complement these approaches by ‘helping researchers with a range
of different backgrounds to reach deeper into the ethnographic description of institutional
processes’ (Hymes 1996: 8). In doing so, I draw upon two sociolinguistic research projects
in which over 60 ethnographic interviews were conducted with men and women politicians

108
 Gender, language, and elite ethnographies

in order to investigate gender and linguistic participation in different UK parliamentary

institutions:1 the House of Commons (HoC); the Northern Ireland Assembly (NIA), the
National Assembly for Wales (NAW) and the Scottish Parliament (SP). As the focus of this
chapter is on methodological concerns, the presentation of the research projects is necessar-
ily selective and partial, and fuller accounts are given elsewhere (Shaw 2000, 2006, 2011,
2013, 2020).
In the following sections I first explain my use of linguistic ethnography as part of a
‘mixed-method’ approach to investigating gender and linguistic participation in parlia-
ments. Then I reflect on some of the challenges and dilemmas faced by the researcher when
‘researching up’ or interviewing people of a higher professional status than oneself, such as
members of the political elites. Using discourse analytic and ethnographic data, I then show
how these mixed qualitative methods can be used to enrich an understanding of gender and
language, both in the minutiae of debate discourse and in exposing wider cultural norms
and attitudes to ‘produce a fuller picture of a social world’ (Dörnyei 2007: 164) through the
‘triangulation’ of findings from different types of data.

Linguistic ethnography in parliamentary institutions

Analyses of the linguistic participation of women in traditionally male-dominated forums,
such as the UK Parliament and the Church of England, has found that women’s public
rhetoric is likely to be fractured by competing, often contradictory, norms and expectations
(Walsh 2001). Further, it has been claimed that ‘institutions are organised to define, demon-
strate and enforce the legitimacy and authority of linguistic strategies used by one gender
– or men of one class or ethnic group – whilst denying the power of others’ (Gal 1991: 188).
The reasons for this are complex and related in part to the ‘fraternal networks’ (Walsh 2001)
that exist in parliamentary institutions and also the ‘visibility’ of women in a traditionally
male-dominated forum (Bourdieu 1991). Traditional parliaments can therefore be viewed
as a ‘linguistic habitus’ in which ‘silence or hyper-controlled language’ is imposed on some
people, while others are allowed the ‘liberties of a language that is securely established’
(Bourdieu 1991: 82). In my view, parliaments are ‘gendered spaces’ in which the setting and
the communicative tasks together become an index of a gendered style (Freed 1996: 67).
Given these observations, one aim of the research was to ‘measure’ the linguistic contri-
butions of men and women MPs in formal parliamentary proceedings to establish whether
the habitus silenced women and privileged men. A second aim of the research was to inves-
tigate the wider cultural linguistic norms of the institutions in order to examine the ways
in which legitimacy and authority related to gender. Gender is viewed as a variable and
contested concept, being both a flexible category in which gender is a ‘doing, an incessant
activity performed’ (Butler 2004: 1), and a category which is partly fixed by the institutional
arrangements based on stereotypical notions of male and female linguistic behaviour.
The decision to use a mixed-method approach reflected a desire to capture and examine
these institutional arrangements in all their complexity, aiming to: ‘describe the some-
times chaotic, contradictory, polymorph of human behaviour’ (Blommaert 2007: 682).
Ethnography was selected as a method in order to uncover the informal institutional rules
in parliaments that: ‘are often taken for granted – usually submerged and barely visible
– and are therefore difficult to study’ (Chappell and McKay 2017: 24, emphasis mine).
Ethnography is suitable for this purpose as an interpretative approach that studies the
actions of participants from their own point of view and considers how these interac-
tions are embedded in wider social structures (Copland and Creese 2015). This approach

109
Sylvia Shaw

holds that the contexts for communication should be investigated rather than assumed and
that the detailed analysis of linguistic data is essential to understanding its significance
(Rampton 2007).
The ethnographic data consisted of archival research (standing orders, the Official
Reports, and parliamentary publications and documents); and visits to the research sites
where observation of debates took place from the public galleries in the different chambers.
This allowed a full view of the debating chambers, rather than the restricted view offered by
video recordings. A view from the public galleries allows the observer to hear more of the
interaction and get a better sense of the rapport or confrontation between speakers to gain a
sense of how the chamber operates. For example, in a debate in 2010 at the NIA at Stormont,
it was possible to see Peter Robinson (the First Minister) sending the then DUP Minister
Arlene Foster out of the chamber to fetch him a glass of water – rather than any of the other
male front-bench Ministers sitting nearer to him. This could indicate that traditional ideas
about women’s domestic roles permeated the DUP. Furthermore, the visits also allowed
observation outside the debating chambers, which gave a broader impression of the culture
of the institutions. For example, political divisions in the NIA at Stormont are also reflected
in the seating arrangements in the Assembly’s refectory. It operates under an informal but
strict layout according to political affiliation and role so that members of one party sit in
a ‘designated’ area, as do the press and domestic assembly staff. So, the ethnographer can
observe that informal segregation together with formal power-sharing appear to perme-
ate all institutional arrangements of the NIA. These observations took place at the same
time as undertaking ethnographic interviews with participants. As I explain in more detail
below, this process allowed for the observation of the ‘frontstage’ performances of debates
in the chamber and gave the opportunity of asking interviewees ‘backstage’ (Goffman 1959;
Wodak 2009) about their reflections on their own and other MPs’ behaviour in the debate
‘performances’ that had just taken place.
Alongside the ethnographic research, a close analysis of the debate floor was under-
taken to investigate the relationship between gender, power, and linguistic participation,
assuming power in these contexts is ‘power as territory: gaining access to discursive space’
(Thornborrow 2002: 27), where ‘power can be construed as one participant’s ability to affect
what the next participant does in the next turn’ (2002: 136). Applied conversation analysis
was used for its formal tools (in the form of units and models of interaction) for analysis,
particularly in relation to turn-taking in spoken interaction (Sacks et al. 1974) rather than
for its strict theoretical stance in its ‘pure’ form that gender can be enacted only by direct
indexicality when speakers directly orient to the category. This analytical approach is par-
ticularly compatible with ethnography (Copland and Creese 2015: 51) because the emphasis
of the analysis is not solely on how the participants obey relevant rules but also ‘on how
they jointly construct the conversation and their shared understanding of what is happening
within it’ (Richards 2006: 13, cited in Copland and Creese 2015: 51).
A distinguishing feature of traditional ethnographic research is its use of observation as
‘the principle source of knowledge about social phenomena’ (Gobo 2008: 190). However,
much linguistic ethnographic research necessarily adopts an ‘ethnographic perspective’
(Green and Bloome 1997: 6) rather than a traditional ethnographic approach (Copland and
Creese 2015: 30). Research with an ‘ethnographic perspective’ recognises that not all the
aspects of the research questions will emerge through observation alone, making ‘formal
interviews valuable data sources’ (Copland and Creese 2015: 30). In my own research,
I adopted an ethnographic perspective where formal, semi-structured interviews supple-
mented the observational and archival data. The reasons for this were that I was seeking

110
 Gender, language, and elite ethnographies

insights about politicians’ own experience of communicative norms in relation to interpreta-

tions I was making about discourse analytic data, and also restricted access to the research
sites (see discussion below) meant that time spent ‘in the field’ to carry out observations
was limited.

Conducting ethnographic interviews with political elites

The ethnographic researcher of political elites needs to recognise the participants as politi-
cal actors who, as interviewees, often play out their professional political roles according
to party political and ideological allegiances. These actors are, on the whole, expert inter-
viewees with years of experience in answering and managing different types of interview
contexts. As Williams notes of the experience of interviewing MPs for research purposes:
‘habits bred in their daily conversations with constituents, journalists or lobbyists seem to
persist in these quite different circumstances’ (1980: 310). This has led to the observation
that ‘political interviews are themselves highly political’ and that politicians as interviewees
can shape the interview in different types of dominant and resistant behaviour that have
been noted to ‘range from monologues of speech, highly defensive off-hand behaviour, to a
delivery of pre-scripted official speech’ (Puwar 1997: 1.1).
The power wielded by interviewees as members of a political elite means that the
researcher is ‘researching up’ from a position of less power and status than that of the
interviewee. This power relationship – in common with other workplace contexts – has
various ramifications for the research process, including the problem of access to interview-
ees and the research sites. These difficulties range from obtaining an interview in the first
place, to the last-minute cancellation and rescheduling of interviews, particularly when they
take place in the bustle of parliamentary business as MPs can be called away to vote or to
respond to unfolding political events. Time constraints mean that the lengths of interviews
are largely unable to be controlled by the researcher. As has been noted by many scholars,
traditional ethnography has, at its core, a commitment to long-term participant observation
‘lurking and soaking’ or immersion within the research site for long periods (Shaw et al.
2015: 7). In relation to parliamentary contexts, this type of access is unrealistic for most
researchers and only the privilege of a very small number of ethnographers who are aligned
with the institutions through disciplinary or personal contacts with MPs or civil servants
(Puwar 1997: 5.4).
The power relationship between the researcher and the researched also calls into question
some of the tenets of feminist research that assume the researcher is at a position of power
over the interviewee and it is therefore incumbent upon them to create a friendly, non-hier-
archical, and emancipatory environment (Gobo 2008: 58). One of the dilemmas posed by
‘researching up’ for feminist researchers is that ‘the emphasis on power-sharing and the vul-
nerability of the researched (…) may not be transferable, indeed may be counter-productive
to the development of feminist theory and practice in research with the “powerful”’ (Luff
1999: 692). My interviews with men and women politicians showed the difficulty inherent
in making any global claim about the power of individuals in relation to the interviewer
according to their professional role. I experienced the highly defensive and off-hand treat-
ment described above from some male and female interviewees and genuinely collabora-
tive, friendly, and constructive ‘treatment’ from others. Some MPs were self-conscious and
cautious in expressing their opinions about speaking in debates to me (as an academic and
linguist), while others held forth with confidence on the intricacies of the conventions of the
institutions. However, given a central claim of the research – that there are ‘institutionalised

111
Sylvia Shaw

advantages and disadvantages’ (Duerst-Lahti and Kelly 1995: 44) for men and women in
political institutions, gender (alongside other social categories such as race, and social class)
can be viewed as a contributory element to the power relations of all the ethnographic
interviews that were conducted. Therefore, I agree with Luff (1999) that interviews with
both men and women MPs have ‘moments of rapport’ involving both understanding and
disjuncture based on ‘a recognition of the fractured and often contradictory subjectivities of
researcher and researched’ (Luff 1999: 687).
Self-disclosure and reflexivity are also established principles of ethnographic inter-
viewing (Kezar 2003: Luff 1999). Reflexivity is about sharing one’s own perspective as a
researcher and is viewed as an ideal rather than a goal as ‘one can never fully know oneself’
(Kezar 2003: 401). Reflexivity refers to the aim of challenging one’s own assumptions and
how they affect the research project, and about sharing some of these reflections with the
person interviewed (Hertz 1997). These relational aspects of interviewing are all associated
with power: ‘Power impacts the nature of the interchange, who guides the process, whose
values shape the interview context, the ability to interpret and make sense of the responses,
and how the data are ultimately used’ (Kezar 2003: 401). When interviewing less powerful
individuals these principles are straightforwardly strategies to equal power relations (albeit
with the complex and often contradictory dimensions of power relations noted above).
However, it is questionable as to whether self-disclosure and reflexivity help to level power
relations when the interviewer is ‘researching up’. As Kezar notes: ‘Can self-disclosure
actually heighten power differentials and provide more ways for the elite to retaliate if they
move in that direction?’ (2003: 406). Limited access also plays a part here as ideally femi-
nist research in an ethnographic context would build trust through empathy, mutuality, and
shared understanding (Wolfe 1996), and this would be achieved through ‘a feminist ethic
of commitment in contrast with the scientific ethic of detachment and role differentiation
between the subject and the researcher’ (Reinharz 1992: 27). This commitment would nor-
mally be expressed through multiple interviews to establish trust with the interviewee – an
objective that is wholly impractical given the time-constraints under which political elites
operate. This is compounded by the sheer number of interview requests received by MPs
– for example women MPs in the 1997 intake in the House of Commons reported being
overwhelmed by requests for interviews (Puwar 1997: 5.1).
Given the constraints of researching the powerful, in particular as manifested by the
problem of limited time in the research site and access to the participants, I found it neces-
sary to reveal little about my own views to the interviewees. On the contrary, I needed to
glean the interviewees’ interpretations of linguistic participation in parliamentary debate
(both generally and in relation to specific examples from the data corpus – see below) with-
out ‘leading’ them by sharing my own views and interpretations first. The limited time for
interviews also meant that exposing differing views from those of the interviewees would
run the risk of spending all the time exploring those differences of opinion rather than focus-
ing on MPs’ knowledge of specific informal ‘rules of speaking’ or wider cultural norms
within the institutions. Another risk of exposing differences of opinion while ‘researching
up’ is that it could jeopardise a researcher’s ‘access to elites in the future’ (Punch 1986) in
the ongoing research process. Reflexivity, however, remains a core concern of feminist and
ethnographic methodologies. While the aim of the complete openness and self-disclosure of
feminist ethnographic work can be seen as an ideal that is contingent on the particularities
of the research context, a commitment to a reflexive approach ‘means that feminist research-
ers should acknowledge their own orientations, bringing to an end claims that ‘objective’
knowledge has been produced with the researcher not influencing the research process in

112
 Gender, language, and elite ethnographies

any way’ (Mills and Mullany 2011: 119). This is particularly important given that ‘other fac-
tors intersecting with gender – such as nationality, race, ethnicity, class and age – also affect
the anthropologist’s field interactions and textual strategies’ (Callaway 1992: 33 cited in
Mills and Mullany 2011: 120). Reflexivity is therefore important at all stages of the research
process from the selection of particular methods, the questions asked and omitted ‘in the
field’, the interpretation of data and the highly selective process of presenting final accounts
of the research.

A mixed-method approach: researching the debate foor

In this section I show how the mixed-method approach, using linguistic ethnography com-
bined with discourse analytic techniques can be used to investigate gender and linguistic
participation in four political institutions. By mixed-method research I refer to the practice
of collecting multiple data types to address a particular research question. Triangulation is
the term that, in the social sciences, ‘is synonymous with combining data sources to study
the same social phenomena’ with the aim that ‘methodological triangulation can help to
reduce the inherent weaknesses of individual methods by offsetting them by the strength of
another’ (Dörnyei 2007: 43). I start by explaining how characteristics of the parliamentary
debate floor relate to these different approaches to the collection of data and then go on to
show how different types of data can illuminate the analysis and interpretation of a particu-
lar event.
The ‘ideal’ or canonical form of debate is ‘the most extreme transformation of conversation
– most extreme in fully fixing the most important (and perhaps nearly all) of the parameters
that conversation allows to vary’ (Sacks et al. 1974: 731). The ideal or ‘legal’ progression of
debates is formalised by explicit rules in the ‘standing orders’ of the parliament, governed by
the moderator (the Speaker or Presiding Officer). These rules embody democratic ideals in
that they ensure speaking turns are divided fairly between politicians from different political
parties, and that only one person speaks at a time in order that their speech can be heard. This
formal ordering of debates is emphasised by the use of strict address forms; having to direct
all speeches to the moderator (to avoid overt confrontation between political opponents); and
the physical standing position of the ‘legal’ speaker with all other interjections from politicians
in a sedentary position prohibited. Together, these formal orders and restrictions are devised
to ‘permit the equalisation of turns’ (Sacks et al. 1974: 730). However, the canonical or ‘legal’
progression of the debate exists alongside an unofficial or ‘illegal’ set of conventions that rou-
tinely operate on the debate floor, including the use of ‘illegal’ address forms and interventions
which vary from full interjections and questions to individual and collective jeering, shout-
ing, and heckling. This means that ‘thus an event that should allow everyone an equal chance
becomes an event in which prior inequalities (e.g. gender, age and ethnicity) can be re-enacted’
(Edelsky and Adams 1990: 171).2
Both the ‘legal’ and ‘illegal’ debate floors need to be accounted for in an assessment of
the linguistic participation of a parliament’s members. The legal floor is particularly acces-
sible as it is routinely and systematically documented in the ‘Official Report’ (the ‘Hansard
Report’ in the HoC). Although this is commonly referred to as a ‘verbatim’ report it actually
represents a selective, orderly, written version of the spoken parliamentary discourse. Illegal
interventions and overtly ‘spoken’ features of the discourse are omitted, address forms are
routinely standardised, and various grammatical adjustments are made to standardise the
official contributions of the participants (see Shaw 2018 for a detailed account of these
editorial changes and omissions). The textual practices that create these documents, as well

113
Sylvia Shaw

as the selectivity of the documents themselves therefore offer valuable ethnographic infor-
mation about the institution’s values, cultural practices, and exclusionary practices. The
Official Reports also give researchers a valuable resource for examining some aspects of
parliamentary participation – they are electronically available and allow the quantification
of some of the different types of speaking turns (speeches, questions, responses, give way
requests, give way interventions, give way refusals, ‘points of order’, and so on). For the
research projects in the different parliaments a quantitative assessment of debate turns found
that in all the institutions apart from the NAW, men and women took legal turns in propor-
tion to their overall representation in the institutions (Shaw 2020).
However, illegal interventions are not systematically represented in the Official Reports
as an unattributed ‘interruption’ is only shown when it gains a response from a ‘legal’
speaker (so that the official discourse remains coherent in the report). Therefore, in order to
assess the participation of MPs in the ‘illegal’ floor, qualitative methods were used. Detailed
transcripts from video recordings of debates were made of the proceedings. An example of
this type of data can be seen in Example 1 below, showing sustained illegal interventions
made by a female Assembly Member in the National Assembly for Wales. The occasion
of this extract was the statement of the Secretary of State for Wales (Cheryl Gillan) on the
Queen’s Speech, and the subsequent debate on the same topic.

Example 1 Sustained illegal interventions in the National Assembly for Wales (16th
June 2010)
CG = Cheryl Gillan; FL = Female Labour Assembly Member; DPO = Deputy Presiding
Officer AM = An Assembly Member

1 CG: I think that the single new back-to-work programme

2 will provide numerous new opportunities for national and
3 local partners (.) to contribute to the coalition
4 Government’s programme (.) to tackle er unemployment (.)
5 FL: well you haven’t convinced me
6 CG: Um(.) well I may not have convinced the Honourable Lady
7 as she says from a
8 sedentary position (.) [but I am] (.)
9 FL [I am not] an honourable lady
10 I am an [Assembly Member]
11 DPO: [um excuse me]
12 CG um well the the honourable Assembly Member (.)
13 FL no I am an
14 Assembly Member
15 CG an Assembly Member (.) an Assembly Member um and I’m
16 sorry I’m I’m not trying to convince her (.) I’m trying
17 to reassure the people that she
18 may have frightened outside this
19 CG cham[ber]
20 FL [no] I am not the one who is
21 frightening [them]
22 CG [that ] we [are looking] (.)
23 DPO [Ann Jones] (.) please
24 CG that we are looking at this area and we

114
 Gender, language, and elite ethnographies

25 [will approach it with compassion] (.)

26 FL [unclear ]
27 CG in closing (.)
28 AM hear hear
29 CG Madam [Presiding Officer Deputy] Presiding Officer I
30 FL [unclear ]
31 CG reiterate how impressed I have been in my first few weeks
32 (turn continues)
(Shaw 2011: 284–285)

Example 1 can be used to illustrate how the ethnographic ‘mixed-method’ qualitative

approach can be used ‘to achieve a fuller understanding of the target phenomena’ (Dörnyei
2007: 164), in this case, illegal utterances, rule-breaking, and the transgression of formal
rules. On this occasion I was present in the debating chamber and made field notes to record
my observations. After the statement and debate I carried out three ethnographic interviews
with Assembly Members (AMs) in the NAW and was able to ask them directly about their
impression of this event (these interviews were subsequently transcribed for analysis). The
speech event itself was video recorded and transcribed as shown in the example and was
therefore available for the analysis of turn-taking and floor apportionment.
Taking each of these qualitative types of data in turn (transcripts of the debate, observa-
tional field notes, and ethnographic interviews) and looking first to the discourse analytic
observations of this extract (a full account of which is given in Shaw 2011), it can be noted that
the female AM initially interrupts the Secretary of State to contest the remark she has made
about the back-to-work programme, using the discourse marker ‘well’ and then saying ‘you
haven’t convinced me’ (line 5). Gillan responds to this illegal intervention and refers to the
interrupting AM as ‘the Honourable Lady’ (line 6), highlighting the AMs illegal intervention
from a ‘sedentary position’. The AM makes another illegal intervention to correct Gillan by
saying ‘I am not an Honourable Lady I am an Assembly Member’ (lines 9–10). Line 12 shows
an example of other initiated self-repair (Schegloff, Jefferson, and Sacks 1977) when Gillan
says ‘The Honourable Assembly Member’ (line 12). The AM interrupts once more to correct
her by saying ‘No, I am an Assembly Member’, to which Gillan says ‘An Assembly member,
an Assembly member’ (line 15) in a further example of other initiated self-repair. Gillan then
accuses the interrupting AM of ‘frightening people’ outside the chamber, which gives rise to
another illegal intervention (line 20–21). The Deputy Presiding officer intervenes but the AM
continues to disrupt the debate floor (lines 26 and 30). Finally, Gillan attempts to control the
floor by a change in conversational alignment or ‘footing’ (Goffman 1981) in which she clearly
signals the final part of her speech by saying ‘in closing’ (line 27). This change in footing is
recognised by a Conservative AM supporter who shouts ‘hear hear’, although the interrupting
female AM illegally intervenes once more while Gillan introduces her closing remarks.
The second source of data is taken from my observational field notes, from which the
following indicative description can be drawn:

On the day the event took place, the viewing gallery was much more crowded than nor-
mal. Cheryl Gillan looked ill at ease as she entered the chamber and my first impression
was that she was struggling to use the headphones that offer English– Welsh transla-
tion. I noted that failing to use this equipment – in itself a symbol of Wales’ distinctive
politics and commitment to inclusivity and openness – would not endear her to the
AMs. I was surprised at her lack of awareness of the procedures of the Assembly and

115
Sylvia Shaw

startled by the extent of the barracking and vehement interventions against her, as they
were much stronger than at any time I had attended the chamber before. I was struck in
particular by her inability to use the correct address forms and noted that she seemed to
have expected the Welsh Assembly to operate in the same way as Westminster. The sus-
tained interventions by the female Labour MP were greeted with a mixture of embar-
rassment and support from fellow AMs, with the Presiding Officer trying a number of
times to call her to order.

These types of ethnographic notes give detailed impressions of the event itself and can form
the basis of subsequent interview questions and discussion points. It is also significant that
while an observer cannot strictly be said to be a ‘participant’ in these events from the gal-
lery, being present in the chamber creates shared understandings that can be used to create
‘moments of rapport’ with subsequent interviewees.
Third, taking the interview data into account (described fully in Shaw 2011) can give the
following insights into the AMs attitudes towards the event, and wider cultural norms in the
Assembly. Three women AM interviewees used this speech as a point of reference when
discussing interactional norms in the debating chamber, and gave their reaction to the per-
formance of Gillan and other AMs taking part in the debate. Although a political opponent
of Gillan, one female AM felt that the heckling of Gillan in the debate was ‘totally pointless
bluster’ and reported being ‘a bit unhappy with the tone of some of the oppositional stuff’
because ‘whatever we feel we have to work with this person, we may not like what they
stand for but we’re not going to progress any of the agendas if you attack someone on a
personal level’. However, all three interviewees were annoyed by Gillan’s failure to adopt
the procedural norms of the assembly. The same female AM observes:

I think we were all irritated with her and she was slightly condescending I think the way
she spoke to us on a few occasions: ‘I’d like to say to the Honourable Lady’. Haven’t
you done your homework you’ve been Shadow Secretary of State for ‘x’ number of
years you know we’re not Honourable Members here and you know no-one has ever
called us that.

A female Labour AM noted that it was not just Gillan’s failure to adopt the correct forms of
address, but a more general lack of preparation and familiarity with procedures that gave a
negative impression:

She was working from notes quite a lot and saying ‘oh sorry hang on I’ve lost my notes’
I thought aw come on you know I wouldn’t want to be Secretary of State for Wales but
she’s chosen to be she wanted to be so yeah she is condescending and patronising. It’s
the way she spoke it was like ‘oh so lovely to be here’.

Gillan’s lack of proficiency in adopting the Assembly’s norms contributes to her negative
assessment by interviewees and emphasises her ‘outsider’ status. It is also clear that gender
(and social class and national identity) plays a central role in the AMs’ assessment of her
performance. One female Labour AM says:

I don’t know but her whole demeanour I suppose was and this may be unfair just because
she’s a woman but her whole demeanour I think was quite sort of Queen Mother-like
you know ‘I’m here aren’t you lucky that I’m here I’ve come to talk to you peasants’.

116
 Gender, language, and elite ethnographies

Here the AM recognises that her negative appraisal of Gillan may be attributable to gender
and reveals the strength of the discourse of gender difference in her appraisal. This dis-
course of gender difference can be seen as a ‘significant “lens” for the way people view
reality, difference being for most people what gender is all about’ (Sunderland 2004: 52).
The description of Gillan as ‘queen mother-like’ is interesting because it refers to British
monarchy as a distinct marker of national identity and at the same time it orients to the role
of ‘mother’, which has been claimed to be one of the very limited roles available to women
leaders (Kanter 1977). It is also significant that the women AMs view the female Minister of
State negatively and this example shows traces of a gendered discourse of ‘women beware
women’ (Sunderland 2004).
These examples illustrate the way in which mixed-method research can contribute to
fuller understanding of an event and enrich a description of the wider norms of the institu-
tions. In this example, women are not silenced, but on the contrary dominate this interaction
yet orient to gendered discourses of differentiation in their interviews, showing that political
institutions are ‘historically variable in their composition and effects’ (Krook and Mackay
2011: 3). The methods thus allow assessments of ‘illegal’ rule-breaking by female AMs
on the debate floor, together with evidence of gendered discourses and assumptions about
male and female linguistic behaviour by the participants themselves. A further benefit of
triangulation in relation to investigating linguistic norms is when the data show divergent
findings or interpretations, which can lead to an enhanced understanding by pointing to
areas for further investigation (Dörnyei 2007: 165). This is certainly important when the
interpretations of discourse analytic observations by the researcher are confounded by the
participant’s own interpretations. This occurred on several occasions in the course of these
research projects. For example, when initially investigating turn-taking in the HoC, I inter-
preted legal ‘give way’ interventions from political opponents as violative interruptions that
were intended to wrest the floor from the legal speaker, and present interactive and substan-
tive challenges to them. However, interviews with MPs revealed that ‘give way’ interven-
tions (even from political opponents) were viewed positively by many MPs as an indicator
of respect and interest in the speech they were delivering – one MP even commenting that
‘at least it shows that everyone isn’t asleep’. These divergent interpretations of events and
norms therefore show that multiple data sources are important in these ‘closed’ parliamen-
tary institutions where it is not possible for the researcher as ‘outsider’ to reliably interpret
interactive phenomena.

Concluding discussion and future directions

This chapter has given a critical feminist account of some of the methodological con-
siderations associated with investigating linguistic participation in elite political insti-
tutions. The results gained from the different data sources continue to raise interesting
questions about mixing qualitative methodologies and contribute to discussions about
both the descriptive (numerical) and substantive representation of women in political
institutions.
It is difficult to assess the transformational effects of asking political questions about
gender, language, and power in elite institutions, but it is possible to underestimate or ignore
this aspect of research. It is important to recognise that taking our research questions and
research agendas into powerful political institutions can have effects on powerful actors. I
do not agree with Wax’s assessment of the role of the elite interviewer that: ‘the fieldworker
is naïve if he (sic) thinks that most of these important personages will really listen carefully

117
Sylvia Shaw

to what he says, much less, believe it’ (Wax 1971 cited in Kezar 2003: 398). My experience,
informed by the different perspectives offered by the triangulation of data, showed that, for
example, after asking one of the institution’s moderators (who wished to remain anonymous)
about how s/he brought politicians to order on the debate floor, his/her subsequent behaviour
in the debating chamber was affected. Three of the four politicians interviewed after the
debate all remarked (without being prompted) on how the moderator’s behaviour in keeping
order in the chamber had been stricter than usual. It is therefore possible that there are effects
on political actors brought about by questions that highlight the micro-details of interaction
and connect these to macro-questions of exclusion and power. This aspect of the research
is extremely pertinent for investigations into language, gender, and sexuality in institutional
contexts, and certainly warrants further investigation. It would also be interesting to extend
this mixed-method research to wider political contexts both within parliament (for example,
in committees), and outside parliament (for example at political rallies, conferences, and
meetings). In this way, it would illustrate how research into language, gender, and sexuality
can influence actors outside the academy and have relevance to the wider world.

Transcription conventions
(.)= micropause of less than a second
[ = beginning of overlap with the utterance above or below
] = end of overlap with the utterance above or below
underline = particular emphasis on word or syllable

Notes
1 The research projects were first conducted in the House of Commons and the data was collected
between 1998 and 2000. Second, the Economic and Social Research Council funded project ‘Gender
and linguistic participation in the devolved parliaments of the UK’ (RES 000223792) collected data
between 2009 and 2010 in the devolved political institutions of the UK and in the HoC.
2 See Shaw 2000 for a fuller discussion of the parameters of formal debates.

Further reading
Snell, J., Shaw, S. E., and Copland, F. (eds) (2015) Linguistic ethnography: interdisciplinary
explorations. Basingstoke: Palgrave Macmillan.
This book demonstrates the scope and breadth of research in linguistic ethnography with useful
introductory chapters on practical and theoretical aspects.
Waylen, G. (ed) (2017) Gender and informal institutions. London: Rowman & Littlefield.
This book represents a feminist institutionalist perspective in political science, giving analytical
frameworks for approaching different types of informal institutions
Wodak, R. (2009) The discourse of politics in action: politics as usual. Basingstoke: Palgrave
Macmillan.
A critical ethnography of the European Parliament, this book aims to go ‘behind the scenes’ of the
institution to uncover the reality of daily political discourse.

Related topics
Anthropological discourse analysis and the social ordering of gender ideology; gender in interaction:
ethnomethodological and ca approaches to gender; feminist conversation analysis: examining violence

118
 Gender, language, and elite ethnographies

against women; identity construction in gendered workplaces; ethnography and the shifting semiotics
of gender and sexuality.

References
Blommaert, J. (2007) ‘On scope and depth in linguistic ethnography’, Journal of Sociolinguistics,
11(5), pp. 682–688.
Bourdieu, P. (1991) Language and symbolic power. Cambridge: Polity Press.
Butler, J. (2004) Undoing gender. New York: Routledge.
Callaway, H. (1992) ‘Ethnography and experience; gender implications in fieldworks and texts’, in
Okely, J. and Callaway, H. (eds) Anthropology and autobiography. New York: Routledge, pp.
60–85.
Chappell, L. and Mackay, F. (2017) ‘What’s in a name? Mapping the terrain of informal institutions
and gender politics’, in Waylen, G. (ed) Gender and Informal Institutions. Lanham, MD: Rowman
& Littlefield, pp. 23–44.
Copland, F. and Creese, A. (2015) Linguistic ethnography: collecting, analysing and presenting data.
London: SAGE.
Dörnyei, Z. (2007) Research methods in applied linguistics: quantitative, qualitative, and mixed
methodologies. Oxford: Oxford University Press.
Duerst-Lahti, G. and Kelly, R. M. (1995) Gender power, leadership and governance. Ann Arbor, MI:
University of Michigan Press.
Edelsky, C. and Adams, K. (1990) ‘Creating inequality: breaking the rules in debates’, Journal of
Language and Social Psychology, 9(3), pp. 171–190.
Franceschet, S. (2017) ‘Disentangling informality and informal rules: explaining gender inequality in
Chile’s executive branch’, in Waylen, G. (ed) Gender and informal institutions. London: Rowman
& Littlefield, pp. 115–136.
Freed, A. F. (1996) ‘Language and gender in an experimental setting’, in Bergvall, V. and Bing, J. M.
(eds) Rethinking language and gender research: theory practice. New York: Longman, pp. 55–76.
Gains, F. (2011) ‘Elite ethnographies: potential, pitfalls and prospects for getting ‘up close and
personal’, Public Administration, 89(1), pp. 156–166.
Gains, F. and Lowndes, V. (2014) ‘How is institutional formation gendered, and does it make a
difference?’, Politics & Gender, 10(04), pp. 524–548.
Gal, S. (1991) ‘Between speech and silence: the problematics of research on language and gender,’
in di Leonardo, M. (ed) Gender at the crossroads of knowledge: feminist anthropology in the
postmodern era. Berkeley: University of California Press, pp. 175–203.
Gobo, G. (2008) Doing ethnography. London: SAGE.
Goffman, E. (1981) Forms of talk. Philadelphia: University of Pennsylvania Press.
Goffman, E. (1990 [1959]) The presentation of self in everyday life. London: Penguin.
Green, J. and Bloome, D. (1997) ‘Ethnography and ethnographers of and in education: a situated
perspective’, in Flood, J., Heath, S., and Lapp, D., (eds) A handbook for literacy educators. New
York: Macmillan, pp. 181–202.
Hertz, R. (1997) Reflexivity and voice. Thousand Oaks, CA: SAGE.
Hymes, D. H. (1996) Ethnography, linguistics, narrative inequality. London: Taylor & Francis.
Jourde, C. (2009) ‘The ethnographic sensibility: overlooked authoritarian dynamics and Islamic
ambivalences in West Africa’, in Schatz, E. (ed) Political ethnography: what immersion contributes
to the study of power. Chicago: University of Chicago Press, pp. 201–216.
Kanter, R. M. (1977) Men and women of the corporation. New York: Basic Books.
Kezar, A. (2003) ‘Transformational elite interviews: principles and problems’, Qualitative Inquiry,
9(3), pp. 395–415.
Krook, M. L. and Mackay, F. (eds) (2011) Gender, politics and institutions. Basingstoke: Palgrave.
Luff, D. (1999) ‘Dialogue across the divides: “moments of rapport” and power in feminist research
with anti-feminist women’, Sociology, 33(4), pp. 687–703.

119
Sylvia Shaw

Mills, S. and Mullany, L. (2011) Language, gender and feminism: theory, methodology and practice.
London: Routledge.
Punch, M. (1986) The politics and ethics of fieldwork. Thousand Oaks, CA: Sage.
Puwar, N. (1997) Reflections on interviewing women MPs. Available at: http://socresonline.org.uk/2/1
/4.html [Accessed 3rd August 2017].
Rampton, B. (2007) ‘Neo-hymsian linguistic ethnography in the United Kingdom’, Journal of
Sociolinguistics, 11(5), pp. 548–607.
Reinharz, S. (1992) Feminist methods in social research. Oxford: Oxford University Press.
Rhodes, R., Hart, P. ’t, and Noordegraaf, M. (eds) (2007) Observing government elites: up close and
personal. Basingstoke: Palgrave Macmillan.
Richards, K. (2006) Language and professional identity: aspects of collaborative interaction.
Basingstoke: Palgrave Macmillan.
Sacks, H., Schegloff, E. A., and Jefferson, G. (1974) ‘A simplest systematics for the organization of
turn-taking for conversation’, Language, 50(4), pp. 696–735.
Schegloff, E. A., Jefferson, G., and Sacks, H. (1977) ‘The preference for self-correction in the
organisation of repair in conversation’, Language, 53, pp. 361–382.
Shaw, S. (2000) ‘Language, gender and floor apportionment in political debates’, Discourse and
Society, 11(3), pp. 401–418.
Shaw, S (2006) ‘Governed by the rules? The female voice in parliamentary debates’, in Baxter, J. (ed)
Speaking out: the female voice in public contexts. Basingstoke: Palgrave Macmillan, pp. 81–102.
Shaw, S (2011) ‘I am not an honourable lady!’: gender and language in the national assembly for
wales. Journal of Applied Linguistics and Professional Practice, 8(3), pp. 275–294.
Shaw, S (2013) ‘An ethnographic investigation into gender and language in the Northern Ireland
assembly’, in Poggi, I., D’Errico, F., Vincze, L., and Vinciarelli, A., (eds) Multimodal communication
in political speech: shaping minds and social action. Berlin: Springer, pp. 39–53.
Shaw, S. (2018) ‘Off the record: the transcription of parliamentary debates for political discourse
analysis’, in Kranert, M. and Horan, G. (eds) Doing politics: discursivity, performativity and
mediation in political discourse. Amsterdam: John Benjamins, pp. 105–126.
Shaw, S. (2020) Women, language and politics. Cambridge: Cambridge University Press.
Shaw, S. E., Copland, F., and Snell, J. (2015) ‘An introduction to linguistic ethnography’, in Linguistic
ethnography: interdisciplinary explorations. Basingstoke: Palgrave Macmillan, pp. 1–13.
Sunderland, J. (2004) Gendered discourses. Basingstoke: Palgrave Macmillan.
Thornborrow, J. (2002) Power talk: interaction in institutional settings. Harlow: Longman.
Walsh, C. (2001) Gender and discourse: language and power in politics, the church and organisations.
Harlow: Longman.
Wax, R. (1971) Doing fieldwork: warnings and advice. Chicago: University of Chicago Press.
Williams, P. M. (1980) ‘Interviewing politicians: the life of Hugh Gaitskell’, Political Quarterly,
51(3), pp. 303–316.
Wodak, R. (2009) The discourse of politics in action: politics as usual. Basingstoke: Palgrave
Macmillan.
Wolfe, D. (1996) Feminist dilemmas in fieldwork. Boulder, CO: Westview.

120
 8
‘Gay, aren’t they?’ An
ethnographic approach to
compulsory heterosexuality
Jodie Clark

Introduction
Ethnography holds a particular appeal for researchers who wish to resist generalisations
about gender and sexuality and focus instead on the local ways in which gender and sex-
uality affect people’s lives. Language, gender, and sexuality ethnographies have offered
insights into the unique gendered and sexed practices in schools (Bucholtz 1999; Eckert
2000; Moore 2006), sports teams (Clark 2012; Sauntson and Morrish 2012), LGBTQ+
groups (Jones 2012, 2016) and drag queen performances (Barrett 1999), to name just a few.
Ethnographic research can also offer a window into the sometimes small but always signifi-
cant ways in which people’s lives are adversely affected by constrictive gender and sexual-
ity norms. By focusing on individual experiences of gender and sexuality, ethnography can
posit an implicit counter-argument to claims that ‘everything is so much better than it used
to be’ in relation to sexism, homophobia, and transphobia.
One focus of ethnographic research is on the norms that participants in various commu-
nities and social spaces abide by. Sometimes these are explicit – they can be explained by
members of the community in interviews for instance – but they are more often under the
radar. Ethnographic methods are often used to identify the unwritten rules of interaction,
acceptance, and belonging. Language, gender, and sexuality ethnographies frequently focus
on the heteronormative frameworks that govern particular social spaces (Bucholtz 1999;
Clark 2012; Eckert 2011).
This chapter focuses on what Adrienne Rich (1980) calls ‘compulsory heterosexuality’
– a set of ideologies that make it very difficult, especially for women, to imagine a life free
from the power imbalances of heterosexual relationships. I draw upon two ethnographic
studies – one of a university women’s field hockey team and one of a university friendship
group – to explore instances in which compulsory heterosexuality effected both physical
and ‘symbolic violence’ (Bourdieu and Waquant 1992: 166). I focus on instances in which
participants were required by their peers to engage in heterosexual activity – specifically,
they were compelled or coerced to kiss men against their will.
Seidman (2009) argues in favour of making a clear distinction between heteronorma-
tivity (as discourses, norms, and social pressures) and compulsory heterosexuality (where

121
Jodie Clark

heterosexuality is an official requirement). From my point of view, to focus on this distinc-

tion is to miss an important perspective. Seidman’s account presents both terms, heteronor-
mativity and heterosexuality, as oppressive forces that impose themselves on individuals.
I propose instead an approach that draws upon ethnographic data to identify those fleeting
instances in which heterosexuality makes itself felt as compulsory, where the result of not
engaging in heterosexual acts results in violence to the self. The opportunity then presents
itself to explore the structure of the social space at that precise moment, and then to ask
questions about how it might otherwise have been structured.
The approach I describe here makes use of the analytic method described in Clark (2016),
which outlines a method for analysing ‘the grammatical structures of participants’ accounts of
their social worlds’ (2016: 9) as a means of imagining possibilities for social transformation.
The methodology supplements ethnography with the methods of critical discourse analysis
(CDA) described in Part V, ‘Sociocultural and critical approaches’, of this Handbook (for
other uses of CDA within gender and sexuality research, see Koller 2015; Morrish 1997 and
Peterson 2010). My approach differs from CDA in that it is designed to orient not to ideology
(claims to truth), but rather to discursively constructed selves and social worlds:

I propose a CDA that asks different questions. … Instead of, what are the claims to
truth, and wherein lies the struggle? I would ask, what is the shape of the social struc-
ture here, and where is the desire for an alternative? What are the ‘selves’ that are
textually constituted here and how might they be otherwise constituted? What are the
possibilities for transformation here?
(Clark 2016: 38)

When these questions are applied to compulsory heterosexuality, the orientation shifts from
understanding heteronormativity as an oppressive discourse (Motschenbacher and Stegu
2013), an inescapable ‘contract’ (Wittig 1992: 34), or a regulatory ‘matrix’ (Butler 2006:
208). Instead, instances in which heterosexuality makes itself felt as compulsory in participant
accounts can be used as windows into the shape of the social world at that precise moment. A
key focus in this chapter will be on the different ‘selves’ that are grammatically constructed in
these accounts. I will explore two different ways in which selves are presented as vulnerable
in the face of compulsory heterosexuality. My analysis demonstrates that grammatical con-
structions of the vulnerable self coincide with grammatical constructions of alternative social
worlds – worlds in which vulnerable selves are protected, and heterosexuality loses its force.
The method of analysing ethnographic data I demonstrate in this chapter is unique from
other approaches in that it does not focus on norms, discourses, belief systems, and ideolo-
gies that structure a particular social space or community. The analysis here is concentrated
on very specific instances within the context that the ethnographic fieldwork provided –
precise moments when compulsory heterosexuality is at issue in participants’ accounts. The
aim here is not to chart how heteronormativity governs a particular social group, but rather
to identify moments within social group interactions in which alternatives to heteronorma-
tivity present themselves.

The ethnographic feldwork

Another way in which the analysis I present in this chapter is different from other ethno-
graphic work is that it draws upon data collected from two different ethnographies, situated

122
 Ethnography of compulsory heterosexuality

within two different social groups. The first is a women’s university field hockey team
(Clark 2012) and the second a small group of students in a module I taught called ‘Language
and social life’ (Clark 2016). Comparing these communities enables me to explore two
very distinct moments of compulsory heterosexuality, to situate these in two different social
contexts and to identify the distinct ways in which the selves and the social worlds are con-
structed in each. Despite the participants’ distinct responses to these moments, it became
clear (as I will show) that the accounts of compulsory heterosexuality consistently reveal,
first, the grammatical construction of a vulnerable self and, second, the grammatical con-
struction of an alternative social world.
The members of the Midland University1 women’s field hockey team were straight cis
women, white British middle-class, between the ages of 18 and 21. They identified as high
achieving hockey players, students, and citizens. They made it clear that to achieve in all
these areas requires not only securing an identity as feminine and straight but also silenc-
ing any discussions of sexual desire (Clark 2012). In other words, their identities depended
upon conforming to heterosexual norms.
The participants in the second group were three students (Beth, Maryam, and Andrew)
in a module I was leading called ‘Language and social life’. Troubled by the ways in which
these three friends had been alienated by other students in the seminar group that I was lead-
ing, I asked them, at the end of the academic year, if they would be willing to talk about their
experiences of feeling marginalised as part of a focus group. During the focus group Beth,
Maryam, and Andrew revealed that what brought them together as friends was a shared
resistance to the norms of the other students on their degree programme. Each of them had
experienced a sense of lack of acceptance from others on their course, which they attributed
to a certain extent to aspects of their identities: for Beth, it was her unapologetic working-
class status; for Maryam, it was her ethnicity (Pakistani-British) and her religion (Muslim);
for Andrew, it was his disability (which affected his mobility and required him to sometimes
use a crutch) and his non-normative gender (transgendered man).
The similarities between the two groups (their ages, for instance, as well as their engage-
ment in university communities of practice (Eckert and McConnell-Ginet 1992)) serve as
a backdrop against which to highlight the points of contrast around compulsory hetero-
sexuality. The hockey players expressed explicit homophobic attitudes throughout the eth-
nography (see Clark 2012 for an analysis of these). The ‘Language and social life’ group,
on the other hand, consistently voiced their disapproval of any exclusionary attitudes and
advocated in favour of diverging from the mainstream. The fundamental differences in per-
spective of the two groups offer a unique backdrop from which to explore their responses to
specific moments of compulsory heterosexuality.

The role of the ethnographer

An important part of the complex decision-making process involved in doing ethnography
involves the researcher’s relationship to the community or the social space they are studying.
Most ethnographic methodologies define these relationships in terms of four types of role,
defined in Gold’s (1958) work: the complete participant, participant as observer, observer as
participant, and complete observer. As O’Reilly (2009) points out, however, most ethnography
requires some form of participation and some form of observation, and roles are likely to shift
as the ethnography progresses. In my ethnography of the university field hockey team, I ini-
tially understood myself to fit within the category of ‘observer as participant’ – I was not part

123
Jodie Clark

of the hockey team, nor did I know any of the team members before asking their permission to
conduct the study. As I became more involved in their activities – attending socials, supporting
them at matches and, indeed hosting meals – I moved more into the ‘participant as observer’
role. The duration of the hockey-team ethnography was an academic year, during which time
I observed the team’s activities at training sessions, matches, and social events. I collected
conversational data from ten members of the team over a period of seven ‘data-gathering
evenings’, in which I served meals to self-selected groups of two or three participants and
recorded their conversations. I allowed the participants to steer the conversation for these eve-
nings, contributing just enough to stay involved in the conversation. The data-gathering eve-
nings produced a corpus of about 18 hours of conversational data, a corpus of 225,000 words.
My participation in the second ethnography followed a different trajectory, one in which
I moved from being a ‘complete participant’ (a member of the community I was investigat-
ing, as their tutor) to a ‘participant observer’ (a researcher seeking answers to questions that
had emerged as part of my institutional participation). ‘A complete participant’, O’Reilly
explains, ‘is not an ethnographer; he or she is a participant. If she decides to research the
group or culture in which she participates, she becomes a participant observer’ (2009: 154).
The focus group marks the moment when I became participant observer.
The second ethnography was much smaller in scope from the first: I had known and
observed the participants for five months before inviting them to participate in the focus
group. As for the first ethnography, I allowed the participants to steer the conversation, but
asked a few open-ended questions designed to elicit narratives about belonging or exclu-
sion. The focus group lasted 2 hours and generated a corpus of about 20,000 words, which
made it just under one-tenth the size of the hockey-team ethnography.
Beginning in the ‘observer as participant’ role, as I did in the hockey-team ethnography,
required me to engage in quite a lot more information gathering, explicit observation, and
permission-seeking. I was an outsider to this community, and indeed, to the larger world of
university sport in which this community was situated, so I had to gain access by introduc-
ing myself to prominent members of the community, including the team captain and the
president of the Student Athletic Union. I was also dependent upon the willingness of team
members to allow me to participate in their team-only social events and to attend the meals
I had arranged. My initial position as outsider required me to fill the gaps in my knowledge
about the team and its context by engaging in a large amount of time-consuming fieldwork,
research, recording, and transcription. As Eckert remarks, ethnography ‘requires a consider-
able time investment, and frequently more of an investment than an academic [or student!]
is free to make’ (2000: 74).
The second study I conducted provides some ways of using an ethnographic approach
when it is not possible to engage in such intensive, time-consuming fieldwork. As I men-
tioned above, my role as tutor on the participants’ course of study made me a ‘complete par-
ticipant’ in the community I chose to later to observe, which meant there were fewer issues
of access and information gathering. The challenge for smaller-scale ethnographic studies
is less about time and more about perspective and critical distance. As Hammersley and
Atkinson explain, ‘it is in the “space” created by this distance that the analytic work of the
ethnographer gets done. Without that distance, without such analytic space, the ethnography
can be little more than the autobiographical account of a personal conversion’ (2007: 90).
It is worth pointing out, however, that to do research in language, gender, and sexuality
is already to engage in the intellectual and critical distance that Hammersley and Atkinson

124
 Ethnography of compulsory heterosexuality

describe. The priority within Queer approaches, for instance, is to see heterosexuality as
constructed in discourse. The term ‘Queer Linguistics’ (Motschenbacher and Stegu 2013)
is often used to encompass research that critically analyses how language use reinforces or
challenges heteronormativity. Many scholars working within Queer linguistics have drawn
upon ethnographic methods as a means of achieving this critical angle.
Because ethnography privileges sustained engagement with participants over a
longer period of time than survey methods, they usually generate richer data in relation
to each participant in the study. The rich data my own ethnographies produced made
it possible for me to draw connections between participant narratives and community
norms and practices, and thus to recognise instances in which symbolic violence was
being chronicled. As Bourdieu and Waquant (1992: 166) point out, ‘symbolic violence’
can work more ‘efficiently’ than physical violence as a form of social control: it can
cause individuals to undermine those aspects of self that they consider to be valuable.
Symbolic violence is not as easy to detect as physical violence, because participants
may not be able to recognise it as such. The ethnographic data from both studies offered
a rich source of participant narratives in which compulsory heterosexuality inflicted
symbolic violence.

Grammatical analysis and social change

A familiar difficulty for the ethnographer is how to analyse the vast amounts of data that eth-
nographic methods generate. As Angrosino points out, the analysis of ethnographic research
is most often ‘“custom-built” to suit the needs of particular projects’ (2007: 74). In an earlier
work (Clark 2012) I discuss how heteronormativity serves as a structuring principle for
the local and institutional practices of the hockey team. In this chapter, I am focusing on
participant narratives in both groups that depict some form of compulsory heterosexuality
– what I have called ‘moments’ of compulsory heterosexuality. The shift from analysing
norms to analysing participant accounts of their social worlds is motivated by a desire for
empirical research to orient more explicitly towards social transformation. My argument is
that ‘critical readings of individual accounts reveal traces of new, yet-to-be-imagined social
structures’ (Clark 2016: 8) and that by bringing these new possibilities to light, researchers
will be contributing to social change.
As I mentioned above, the critical readings I offer are informed by grammatical analy-
sis, specifically Halliday’s (2014) Systemic Functional Grammar, which describes lan-
guage in terms of three metafunctions: ideational (how human experience is construed),
interpersonal (how social roles and relationships are construed), and textual (how textual
coherence is achieved). I argue that this type of grammatical analysis ‘can reveal the
structure of a given text (textual metafunction), the specific constitution of each “self”
construed or invoked in a text (interpersonal/ideational metafunction) and the “shape”
of the world that is produced in a segment of discourse (ideational metafunction)’ (Clark
2016: 36). In this chapter I focus particularly on the ideational metafunction, drawing
upon Halliday’s framework for analysing transitivity. An analysis of the transitivity of
a chunk of language categorises how different types of experience are construed as pro-
cesses (through verb phrases) and participants (usually through noun and prepositional
phrases). I will demonstrate how these categorisations construct different configurations
of selves and social worlds.

125
Jodie Clark

The grammar of compulsory heterosexuality

As I mentioned above, drawing upon two ethnographies offers the possibility of exploring
distinct responses to similar moments – in this case, moments of compulsory heterosexual-
ity. I was fascinated to discover that the conversational data from both the hockey team and
the ‘Language and social life’ ethnographies offered moments in which participants told
stories about women being coerced into kissing men. At first glance, the narratives seem
rather different. Consider, for instance, Extract 1, which is part of a conversation between
three members of the hockey team – Speedo, Emma, and Sullivan. They are all middle-class
cisgender women between 20 and 21 years old. Speedo and Emma are from the English
Midlands; Sullivan is from Northern Ireland. Speedo is team captain and is in her third and
final year at university; Emma and Sullivan are in their second year. Speedo is telling Emma
and Sullivan about travelling home on the bus after a match with the women’s and men’s
football teams, where the ‘football boys’ tried to get the ‘football girls’ to kiss them. When
the women refused to kiss them, the men came to the conclusion that they were gay, which
Speedo confirmed.

Extract 12
1 Speedo: they ((the football boys)) were trying to make the football girls snog
2 one of them [and I was like oh, it’s not gonna happen]
3 Sullivan: [Oh were there football girls in the bus]
4 Speedo: It’s not gonna happen. They’re like gay, aren’t they. And I was
5 like, mmhm, yeah? [((laughs))] I KNOW them
6 Sullivan: [((laughs))] [All of them are!]

It takes Speedo only a few seconds to tell the story, and the response to it is light-hearted, as
evidenced by the laughter on Lines 5 and 6. Indeed, as will be shown later, the focus of the
subsequent conversation is on which players in women’s football and hockey are gay – not
on the men’s sexual advances.
Extract 2, from the ‘Language and social life’ ethnography, seems to tell a very different
story. Beth is retelling (for my sake) an account she has already shared with Andrew and
Maryam, about her experiences of being bullied at school as a child. Beth describes being
part of a ‘wrong crowd’ when she was in school. The ‘wrong crowd’ was a group of girls
with a set of rules about how to fit in, what to wear, how to do your hair, who to go out with.
The consequences for not abiding by these, for Beth, was physical and emotional abuse.
Beth, Andrew, and Maryam are all 20 years old. Beth and Andrew are white British, from
the same working-class village in the north of England. Maryam is Pakistani British from a
working-class city in the north of England. Beth and Maryam are cis women; Andrew is a
trans man. Extract 2 comes from Beth’s account of her school friends trying to force her to
kiss a boy. When she refused, her friends became physically violent.

Extract 2
1 Beth: Erm (0.4) they tried to make me go out with this boy that I didn’t want to-
2 (0.7) like, I didn’t want to be near him, he were nasty. (0.4) And then when I
3 wouldn’t kiss him, cos I- I didn’t want to, I mean, I were like ↑thirteen and he
4 were a dick (0.7) a:nd er when that didn’t happen they shoved me off a kerb?
5 (0.4) And I cut all me leg, I’ve still got a scar.

126
 Ethnography of compulsory heterosexuality

The tone is very different in Beth’s account of compulsory heterosexuality than in Speedo’s.
Beth’s story has significance to her personally – it represents a specific memory from her
childhood that she has shared at least once before. Beth presents herself as the victim of the
sexual coercion here. Speedo’s story, in contrast, is of something that happened only a few
days before she tells it, and she presents herself in Extract 1 as an observer, not as a main
character in the narrative.
A grammatical analysis of some of the clauses in each of the accounts, however, reveals
some interesting similarities. In both accounts the ‘compulsory’ moment is grammatically
construed in what Halliday calls a ‘three-participant causative’ (2014: 579), where one par-
ticipant – the ‘initiator’ – causes another – the ‘actor’ to act on a final participant – ‘the goal’,
as shown in Table 8.1.
A further similarity is revealed through an investigation of the process types in the clauses
of each account. Material clauses (processes of doing and happening) predominate in Beth’s
account, but there are also some mental (thinking and feeling) and relational (being) clauses,
as shown in Table 8.2.
Speedo’s story, too, consists of mostly material clauses, with one mental and one rela-
tional clause. Her account also includes verbal (saying) clauses, as shown in Table 8.3.
The most significant parallel between the two stories comes into view through a compari-
son of the material processes in each, as shown in Table 8.4. Restricting the accounts only to
their material clauses suggests a striking similarity in the grammatical form of compulsory
heterosexuality. In both cases ‘heterosexuality’ takes two forms: in the first instance, as a
transitive verb indicating sexual interaction: ‘go out with’, ‘kiss’, and ‘snog’; in the second
as the intransitive verb ‘happen’. The ‘compulsory’ element takes the form of the causative
‘to make’. In addition, in both stories the causative is extended by a process of conation, or
trying to do something (Halliday 2014: 572).

Table 8.1 Grammatical analysis of some clauses

Speedo: they were trying to make the football girls snog one of them
Beth: they tried to make me go out with this boy
Initiator Actor Goal

Table 8.2 Investigation of the process types in the clauses of each account

Clauses in Beth’s account Process types

they tried to make me go out with this boy material

when I wouldn’t kiss him material
when that didn’t happen material
they shoved me off a kerb material
I didn’t want to mental
I didn’t want to be near him mental
I didn’t want to mental
he were nasty relational
I were thirteen relational
he were a dick relational

127
Jodie Clark

Table 8.3 Clauses in Speedo’s account

Clauses in Speedo’s account Process types

they were trying to make the football girls snog one of them material
it’s not gonna happen material
it’s not gonna happen material
I KNOW them mental
gay, aren’t they relational
I was like verbal
they’re like verbal
I was like, mmhm, yeah verbal

Table 8.4 Material clauses in Speedo’s and Beth’s accounts

Speedo’s account Beth’s account

they were trying to make they tried to make

the football girls snog one of them me go out with this boy
when I wouldn’t kiss him
it’s not gonna happen (x2) when that didn’t happen
they shoved me off a kerb

Social worlds and vulnerable selves

Having identifying some parallels in the grammatical structure of these two accounts, we
are now in a position to inquire about the shapes of the social worlds these accounts offer,
including how ‘selves’ are constructed. Consider first how each account construes the par-
ticipants in the scene. Although it is clear that there were multiple people in each scene, the
number of clausal participants in both accounts is reduced to three, as shown in Table 8.5. In
Anna’s account, these consist of the plural noun phrases ‘the football boys’ and ‘the football
girls’ (with their associated third-person plural pronouns, ‘they’ and ‘them’), and the first-
person pronoun ‘I’. In Beth’s account the participants are reduced to ‘they’ (Beth’s school
friends), ‘this boy’ (and the associated third-person singular pronouns, ‘he’ and ‘him’), and
the first-person pronoun ‘I’. Note the tendency in each account to treat the groups as discrete
entities; Beth never singles out any of the members of her friendship group (here or at any
other time in the conversation), and Anna only does so one time, with her use of the indefi-
nite pronoun ‘one’ as it is used in relation to the group (‘one of them’).
An important contrast can be seen in how each account constructs the participants
involved in the act of coercion. Both set up an antagonism between two parties. For Speedo
these consist of a heterosexual binary between ‘boys’ trying to kiss ‘girls’ who do not want
to. In Beth’s account, on the other hand, the antagonism is not between male and female par-
ticipants, but between two female participants, Beth and her group of female friends. What
is perhaps more noteworthy is that the two parties in Speedo’s account are two groups (the
two football teams), whereas Beth’s narrative constructs a singular self (‘I’) as at the mercy
of the group she is ostensibly a part of. In other words, Beth’s account is shaped in relation
to a singular self that is vulnerable to the violence caused by a group. Speedo’s account of
compulsory heterosexuality does not seem to involve a vulnerable self.

128
 Ethnography of compulsory heterosexuality

Table 8.5 Participants in Speedo’s and Beth’s accounts

Participants in Speedo’s account Participants in Beth’s account

(the football boys)/they/them they (Beth’s school friends)

the football girls/they/them
one of them (one of the football boys) this boy/he/him
I (Speedo) I/me (Beth)

Table 8.6 A comparison of the clauses that serve the ‘justification’ function

Beth’s account Speedo’s account

I didn’t want to [go out with this boy] (mental) gay, aren’t they (relational)
I didn’t want to be near him (mental)
he were nasty (relational)
I didn’t want to [kiss him] (mental)
I were thirteen (relational)
he were a dick (relational)

Consider next what happens in both accounts when the sexual act doesn’t happen. In
Beth’s account, the consequence is physical harm: ‘when that didn’t happen, they shoved
me off a kerb’. This ‘not happening’ is paralleled in Speedo’s story, but with the future
modality in the verbal clause where Speedo’s talking to the football boys: ‘I was like, “oh,
it’s not gonna happen. … It’s not gonna happen”’. Rather than describing the consequences
of it not happening, Speedo offers an explanation as to why ‘it’s not gonna happen’. The two
‘sayers’ (Halliday 2014: 303) in the dialogue (‘I’ and ‘they’) co-construct a justification:
‘They’re like, “Gay, aren’t they?” And I was like, “Mmhm, yeah?”’
Beth’s story, too, provides a justification for why ‘that didn’t happen’, but hers is
expressed through mental and relational clauses, all of them oriented to her own reasons
why she’s not kissing the boy. A comparison of the clauses that serve this ‘justification’
function, shown in Table 8.6, is revealing.
Justification for her refusal to do the heterosexual act requires a good deal of discursive
reinforcement for Beth. In the first instance, it occurs through the repeated use of the mental
clause ‘I didn’t want to’, and in the second through relating undesirable attributes to the boy
(‘he were nasty’, ‘he were a dick’). The only attribute she relates to herself as a justifica-
tion for not kissing him is her age (‘I were 13’). Speedo’s justification, on the other hand,
requires no mention of what the football girls desire, nor whether the football boys are desir-
able; she needs simply to assign the football girls (‘they’) with the attribution ‘gay’. When
Speedo makes it clear to the football boys that the football girls are gay, they seem to stop
pushing it. The selves here do not seem vulnerable to the type of violence Beth describes
having to suffer. Indeed, at this point in Speedo’s narrative the heterosexuality may not even
seem compulsory: the boys try to kiss the girls, the girls refuse, no one gets hurt – or no one
seems to, anyway.
A closer look, however, reveals that a vulnerable self does emerge in Speedo’s account.
It occurs in the bit of the conversation that occurs just prior to Extract 1, where Speedo

129
Jodie Clark

describes the boys inquiring and speculating about the sexualities of the other women on
the bus.

Extract 3
1 Speedo: All the boys kept going on, how many of the girls are
2 lesbians in the bus and I was just like, oh and they eventually
3 came and they were like you clearly are. ((laughing)) And I
4 was like, [what?]
5 Emma: [((laughs))]
6 Sullivan: I’ve got a [boyfriend!]
7 Speedo: [You’ve] done it! (0.4)
8 And I was like [nnyah]
9 Emma: These [weren’t] hockey [guys though]
10 Speedo: [nn yeah ]
11 Emma: Oh.
12 Sullivan: But they don’t know [Speedo ]
13 Speedo: [They were] freshers
14 Sullivan: They were freshers
15 Emma: Oh they were [freshers]
16 Speedo: [But they were] with the- football

Compulsory heterosexuality does not show up straightforwardly in this extract as a grammati-

cal construction with a three-participant causative. Nevertheless, it is here that we can find a
parallel to the vulnerable self that shows up in Beth’s story in Extract 2. Consider the partici-
pants in Speedo’s account, as listed in Table 8.7. The singular self occurs here as the first- and
second-person singular pronoun, and, as in Beth’s story, it is pitted against an antagonistic
group. The antagonism here occurs not through material processes, but through a series of ver-
bal clauses, in which ‘the boys’ interrogate Speedo’s sexual identity, as can be seen in Table 8.8.
It is worth pointing out, however, that the verbal processes here do not depict a straight-
forward antagonism between the self and group, as they did in Beth’s account. Beth’s narra-
tive constructed her resistance to the group’s coercion through a series of mental processes:
‘I didn’t want to’, ‘he were nasty’, ‘he were a dick’. Speedo’s account does not offer any
insights into her own mental processes: she does not indicate whether she agreed or disa-
greed with the boys’ assessments about her sexuality, and the verbal responses she reports
suggest a lack of commitment. Indeed, the self constructed here is one that resists being
labelled as either straight or gay. I would argue that it is this self – the self that resists such
labelling – that is vulnerable in this account. It is vulnerable not only to the questioning of
the boys on the bus, but also to the responses to her account that Emma and Sullivan offer
in the rest of the extract.

Table 8.7 Participants in Speedo’s account

Participants in Speedo’s account in Extract 3

all the boys/they

the girls … in the bus
I/you (Speedo)

130
 Ethnography of compulsory heterosexuality

Table 8.8 Verbal clauses interrogating Speedo’s sexual identity

Sayer Verbiage

the boys how many of the girls are lesbians in the bus
you clearly are
you’ve done it
Speedo oh
what?
nnyah

This vulnerability becomes visible with a closer look at how Speedo’s narrative becomes
a collaborative telling at line 6. Note the different constructed dialogue responses to the
moment of the story in which Speedo is labelled a lesbian (‘you clearly are’). Sullivan joins
in at this moment, supplying a line of dialogue on Speedo’s behalf – ‘I’ve got a boyfriend’ –
which is likely designed to put an end to further questioning about her sexuality.
Perhaps more revealing than the dialogue that Speedo attributes to herself are those she
attributes to the footballer who is interrogating her. In line 3 she has the footballer’s accusa-
tion take the form of a relational clause: ‘you clearly are [a lesbian]’. Once Sullivan pro-
vides what might be considered evidence that Speedo is not a lesbian (having a boyfriend),
Speedo repeats the initial accusation to take the form of a material clause in line 7: ‘you’ve
done it’. This shift can be interpreted as a subtle way of countering Sullivan’s assumption
that being in a heterosexual relationship precludes the possibility of being gay.
Speedo’s narrative in Extract 3 can be understood as an act of resistance to Sullivan’s
attempt to impose a heterosexual identity upon her. Not only does Speedo resist being
straightforwardly labelled as straight, she also resists the silencing of her non-normative
sexual experiences, as becomes clear with the continuation of Speedo’s narrative, in Extract
4. Here Speedo is revealing to Emma and Sullivan that she has ‘tried it’ (in other words, she
has experimented sexually with women in the past), but her previous team members never
judged her for doing so.

Extract 4
1 Speedo: But then it is- you do- the reason I- (0.2) a lot of the time I tried
2 it was because the other people [were]
3 Emma: [Everybody] did it, it’s peer
4 pressure, I understand [that]
5 Speedo: [It wasn’t] peer pressure!
6 [It was just what I was seeing] I was like [oh, you know]
7 Emma: [Not peer pressure it’s just] [you’re in that]
8 environment [aren’t you]
9 Sullivan: [hhh]
10 Emma: Yeah, but it’s not WRONG, is it [if s-]
11 Speedo: [THAT’S THE THING,]
12 if everyone else did it it wasn’t wrong, it didn’t matter, who
13 cares, if somebody did it now in front of like half of our friends,
14 they’d all be like oh my God

131
Jodie Clark

Note Speedo’s attempt to give voice to her non-normative sexual experiences in Lines 1−2:
‘a lot of the time I tried it was because the other people were’. Emma’s response attempts to
reframe Speedo’s experience to make it understandable to her. By calling it ‘peer pressure’,
Emma can make sense of it (‘I can understand that’) because it fits within a framework that
she and other teammates have formulated in which university students experiment with gay
sex because they are insecure and have not yet matured (Clark 2013).
With Speedo’s rejection of Emma’s move toward intelligibility (‘It wasn’t peer pres-
sure!’, Line 5) comes a discussion of an image of community that Emma may not have
ever considered: a community in which sexuality is not constrained by norms. In this
view of community, heterosexuality is not a structuring principle, nor are sexual experi-
ences subject to ‘pressure’; instead ‘it wasn’t wrong, it didn’t matter, who cares’ (Lines
12–13).
Speedo’s acts of resistance here reveal a ‘self’ that her friends attempt to silence in this
conversation. In my discussion of Beth’s account, I described a ‘vulnerable self’, vulnerable
by virtue of being subject to physical violence from a norm-enforcing community. I would
argue that Speedo constructs a self in her narrative that is vulnerable to symbolic violence
from the silencing and from disavowing her sexual experiences.

Alternative social worlds

My approach here has been to identify instances in which heterosexuality makes itself felt
as compulsory, where heterosexuality is compulsory by virtue of its imposing violence to
the self. I have argued that such accounts help us understand grammatical constructions of
selves that are vulnerable in the face of compulsory heterosexuality.
After having constructed an account of a vulnerable self in the face of what she calls ‘the
wrong crowd’, Beth goes on to describe and construct another type of social world, one that
protects her from the types of attacks she encountered with her former friendship group. She
recounts joining a drama group, where she met people with whom she was to form ‘last-
ing friendships’ (Clark 2016: 64). ‘I just knew,’ she said, ‘that I needed to keep with these
friends that I made, cos they were- they still are so protective over me’ (Clark 2016: 65).
With the arrival of the new friendship group, Beth said, the ‘bullying just stopped. […] It’s
like they saw me get confident and get a good group of friends around me and just backed
away’ (Clark 2016: 65).
The image Beth presents here is of use to gender and sexuality scholars, especially in
relation to the account of compulsory heterosexuality discussed in this chapter. The social
world Beth describes is not one that resists the force of compulsory heterosexuality, but
rather one that supports and protects the integrity of each member’s sense of self.
Speedo’s account also offers a new way of thinking about community. In the same way
that for Beth, her ‘good group of friends’ served as ‘an envelope to protect the integrity
of the self’ (Clark 2016: 71), Speedo’s nostalgic account of a more sexually progressive
hockey team serves as a means of making visible a disavowed, unintelligible self.

Future directions
I have argued that ethnographic methods make it possible to identify the adverse effects of
gender and sexuality norms at particular moments in particular social spaces. I have analysed
two instances of compulsory heterosexuality from my own ethnographic data, both result-
ing in violence, one physical, one symbolic. As I mentioned above, instances of symbolic

132
 Ethnography of compulsory heterosexuality

violence are harder to spot, especially in largely norm-enforcing communities. Identifying

these moments is well worth the effort, though, because it sheds light on the specific, localised,
and largely unnoticed ways in which people are adversely affected by heteronormativity.
More importantly, though, focusing on accounts of symbolic or physical violence makes
it possible to ask questions about how the self and the social world are grammatically con-
structed in these accounts, and to identify the alternative structures that these accounts pro-
duce. My analysis of Beth’s and Speedo’s accounts of compulsory heterosexuality reveals
a fascinating pattern. They both first constructed a self that is vulnerable to the norms of
the community. They then made reference to an alternative type of community, one that is
accepting and protective of this vulnerable self.
Much gender and sexuality ethnographic research focuses on community-level resist-
ance to compulsory heteronormativity. That said, as Hall (2013) points out, such research
focuses on how these social spaces are structured according to locally constructed norms.
In Barrett’s (1999) ethnographic study on drag queen performances, for instance, the norms
take the form of a culture-specific notion of authenticity, which become the criteria upon
which these performances are evaluated. Similarly, Jones’s (2012, 2014) ethnography of a
British lesbian hiking group reveals a set of norms that centre around community-specific
ideas about what it means to be an authentic lesbian. In Jones’s (2016) ethnography of an
LGBT youth group the normative pattern is oriented to the racialised marginalisation of
others. These studies present a pattern in which social spaces will inevitably be governed
by sets of norms.
I would encourage language, gender, and sexuality ethnographers to widen their focus,
such that they look not only to the norms that structure the community, but also to those
fleeting moments in the data in which participants construct alternatives to oppressive struc-
tures. A close look at Beth’s and Speedo’s responses to instances of compulsory heterosexu-
ality offers a new way of thinking about how selves might be situated in social spaces. Their
accounts make it possible for gender and sexuality researchers to consider the possibility
that social worlds might be configured in such a way as to protect the integrity of a vulner-
able self, rather than to enforce norms upon individuals.
Such a perspective, combined with the ethnographic depth and the analytic rigour of
grammatical analysis, offers gender and sexuality researchers new ways of understanding
the complexities of heteronormativity and the possibility of imagining how these might be
transformed.

Notes
1 The name of this university, and the names of all participants are pseudonyms.
2 The transcripts follow my modified form of the conventions used for conversation analysis (CA):

[ ] Overlapping speech
Underlining Emphasis
CAPITALS Loud speech
(0.4) Length of a pause in seconds
(.) Pause less than one-tenth of a second
((laughs)) Transcriber’s descriptions or comments, contextual information
((...)) Words or lines omitted
() Indecipherable
(word) Transcriber’s best guess at what was said
sto::p Colons indicate elongation of a sound (number of colons corresponds to length
of elongation)

133
Jodie Clark

hhh Out-breaths
.hhh In-breaths
(as with colons, number of h’s corresponds to length of out-breaths or in-breaths)
, Weak, ‘continuing’ intonation
? Rising, ‘questioning’ intonation
. Falling intonation

Further reading
Clark, J. (2012) Language, sex and social structure: analysing discourses of sexuality. Basingstoke:
Palgrave.
A detailed explanation of how I conducted the hockey-team ethnography can be found here. The
book describes the structuring principles of the heteronormative practices of the women’s hockey
team.
Clark, J. (2016) Selves, bodies and the grammar of social worlds: reimagining social change. London:
Palgrave.
This book contains more background on the ethnographic study of Beth, Maryam, and Andrew that
I describe in this chapter. It offers insights into how to draw upon ethnographic and discourse analytic
research to reimagine social structure.
Livia, A. and Hall, K. (1997) Queerly phrased: language, gender, and sexuality. New York: Oxford
University Press.
This edited collection offers a wide range of contexts in which questions of identity, community,
selfhood, and normativity are explored through the lens of language, gender, and sexuality. Many of
the chapters in this volume draw upon ethnographic research.
Motschenbacher, H. and Stegu, M. (2013) ‘Queer linguistic approaches to discourse’, Discourse &
Society, 24, pp. 519–535.
This article outlines the basic principles of Queer theory and reviews its applications in linguistic
and discourse-oriented research.
O’Reilly, K. (2009). Key concepts in ethnography. London: Sage.
This text offers a concise and critical explanation of issues and terminology associated with
ethnographic research.

Related topics
Language, gender and sexuality: reflections on the field’s ongoing critical engagement with the
sociopolitical landscape; applying queer theory to language, gender and sexuality research in schools;
identity construction in gendered workplaces; using communities of practice and ethnography to
answer sociolinguistic questions; gender and sexuality normativities.

References
Angrosino, M. (2007) Doing ethnographic and observational research. Los Angeles: SAGE.
Barrett, R. (1999) ‘Indexing polyphonous identity in the speech of African-American drag queens’,
in Bucholtz, M., Liang, A. C., and Sutton, L.A (eds) Reinventing identities: the gendered self in
discourse. Oxford: Oxford University Press, pp. 313–330.
Bourdieu, P. and Waquant, L. (1992) An invitation to reflexive sociology. Cambridge: Polity Press.
Bucholtz, M. (1999) ‘“Why be normal?”: language and identity practices in a community of nerd
girls’, Language in Society, 28, pp. 203–223.
Butler, J. (2006) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Clark, J. (2012) Language, sex and social structure: analysing discourses of sexuality. Basingstoke:
Palgrave.

134
 Ethnography of compulsory heterosexuality

Clark, J. (2013) ‘“Maybe she just hasn’t matured yet”: politeness, gate-keeping and the maintenance
of status quo in a community of practice’, Journal of Politeness Research, 9, pp. 211–237.
Clark, J. (2016) Selves, bodies and the grammar of social worlds: reimagining social change. London:
Palgrave.
Eckert, P. (2000) Linguistic variation as social practice. Oxford: Blackwell.
Eckert, P. (2011) ‘Language and power in the preadolescent heterosexual market’, American Speech,
85, pp. 85–97.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Think practically and look locally: language and gender as
community-based practice’, Annual Review of Anthropology, 21, pp. 461–488.
Gold, R. L. (1958) ‘Roles in sociological field observations’, Social Forces, 36, pp. 217–23.
Hall, K. (2013) ‘“It’s a hijra!” queer linguistics revisited’, Discourse & Society, 24, pp. 634–42.
Halliday, M. A. K. (2014) Introduction to functional grammar. London: Routledge.
Hammersley, M. and Atkinson, P. (2007) Ethnography: principles in practice. London: Routledge.
Jones, L. (2012) Dyke/girl: language and identities in a lesbian group. Basingstoke: Palgrave.
Jones, L. (2014) ‘“Dolls or teddies?”: constructing lesbian identity through community-specific
practice’, Journal of Language and Sexuality, 3, pp. 161–190.
Jones, L. (2016) ‘“If a Muslim says ‘homo’, nothing gets done”: racist discourse and in-group identity
construction in an LGBT youth group’, Language in Society, 45, pp. 113–133.
Koller, V. (2015) ‘The subversive potential of queer pornography: a systemic-functional analysis of a
written online text’, Journal of Language and Sexuality, 4, pp. 254–271.
Moore, E. (2006) ‘“You tell all the stories”: using narrative to understand hierarchy in a community of
practice’, Journal of Sociolinguistics, 10, pp. 611–640.
Morrish, E. (1997) ‘“Falling short of God’s ideal”: public discourse about lesbians and gays’, in Livia,
A. and Hall, K. (eds) Queerly phrased: language, gender and sexuality. Oxford: Oxford University
Press.
Motschenbacher, H. and Stegu, M. (2013) ‘Queer linguistic approaches to discourse’, Discourse &
Society, 24, pp. 519–535.
O’Reilly, K. (2009) Key concepts in ethnography. London: SAGE.
Peterson, D. J. (2010) ‘The “basis for a just, free, and stable society”: institutional homophobia and
governance at the family research council’, Gender and Language, 4, pp. 257–286.
Rich, A. (1980) ‘Compulsory heterosexuality and lesbian existence’, Signs, 5, pp. 631–660.
Sauntson, H. and Morrish, L. (2012) ‘How gay is football this year?: identity and intersubjectivity in
a women’s sports team’, Journal of Language and Sexuality, 1, pp. 151–178.
Seidman, S. (2009) ‘Critique of compulsory heterosexuality’, Sexuality Research & Social Policy, 6,
pp. 18–28.
Wittig, M. (1992) The straight mind and other essays. Boston: Beacon Press.

135
 9
Anthropological discourse analysis
and the social ordering of gender
ideology
Susan U. Philips

Introduction
In this chapter I explain how I carried out anthropological discourse analysis of gender
ideology in the Polynesian Kingdom of Tonga, a tiny nation in the South Pacific. I show
how anthropological discourse analysis examines the relationship of forms of talk to their
position within the broader social order, as well as to their immediate contexts of speech
production. The ability to provide such analysis depends on a methodology that records not
just one form of talk, but multiple forms of talk. When this approach is applied to the study
of gender ideology, it reveals how ideas about women vary systematically across social
domains. Anthropologists are also interested in how the same gender ideology occurs over
and over again in repetitions of a single form of talk, or in the reproduction of culture in
discourse. In this respect anthropological discourse analysis complements contemporary
approaches to the study of gender and language that stress the processual, emergent, and
contingent orienting to gender in face-to-face interaction.
In the discussion to follow I first explain how diversity in gender ideology came to be an
issue in feminist cultural anthropology. I then focus attention on the ethnographic method-
ology used by anthropologists to study gender. This includes the role of recording units of
interaction or discourse units within that ethnographic methodology. Finally, after a descrip-
tion of the design of my research project in Tonga, I provide an account of gender ideology
in Tonga. I contrast two published microanalyses of gender ideology from two recorded
contexts: courtroom cases of bad language and the singing, and comment on singing, in
a workgroup where women were putting pieces of bark cloth together and decorating the
larger piece they were making. The main purpose of this comparison is to show how differ-
ent kinds of analysis can be done using different parts of a data base. The possibilities for
analysis depend on how the larger project was designed and on aspects of the original data
collection and processing of data in the field.

Gender ideology
The concept of gender ideology emerged during the women’s movement of the 1970s. The
women’s movement, formally called the Women’s Liberation Movement, began in the

136
 The social ordering of gender ideology

United States and spread rapidly to other countries. Today this movement is often referred
to as ‘second-wave feminism’, to relate it back to the first wave of feminism in the nine-
teenth century. A much wider range of issues was raised in the 1970s when every aspect of
women’s lives was subjected to feminist critique, including very general ideas about men
and women.
The term ‘ideology’ had deliberate Marxist connotations. This was particularly true of
the influential pioneering feminist theory of Firestone (2003 [1970]). The idea was that
men dominated women just as the bourgeoisie dominated the working class. This domina-
tion was accomplished partly through ideology maintaining that women’s subordination
was both natural and in women’s interest when, in reality, it was neither. In current discus-
sions of ideas about women some scholars use the term ‘discourses’ in the Foucauldian
sense (Foucault 1971) to talk about conventionalised sets of ideas about women and men
that appear in language use. Others, like myself, retain the concept of gender ideology to
foreground its connotations of the obscured power relations entailed in ideas about women
and men.
Negative visions of women harm them psychologically and are used to justify their sub-
ordination. Multiple gender ideologies exist in all societies and some gender ideologies
clearly benefit women more than others. Those beneficial to women can be used creatively.
For example, in the United States, when fathers are urged to support stronger laws against
sexual assault by being asked how they would feel if their daughters were raped, they are
being asked to extend positive concepts of the father–daughter relationship to all women.
The expansion of positive views of women and the denial of negative views of women, then,
is a necessary part of a feminist agenda for enhancing the well-being of women worldwide.
The cross-cultural documentation of gender ideology provides the basis for understanding
the nature of both positive and negative views of women.
Gender ideology was taken up from the women’s movement by feminist anthropolo-
gists in the early 1970s as a topic of interest. Feminist anthropologists are committed to
the relevance of feminism for all women everywhere, and not just in Western European
societies. In feminist anthropology the organisational distinction between public and pri-
vate domains and women’s exclusion from public domains was early on characterised as
central to women’s subordination worldwide and as justified by gender ideologies (Philips
2001, 2014).
Gender ideologies are often specific to the activities in which women participate, and
to whole social domains or institutions made up of multiple activities. The location of a
particular activity within a larger system can play a role in whether a given way of thinking
and talking about women can travel, can move or spread from one institutional domain to
another (Philips 1998, 2004a). Consider the spread of critiques of male hegemony due to the
women’s movement of the 1970s. These critiques moved rapidly from community activism
into academia, law, education, and medicine, but how the critiques have been central ideo-
logically differs from institutional domain to institutional domain.
For example, in the educational domain there was and still is a concern with girls’ and
women’s access to the same forms of knowledge as men have, and to equality of treatment
in and by educational institutions. In contrast, consider the legal domain. Here one key femi-
nist issue, and the one given greatest attention in the language and law literature, has been
to stop various forms of aggression towards women through the greater criminalisation of
such forms that were more acceptable in the past. In this language and law literature, the key
issue in both legal behaviour and in gender ideology has been women’s moral credibility
and the way it is undermined, particularly in cases of rape (Matoesian 1993).

137
Susan U. Philips

There are also domains where feminist ideologies have had difficulty penetrating. Some
would say that would be all domains, yet, clearly, we see more response to feminist issues
in some domains than others. Worlds of engineering, computer science, and business are far
more resistant to the presence of women than is true of law, medicine, and academia. As
McElhinny (2014) has argued, more thinking organisationally, more pursuit of social order,
is needed in the study of gender and language. Linguistic and cultural anthropologists have
developed the idea that gender ideology or views of women emerge differently in different
social contexts by methodologically recording multiple forms of talk which vary in some
socially significant way.

Ethnography and discourse analysis in linguistic anthropology

Anthropologists typically gather discourse data while engaged in ethnographic field work.
Ethnography as a methodology emerged conceptually within anthropology in the early
twentieth century and has spread to other disciplines over the last half century. With this
spread, the meaning of ethnography has undergone change and has come to centrally mean
different things to researchers in different disciplines.
In the anthropological subdisciplines that deal with living people, ethnography as a meth-
odology has several core or classic features not all of which may be present in a particular
research project. Direct presence, or the immediate personal involvement with people in
the field situation, is the key feature (as opposed to, say, mailing questionnaires to people
or focusing on documents in archives). Long-term continuous presence in the field site
of ethnographic fieldwork and learning the local first language are important features of
ethnographic fieldwork. Immersion in the local culture so that one occupies a social organi-
sational position within the society in some sense is also considered highly desirable. This
often means living with a local family while in the field.
Participant observation in local activities is considered the specific method most char-
acteristic of ethnography (as opposed to the interviews or experiments that anthropology
shares with other disciplines). To engage in participant observation a researcher joins in or
becomes part of the activities relevant to the research topic of the project. The aim is to be
involved in a way that does not significantly transform or disrupt the activity in which one
participates. Often the easiest way to be a participant and a selectively focused observer at
the same time is as a member of an audience. Selective focus of attention is guided by the
theoretical orientation of the research project. For example, in a study of gender ideology
there is a focus of attention on how men and women talk and on how people talk about men
and women, often regardless of the situation one is in. Field notes written during and after
an activity are the standard way to make a record of the focus of attention sustained during
participant observation. These properties of ethnographic fieldwork are considered relevant
for a wide range of kinds of projects.
Various subdisciplines of anthropology also have more specific activities tailored to their
particular theoretical and substantive agendas. For example, cultural and biological anthro-
pologists interested in the relation between adaptation to an environment and other aspects
of culture and biology might take soil samples or count crop yields.
In linguistic anthropology our specialised method is the recording, transcription, transla-
tion, and computer entry of socially occurring speech. By ‘socially occurring’ I mean activi-
ties that would take place whether the researcher was present or not, activities that are part
of the social fabric of a place. Audio and video recordings are technological extensions of

138
 The social ordering of gender ideology

participant observation. They preserve communicative behaviour in far greater detail than
field notes, although they cannot take in an activity in the way that being present can for the
researcher. Long-term ethnographic experience provides information about the positioning
of the activity being recorded within a broader social order. The broader ethnographic expe-
rience also offers insight into how the activity is related to other activities.
This comparative thinking about the activities in which speech occurs is motivated by
linguistic anthropological theory about the nature of language that was initially put forth in
the ethnography of communication (Hymes 1964). It is part of the nature of language that its
central role in human society is to communicate, to transfer information from one person to
other people. Culture is constituted through language use, as well as through other semiotic
means. Social life itself is differentiated into various institutional domains and all humans
draw selectively upon their linguistic resources to accomplish different social actions. From
the inception of the ethnography of communication, with its emphasis on studying language
ethnographically, there has been a fundamental methodological commitment to comparison
of different speech genres or forms of talk as a way of revealing what specific linguistic
forms are implicated (how linguistic resources are drawn upon selectively) in the constitu-
tion of different social realities.
Units of interaction that are bounded in some way are the basic units of research focus
in this tradition because they are viewed as discourse units of meaning. Units of interaction
can of course vary in complexity and scale. They also vary in how boundaries, i.e. begin-
nings and endings, are created and made sense of. For example, in trial courts in the United
States a trial can be considered a unit of interaction, yet within that unit there are smaller
named units, such as the opening, testimony, and the closing. A trial opening will consist of
opening speeches by both the prosecution and defence. The boundaries between larger units
are often marked by actual physical shifts in the spatial positioning of participants who have
social identities specific to the activity. For example, in a trial where one witness’s testimony
constitutes a unit, that witness leaves the witness box when her testimony ends and she is
replaced by another witness.
Different speech genres, forms of talk, or units of interaction will be enacted through dif-
ferent frequencies of particular linguistic forms. Comparative analysis of multiple instances
of the same form of talk or of different forms of talk reveals how through language different
cultural actions are accomplished. Close examination of different forms of talk makes it
clear that such units, like a meeting, a dinner, a song, or a prayer, not only accomplish dif-
ferent social purposes through the selection of particular linguistic forms, but also through
the organisation of turns at talk relative to participant identities and through the sequen-
tial structure of the discourse itself. The cultural meaning of an activity is not constituted
through the particular selection of linguistic resources alone.
I have emphasised here that the recording of multiple forms of talk and their comparison
distinguishes anthropological discourse analysis from other kinds of discourse analysis. In
other respects anthropologists have a great deal in common with other discourse analysts
in our approach to discourse. Under the influence of conversation analysis anthropologists
have relied heavily on recordings that are transcribed, on diacritics in the transcripts that
capture some qualities of speech, and on analysis through repeated readings and reformat-
tings of the transcripts rather than through repeatedly listening to the recordings themselves.
Like conversation analysts, anthropologists have also been influenced by ethnomethodol-
ogy and see social realities as constituted through talk in face-to-face interaction. Like criti-
cal discourse analysts, anthropologists who analyse discourse are interested in power, in the

139
Susan U. Philips

way relations of domination and subordination are constituted through discourse (Philips
1998, 2004b). And like social interactionists who share our ethnography of communica-
tion heritage, anthropologists have a theoretical commitment to the idea that language has
meaning through its indexical relationship to the context in which speech is uttered, or to
its situatedness.
In contrast to other discourse analysis traditions, however, anthropologists tend to work
in non-European cultures and on non-Indo-European languages. Here the broader ethno-
graphic method of participant observation is crucially drawn upon in making sense of spo-
ken discourse and in locating multiple forms of discourse within a broader social order.
In thinking about how speech is socially organised above the level of its immediate inter-
actional context, linguistic anthropologists draw upon sociocultural distinctions that have
proved useful in cultural anthropology and sociology. General organisational concepts such
as ‘social domains’ and ‘institutions’ have been part of sociolinguistics as broadly defined
for decades. The distinction between private and public social domains has been important
in cultural anthropology and linguistic anthropology in describing the gendered organisa-
tion of speech and the way such organisation can subordinate women. As we will see in the
next section, the distinction between state and civil society and the role of political economy
in the gendering of speech have also been important social organisational concepts for me
as well as for other students of gender.
I turn now to a discussion of the research design that led to my characterisations of gen-
der ideology in Tongan discourse.

The research design of my Tongan study of language use

The data I have on Tongan gender ideology comes from a research project I carried out
in the Polynesian Kingdom of Tonga. Culturally and linguistically Tonga has a great deal
in common with other Polynesian societies in the region, including Samoa, Tahiti, and
Hawai’i. Tonga as a nation state is comprised of a group of islands in the South Pacific
with a population of approximately 100,000. An equal number of Tongans live overseas in
a diaspora located primarily in former British colonies, including Great Britain itself, the
United States, New Zealand, and Australia. Tonga was a colony of Great Britain for a period
during which British governmental and legal forms were introduced into the country. And it
is a constitutional monarchy up to this day.
The primary purpose of the research project that provided the data on Tongan gender
ideologies was to examine the way convincing evidence is created through language use in
Tongan courtrooms. What Tongan linguistic resources were mustered to convey greater or
lesser certainty or persuasiveness about what really happened in court cases where guilt or
innocence is decided? My thinking was that to describe what was characteristic of evidence
in Tongan courtroom resolutions of conflict, it would be useful to compare language use in
different court jurisdictions with each other, and also to compare courtroom speech with
non-courtroom speech. For non-courtroom speech, I sought language use in other formal or
highly structured public activities, particularly meetings where conflicts might similarly be
dealt with. I also aimed to record more conversational activity to see if non-court and court
activities were similar to each other yet different from conversation by virtue of their formal
public nature.
I was particularly interested in comparing language use in the higher versus lower trial
courts. In Tonga the lower-level Magistrates’ Courts that heard misdemeanors and minor

140
 The social ordering of gender ideology

civil matters were all carried out in the Tongan language and were presided over by Tongan
magistrates who did not have formal legal degrees. The higher-level Supreme Court heard
felonies and major civil suits, was bilingual in Tongan and English, and was presided over
by a British judge with formal legal training. In British-derived legal systems evidence
standards are more strictly imposed in such higher-level trial courts. This raised questions
about whether and how different legal standards for evidence would be manifested in the
two court levels.
In an earlier pilot research project in Tonga I had seen how these two court levels dif-
fered in their locations in ways that correlated with a political economic concept of social
organisation. Gundar-Frank’s concept of political economy (1967) directly or indirectly had
already influenced other linguistic anthropological research relating language, gender, and
political economy that I have reviewed (Philips 2004b). Frank characterised nation states
in Latin America that had been colonised by Europeans as consisting of communities of
increasing size and institutional complexity. Each community had economic resources that
were being extracted from it by the next largest community under the coercion of Europeans,
with all wealth ultimately flowing to European economic centres.
I had seen the way Alaskan indigenous communities were organised into villages and
regional centres, and I had seen how actual formal law only extended institutionally as far
‘down’ as the regional centres in Alaska. In Tonga I saw again that courts functioned dif-
ferently in regional centres and villages. The villages on the main island of Tongatapu had
homes, churches, schools, multi-purpose halls or large sheds, and small thinly stocked road-
side stands that sold canned food and cigarettes. In addition to such structures, the regional
centre of Nuku’alofa had supermarkets, a large open-air market for local produce, banks,
government offices, stores, restaurants, gas stations, and a bus station that was the starting
point for all routes on the island. The other smaller island groups similarly had a village-
regional centre organisational distinction based on the kinds of activities possible in them.
In Tonga, only Nuku’alofa, the capital of the country, had a higher-level trial court,
the Supreme Court, a court of general jurisdiction. The judge from that court went on cir-
cuit to outlying island groups to hear cases. This regional centre also had two lower-level
Magistrates’ Courts with limited jurisdiction that met every day. Most villages did not even
have Magistrates’ Courts. There were three such courts on each side of the main island and
each of them only met for a few hours one day a week.
Based on my awareness of this political economic organisation of Tongan communities,
the first year of the project was located in a village and the second year of the project was
located in the regional centre of Nuku’alofa. In Year One, I lived with a family in a large
village for several months where there was a court and I first observed and then recorded all
proceedings in the local court on a weekly basis. I also recorded a women’s work group in
which women put pieces of bark together and stencilled designs on them with a vegetable
dye. My purpose was to get a sample of women’s conversation. I had learned during an ear-
lier pilot project that although people in the village were comfortable with, and some were
even keen on, having public activities recorded, they were not so eager to have their private
lives recorded inside their homes. In addition, socialising that was not on a large scale was
usually gender segregated. Such socialising might take place in a work group in a back
yard. Or it could be a party on the verandah where the narcotic root kava was drunk dur-
ing conversation and singing. So I knew my recording of more casual conversation would
be gender segregated. It was these understandings that led me to the women’s work group,
called a koka’anga.

141
Susan U. Philips

Following those few months that I lived in this village, a University of Arizona gradu-
ate research assistant moved into the village and took on responsibility for recording other
kinds of non-court speech activities that included village-level meetings of various kinds,
church activities, feasts, and men’s conversation in gender segregated kava parties and a
work group. At my request this research assistant made a point of recording women as well
as men in public activities, even though their presence in public was limited.
Year Two was based in the regional centre. I spent the first few months recording pro-
ceedings in the higher-level Supreme Court and then in the final phase a second gradu-
ate student assistant moved into the regional centre and recorded proceedings in the two
lower-level Magistrates’ Courts. In these four phases of the research as I alternated with my
research assistants, we timed our presences so that we always overlapped, so that before one
of us left the next one would arrive.
During all phases of this two-year period we hired local Tongans to transcribe, translate,
and computer enter subsets of the audio recordings. Roughly about half of the audio record-
ings, or about 50 hours of recordings, were processed in this way.
In sum, this project was designed to allow for several different kinds of comparisons.
These comparisons were both potentiated and constrained by my political economic under-
standing of Tongan social order: comparison between Magistrates’ Courts in the village
versus the regional centre; comparisons between the Magistrates’ Courts and the Supreme
Court; comparisons between village-level court talk and village-level non-court formal pub-
lic activities; and comparisons between conversational speech and more formally organised
speech at the village level. This did not mean that I fully intended to pursue all of those
comparisons; rather, it meant that I could make such comparisons if it appeared that to
do so would shed light on what was characteristic of Tongan legal language. It was at this
stage of the research process, then, that the potential for anthropological discourse analysis
involving the comparison of multiple forms of discourse, recognised as socially ordered,
was created.

The gender ideology data

When all of the data was considered together, it was clear that three gender dyads were
being made salient by Tongans in discourse across a range of situations, including many
in which I participated without recording. These were the sister–brother relationship, the
husband–wife relationship, and the sweetheart–sweetheart relationship (Philips 2014). In
Besnier’s earlier (1997) discussion of Tongan faka-leiti, ‘men who act like women’, one
can see that the men talked about all three of these relationships: the duty to the sister not to
shame her and the pain that the sweetheart relationship could not become that of husband
and wife. In later work (2004), Besnier characterised these dyads as traditional and argued
that there were many kinds of modern identities such as flea-market seller or pawnshop
dealer deriving from transnational relationships with overseas Tongans that simply could
not be captured by those dyads. Nevertheless their generative power is still robust. Earlier
research on gender identities in Samoa and Tonga by cultural anthropologists had given
attention to the sister and wife identities, but had not recognised the import of the sweetheart
relationship. In addition, these identities were often discussed as if they floated free, rather
than being focused on and made explicit in particular social contexts that were part of the
larger social organisation of Tongan life. A linguistic anthropological approach that relies on
recording socially occurring activities is able to provide this grounding.

142
 The social ordering of gender ideology

The three dyads were predictably talked about or oriented to in socially organised specific
forms of talk and social situations. The sweetheart relationship was salient in love songs that
saturated the radio and were composed and sung live in and for a range of situations. The
husband–wife relationship appeared early on in my data collection, but it was the least sali-
ent of the three dyads overall in the data base. The sister–brother relationship was the dyad
that came up in the greatest range of circumstances. The brother is said to be subordinated
to the sister and should adhere to her wishes, whereas in the wife–husband relationship
the husband is clearly dominant. This means the sister identity is the most productive and
prominent identity to draw upon metaphorically in advocating positive treatment of women.
There were four key sources of information that I drew upon to make sense of the record-
ings themselves. First, there was my actual participation in a wide range of Tongan situ-
ations, and participation with my research assistants in the activities I recorded. Second,
my female assistants and I took notes during recording and afterwards we reviewed what
had gone on and I asked them questions about their knowledge of the activity we had just
recorded. I also reviewed some data with two highly respected older Tongan consultants
considered to be specialists in Tongan language and culture, one male and one female. With
the male who was considered the most respected intellectual in Tonga, I reviewed data from
the village Magistrate’s Court. With the female, who had been the principal of the most
highly regarded girls’ high school in the country, I worked on translating a play she had
written for the girls in her school that modelled traditional female behaviour.
Third, during the processing of the speech data with my assistants, through transcrip-
tion, translation, and computer entry of data while in the field, I asked them questions about
the material we were working with and they also frequently brought to my attention issues
they thought they needed to explain to me. Fourth, I carried out close textual analysis of
transcripts and translations of subsets of the data we had worked with, primarily while back
in the United States, bearing in mind the other sources of information already drawn upon.
Close textual analysis of gender ideology was carried out on two subsets of data: four
cases of ‘bad language’, or lea kovi, from the Magistrate’s Court data (Philips 2000) and
one four-hour women’s work party from the village where I lived during my first field trip
(Philips 2007). As I noted in the introduction, my main purpose in describing these two
analyses is to show how the location of a subset of data within a larger database both con-
strains and facilitates the kind of analysis that can be done with it.

Magistrate’s Court data

Overall, more data from the Magistrate’s Courts was recorded, transcribed, translated,
computer entered, and analysed than from any other setting. Approximately 32 hours were
recorded and 16 hours were processed in the field, with about half of the data coming from
the village court and half from the town court. The internal sequential structure of the pro-
cessed Magistrate’s Court cases was analysed. Of the 86 cases complete from charge to
sentencing, only what were considered the three more serious criminal charges, based on
sentencings, had a discourse unit within the sequence called in Tongan, ‘akonaki, ‘counsel-
ling’. In this moralising section defendants were told why what they did was wrong. These
three charges were theft, assault, and bad language (Philips 2004a).
Women only appeared in the Magistrate’s Courts as victims of bad language or assault,
not as perpetrators (except for one case of theft) or in any other court role such as magistrate,
clerk, lawyer, police prosecutor, or policeman. In bad-language cases men were described

143
Susan U. Philips

as yelling angrily in public, using bad language, swearing, or threatening women. In all four
of the cases I examined closely, and in others not examined closely, the brother–sister rela-
tionship was mentioned or heard by my research assistants as referring to the brother–sister
relationship (Philips 2004a, 2000).
Here is an example of the magistrate moralising about a bad-language case:

1) Ko e ‘ulungaanga palakū
That was a disgusting thing
Na’a ke fai
You did.
Na’a o kinaua,
They two went,
Ko si’i ongo faifekau
The two little missionaries
ke fakamafola
to spread the gospel
‘e fu’u kape mai.
And you swore repeatedly
‘Ikai ke ‘ilo na’a
You didn’t know
Ko homou kāinga.
If they were your relatives.
(Philips 2004a: 240)

The relationship between ‘brother’, tuonga’ane, and ‘sister’, tuofefine, is extended to cross-
gender cousins indefinitely (i.e. to what in English are called second cousins, first cousins
once removed, third cousins, etc.). The relationship has its own language ideology. One is
not supposed to engage in even allusive ribald scatological talk when brothers and sisters or
cousins are co-present out of respect for this relationship, and the siblings should likewise
not do so with each other, sometimes to the point of significant mutual avoidance.
In actuality none of my four closely analysed cases involved sisters and brothers, although
one involved distant cross-gender cousins. The other cases I did not examine closely, includ-
ing cases not processed, seemed to involve angry boyfriends yelling at their girlfriends in
public. Yet magistrates either told defendants that the bad language was wrong because
relatives and, more specifically, brothers and sisters, might have been present, or that it is
necessary to ‘stay mutually respectful’, nofo feulu’ufi, which my assistants heard as a direct
reference to the brother–sister relationship. These young adult female research assistants
said it is common for people to be exhorted to stay mutually respectful in public gatherings
and this expression is heard as specifically referring to cross-gender relations modelled on
the sister–brother relationship.
Since, Tongans explained to me, one cannot personally know all of the members of
the opposite sex with whom one is in this tabooed relationship, particularly when going
outside one’s own village, it is necessary in public to treat all as potential cross-gender
siblings – or, since the view seems more that of men than women, to act as if any woman

144
 The social ordering of gender ideology

could be a classificatory sister, and treat her like a sister. This then becomes the cultural
foundation for respectful cross-gender relations in civil society. Since the primary goal of
the Magistrate’s Courts is to maintain social order, and discourage disorder, this particular
morality that rests on the brother–sister relationship has a readily comprehensible logic.
Broad Tongan gender ideology conceptualises the brother as subordinated to the sister.
Examples of this subordination often involve the sister asking the brother for things that
belong to him, such as his crops, or pieces of his furniture. This means the sister female
identity, and not that of the wife or the sweetheart, is the most productive and prominent
identity to draw upon metaphorically in advocating positive treatment of women. This is
similar to the earlier example where fathers in the United States are asked how they would
feel if it was their daughters who were raped when trying to get support for more effective
legal treatment of rape cases.
The strength of the micro-analysis done on the four bad-language cases stems from the
close analysis of the entire Magistrate’s Court database and from the multiple occurrences
of bad-language cases where the brother–sister relationship is invoked. I consider it a weak-
ness that I was not able to interview the parties to the conflicts at issue. I would like to
have seen how they talked about the conflicts. I would also like to have obtained social
background information on the men and the women to determine whether bad-language
cases (and women taking men to court) are associated with particular life situations. Clearly
young women predominated in these cases and there is a view expressed by some Tongans
that disrespect to the woman disrespects the entire family and cannot be allowed to pass. I
could not, however, go beyond this in efforts to understand the cases.

Women’s work group data

The women’s work group, as noted earlier, was recorded in order to get a sample of women’s
conversation. Because the legal proceedings were the focus of the research and not wom-
en’s speech, only one work group was recorded in this project. However, it was considered
important and interesting enough to process in the field. The work group lasted four hours,
which means there was much less data to consider closely here than in the Magistrate’s
Court data of approximately 16 hours.
In this activity, called a koka’anga, ‘bark cloth work group’, 12 women in the village
were making a large sheet of tapa, ‘bark cloth’, about 70 feet long, out of many small pieces
brought by the women. Gifting of tapa is important on ritual occasions such as funerals and
college graduations, among other uses. As the women stuck the pieces together they sten-
cilled a design on them. Each month a different woman in the group took the final product
home as hers. In the paper I wrote about this activity (Philips 2007), I explained that the
women and their work symbolically invoked the powerful image of a traditional activity
that produced women’s wealth. At the same time their work was materially marginal in that
the value of the work product was not great relative to what the women could be paid if they
had work that paid an hourly wage or salary.
At the most general level, the activity consisted of alternations between conversation
and the singing of love songs. Because there were so many women in the group and a great
deal of thumping from applying the stain over the stencils and rolling over the bark cloth,
the conversation was difficult to process. We were still able to identify enough segments of
speakers and speech to get a general idea of their conversations. The love songs, sung by all
together, were much clearer.

145
Susan U. Philips

The love songs are usually mentioned by Tongans as composed by men to their girl-
friends, although women do compose love songs and love songs are sung in a range of
situations. They generally elevate the person addressed in the song and speak of the longing
and/or loss experienced by the composer/singer for whom the loved one is unreachable,
temporarily or permanently. The phrases of the work group songs themselves were gender
neutral, so could be imagined as referring to a same-gender relationship as well as a cross-
gender relationship.
Here is the transcription and translation of one verse and chorus of a love song from the
work group:

2) First verse
He uisa ‘e kuo kafo si ‘oku loto Oh my wounded heart cries out
He me’a vivili ko e māvae From this unending separation
He’uisa ‘e kuo kafo, si ‘oku loto Oh my wounded heart cries out
He me’a vivili, ko e māvae From this unending separation
Chorus
He ‘oiauā tangi ‘oiauā Oh cry oh
Kuo ‘ikai te u ‘ilo pe ‘e anga-fēfē I cannot understand a thing like this
He ta’e ‘alo’aloa si’i tokelau How could it have come from the gentle
northern wind
He ngahau kuo uhu si ‘oku mafu The arrow that pierces my heart
(Philips 2007: 60)

These songs are metaphorical, sweet, and saturated with tender romantic feeling when one
hears them on the radio or sung while a young woman dances a tau’olunga, a traditional
individual dance. In the context of the koka’anga, however, the songs were sung vigorously
with a heavy beat that matched the pounding and rolling of the tapa. The women, especially
the older women, called out reports that this or that other woman in their group was long-
ing for a man, or missed her man overseas, hinting at unlikely infidelity, and even point-
ing at one’s own private parts. They were imagining the words of the songs of longing as
coming from a woman, not a man. These remarks were greeted with loud raucous laughter
and understood as ways of teasing and joking with the older women who were accused of
dreaming of and longing for a man. There was also the added benefit of embarrassing the
younger women. Because the sweetheart relationship is supposed to be chaste, all the joking
and teasing offered a counterpoint that foregrounds sexual attraction and even the possibility
of infidelity. The same people are offering both the tender words and the bawdy commentary
on the words, providing two different visions of the sweetheart relationship, but with both
views treating it as outside of and separate from marriage. Here an advantage of this data
is that is allows the researcher to arrive at a different understanding of the gender ideology
provided by the women’s style of singing and commentary than could be gleaned from other
contexts where the genre of love songs is performed.
A key difference from the court research was that all the women in the group lived in
the village where I was living and where my research assistant had lived all her life. She
knew all the women in the group and her sister was a member, so I was able to obtain social
background information on the women in a way I was not able to do with the parties to

146
 The social ordering of gender ideology

court cases. This included their age, whether they were married, had children, what their
husbands did for work, and to some extent what the household composition of their homes
was. This data supported my claim that the women engaged in the work were relatively poor
compared to some in the village. They were also predominantly from a minority church
denomination that had been politically and economically marginalised for at least decades
and its members were thought to adhere to traditional Tongan ways more than members of
other religious denominations. The work group activity was in these specific ways socially
positioned.
There was a weakness in the analysis possible from this data because I only had recorded
the one work group, and I did not know how much I could generalise about what I witnessed,
whereas I was confident from the more substantial Magistrate’s Court data that I was docu-
menting a common phenomenon. I was fairly confident that I would see the same counterpoint
between words and the way they were sung among women of the same religious denomination
because denomination is, as I have already noted, predictive of a range of aspects of life. I do
not, however, know when sexual joking is appropriate across social domains.

Discussion
There are several general points illustrated by this comparison of two gender ideology
analyses. There are multiple distinctive gender ideologies in any given society. Gender ide-
ologies are context- and genre-specific, although they may have underlying principles in
common. In this case both the bad language and the sexual joking have scatological ele-
ments, but these are framed very differently. Understanding of a familiar gender ideology
can change when it is examined in more than one context. This was the case with the love
songs, which are no longer demure as cast by the singers in the work group. It was also true
of the bad-language cases in that the context of the Magistrate’s Court generalises the prin-
ciple of respect for the sister to all cross-gender public activities in a way I could not glean
from talk about real family lives.
Methodologically it would not be possible to carry out such comparative analyses of
this kind without recording different types of social activities. It would also not have been
possible without the broader ethnographic methodology that helped make sense of those
activities. Even then, however, the overall research agenda created constraints on the kinds
of analyses that were possible. That same agenda also made possible or potentiated certain
kinds of analysis.

Future directions
Ethnographic methodology is particularly important for cross-cultural studies of gender,
language, and sexuality. Given a commitment to the relevance of feminist critiques and
desires for change in all societies, there is a need for more research in non-Western cul-
tures where non-Indo-European languages are spoken. Methodologically the ability of the
researcher to anchor her claims empirically by reference to actual recordings of speech is
again particularly important in the study of other cultures.
I think of the research I have described here as ‘synchronic’ in that it treats all of the
Tongan data as happening at a given point in time, even though the research project took
place over two years. My own broader involvement with Tonga and Tongans in both Tonga
and the United States took place over more than a 20-year period and I saw many changes

147
Susan U. Philips

in Tongan society during that time. An important future direction for ethnographic methods
in the study of gender and language is the study of changes over time, or ‘diachronic’ stud-
ies. This is particularly important where there is a strong sentiment for change that directly
affects women’s well-being, as in many Western nations. For example, the recent #MeToo
movement in the United States involved strong demands for women to be free of sexual
harassment in the workplace. A business that has a poor reputation regarding its treatment of
women, yet asserts a commitment to change, can be studied ethnographically at two points
in time to determine whether and how change is taking place.
Ultimately, at the heart of this anthropological methodology is the idea of thinking com-
paratively to ask how the speech produced in one real-time activity is both similar to and
different from speech in another real-time activity.

Acknowledegments
The research described here was funded by the National Science Foundation Linguistic
Program. I thank the Government of Tonga for permission to carry out this research, and
the Honourable Geoffrey Martin, then Chief Justice of the Supreme Court, who oversaw the
court research. Among the many Tongans who supported this work I would particularly like
to express my appreciation for the many kindnesses of the late Vili Salakielu, Loukinikini
‘Ahio, Kalasi Salakielu, and Toti and Liliani Maile.

Further reading
Dick, H. (2018) Words of passage: national belonging and the imagined lives of Mexican migrants.
Austin: University of Texas Press.
Dick describes gender differences in how people in a Mexican city use talk about migration to the
United States. Methodologically the author shows how interview data can capture the same kinds of
gender representations heard in everyday talk.
Hoffman, K. (2008) We share walls: language, land, and gender in Berber Morocco. Malden:
Blackwell Publishing.
Hoffman shows how the political economy of southern Morocco takes men out of Berber mountain
villages for migrant labour in non-Berber towns while women stay in the villages. This labour
organisation is related to ongoing changes in gendered forms of song and speech that in turn also
organise Berber and Arabic languages.
Kulick, D. (1992) Language shift and cultural reproduction: socialization, self and syncretism in a
Papua New Guinea village. Cambridge: Cambridge University Press.
Kulick shows how the local village-level indigenous language is associated with negatively
evaluated women’s genres of speech. Tok Pisin, a language of wider communication, is associated
with positively evaluated men’s genres and the wider national political economy, resulting in language
shift.
Mani, L. (1998) Contentious traditions: the debate on sati in colonial India. Berkeley: University of
California Press.
This book documents multiple representations of widow-burning in written genres of discourse
drawn from diverse colonially relevant forms of social organisation. Europeans used the treatment of
women in colonial contexts to justify colonial cultural transformations.
McElhinney, B. (ed) (2007) Words, worlds, and material girls: language, gender, globalization. New
York: Mouton de Gruyter.
This edited collection show how gender is articulated with changing global economies through the
ordering of multiple linguistic varieties and forms of talk.

148
 The social ordering of gender ideology

Related topics
Interactional sociolinguistics: foundations, developments, and applications to language, gender, and
sexuality; text trajectories and gendered inequalities in institutions; the accomplishment of gender
in interaction: ethnomethodological and conversation analytic approaches to gender; semiotic
representations of women criminals; ethnography and the shifting semiotics of gender and sexuality.

References
Besnier, N. (1997) ‘Sluts and superwomen: the politics of gender liminality in urban Tonga’, Ethnos,
62, pp. 5–31.
Besnier, N. (2004) ‘Consumption and cosmopolitanism: practicing modernity at the second-hand
marketplace in Nuku’alofa, Tonga’, Anthropological Quarterly, 77, pp. 7–45.
Firestone, S. (2003 [1970]) The dialectic of sex: the case for feminist revolution. New York: Farrar,
Straus and Giroux.
Foucault, M. (1971) ‘The discourse on language’, in Archaeology of knowledge, translated by A. M.
Sheridan Smith. New York: Pantheon, pp. 215–237.
Frank, A. (1967) Capitalism and underdevelopment in Latin America. New York: Monthly Review
Press.
Hymes, D. (1964) ‘Introduction: toward ethnographies of communication’, American Anthropologist,
66 (6 Part 2), pp. 1–34.
Matoesian, G. (1993) Reproducing rape: domination through talk in the courtroom. Chicago:
University of Chicago Press.
McElhinny, B. (2014) ‘Theorizing gender in sociolinguistics and linguistic anthropology: toward
effective interventions in gender inequity’, in Ehrlich, S. Meyerhoff, M., and Holmes. J. (eds) The
handbook of language, gender, and sexuality, 2nd edn. Oxford: Wiley Blackwell, pp. 48–68.
Philips, S. (1998) ‘Language ideologies in powerful institutions’, in Schieffelin, B., Woolard, K., and
Kroskrity, P. (eds) Language ideologies: practice and theory. New York: Oxford University Press,
pp. 211–225.
Philips, S. (2000) ‘Constructing a Tongan nation state through language ideology in the courtroom’,
in Kroskrity, P. (ed) Regimes of language. Santa Fe: School of American Research, pp. 229–257.
Philips, S. (2001) ‘Cross cultural aspects of gender ideology’, in Smelser, N. J. and Baltes, P. B. (eds)
International encyclopedia of the social and behavioral sciences, vol. 9. Oxford: Elsevier Science
Limited, pp. 6016–6020.
Philips, S. (2004a) ‘The organization of ideological diversity in discourse: modern and neotraditional
visions of the Tongan state’, American Ethnologist, 31(2), pp. 231–250; reprinted in van Dijk, T.
(ed) (2007) Discourse studies, New Delhi: SAGE.
Philips, S. (2004b) ‘Language and social inequality’, in Duranti, A. (ed) A companion to linguistic
anthropology. Oxford: Blackwell Publishing, pp. 474–495.
Philips, S. (2007) ‘Symbolically central and materially marginal: women’s talk in a Tongan work
group’, in McElhinny, B. (ed) Words, worlds, and material girls: language, gender, globalization.
New York: Mouton de Gruyter, pp. 41–75.
Philips, S. (2014) ‘The power of gender ideologies in discourse’, in Ehrlich, S., Meyerhoff, M.,
and Holmes, J. (eds) The handbook of language, gender, and sexuality, 2nd edn. Malden: Wiley
Blackwell, pp. 297–315.

149
 10
Using communities of practice
and ethnography to answer
sociolinguistic questions
Ila Nagar

Introduction
My research focuses on what informs identity and how identity is formed in interaction. This
chapter provides a diachronic view of my interactions with jananas and engages with socio-
linguistic and ethnographic methods as reliable tools to study marginalised communities. In
doing so, I engage with my work with the janana community but the primary focus is the
theoretical and methodological frameworks that guided my research. I study a community
of men who self-identify as jananas, also called kotis, in Lucknow, India (Cohen 2005; Hall
2005). Jananas are assigned male at birth, fall in the lower rungs of the socio-economic ladder,
desire other men, possibly engage in sex work with other men, and simultaneously embrace
heteronormative sexuality (with their wives and families) and reject it (as jananas with other
jananas or with their male sexual partners). Thus, they occupy a complex space in the gender
and sexuality continuum. I have been working with jananas since 2003. An essential guide-
line for understanding sexualities in South Asian contexts is that sexuality goes beyond mere
sexual practice and marginalised people manage it in varied ways that are locally meaningful.
Standards with which we view some sexual identities might not apply to others and vice versa.
I use ethnography and communities of practice (Eckert and McConnell-Ginet 1992) as
methodological frameworks to study the janana community of practice (CofP). Ethnography
is defined as theoretical and methodological assumptions that guide a researcher in the field.
Ethnography as a theoretical framework for being in the field, writing about research findings,
and ways of collecting data has been widely used across disciplines and in language, gender,
and sexuality research (see Zimman and Hall 2016 for an overview of this research). CofP
is a theoretical framework that, within the study of language, helps researchers understand
motivations behind linguistic variation. My goal in this chapter is to highlight the implications
of using ethnography and CofPs as methodological frameworks in my work and to under-
line the collaborations between these two frameworks in sociolinguistic research. It has been
more than 25 years since Eckert and McConnell-Ginet (1992) fundamentally changed the way
we understood how language variation works and called for understanding local practices
to explain more global movements in language change. The CofP framework distinguishes
practice from activity, and practice is understood as an aggregate of aspects of practice. These

150
 Communities of practice and ethnography

are practice as meaning, practice as community, practice as learning, practice as boundary,

and knowing in practice (Wenger 1998: 54). For the CofP framework the concept of practice
implies a deeper engagement with the community which goes beyond simply participating
in activities together (Eckert and McConnell-Ginet 2007: 28) and extends to being incorpo-
rated in practices to the extent of being defined by them and also defining them. While early
sociolinguistics mapped language use onto gender, class, socio-economic status, occupation,
and education, sociolinguistic theory in the last two decades has moved towards investigating
local practices that guide specific language usage. Eckert and McConnell-Ginet (ibid.) brought
the communities of practice framework into linguistics from the work of Lave and Wenger
(1991), which was a step towards understanding how people learn in certain situations and
how this learning is a social process, which is a matter of participation in communities of prac-
tice. Eckert and McConnell-Ginet demonstrated that identifying communities of practice has
an application for sociolinguistics. They proposed looking at language use and social indica-
tors as informing one another. By tracing language variation to the local realities of members
of communities, it is possible to understand how local practices govern large-scale language
variation. The benchmark of the framework is social practice, and the ways in which regular
practices influence language use. The framework suggests that language variation happens
within communities of practice which are ‘aggregates of people who come together around
a mutual enterprise’ (Eckert and McConnell-Ginet 1992: 464); something that brings mem-
bers together on a regular basis for a practice that changes their interaction with their world.
Engaging in this community of practice/mutual enterprise leads to specific forms of learning.
According to this framework, an individual’s identity is constructed in terms of what they
experience in their communities of practice and how they (re)present the experience.
This chapter accomplishes the following: (1) it defines the communities of practice
framework; (2) it shows how ethnography and understanding meaning-making within com-
munities of practice go hand in hand; and (3) it demonstrates why the communities of prac-
tice framework in spite of some of the critiques of the framework (Angouri 2018; Davies
2005; King 2014) was an appropriate tool for me to understand meaning in the janana
community. In the next section, I outline major works that inform our understanding of com-
munities of practice and ethnography. I then discuss how, as sociolinguistic researchers, we
can use CofP to find nuances in the gender and sexuality continua and language use in cul-
tures that we do not know well or in cultures to which we belong. Cultural anthropologists
understand culture as historically specific meaning-making. For CofP, culture and meaning
are tied to one another. As Wenger (1998: 54) writes, ‘The negotiation of meaning is a pro-
ductive process, but negotiating meaning is not constructing it from scratch. Meaning is not
pre-existing, but neither is it simply made up. Negotiating meaning is at once both historical
and dynamic, contextual and unique’. I explain how the nuances in gender and sexuality
which are situated in culture allow us to comment on how language use becomes meaning-
ful in communities. I use examples to highlight how the communities of practice framework
and ethnography inform my findings. This chapter addresses questions about identifying
cultures in communities, finding identity in interaction, and how to make local meaning
speak to global ideas about culture, gender, and language use.

Communities of practice and ethnography

A CofP serves as a theoretical tool to determine how people use language to understand
their own place within larger structures. The janana community is not clearly defined by
any boundaries; by which I mean, one cannot locate them by a common space, profession,

151
Ila Nagar

cause, or other commonalities that one can use to identify a community. Nonetheless,
jananas come together around mutually defined practices, and that is what makes them a
CofP. These defined practices are what need to be ‘found’, and ethnography seems to offer
the other fundamentally important tool in helping ‘find’ these practices.
The CofP framework has been used by scholars to understand language use and to inves-
tigate identity in social practice. Moore (2006) uses analysis of narratives to show how girls
in Midland High School negotiate meaning and use seemingly simple social practices like
dance to symbolise their affinity with a particular social group. Mendoza-Denton (2008),
Bulcholtz (1999), and Eckert (2000) have all found similar meaning in social practices
unique to the groups that they studied in high schools and communities formed by teenagers
around school activities. An entire issue of the journal Language in Society was dedicated
to identifying and analysing the usefulness and applicability of the framework. In this issue,
scholars found that the framework could be useful in explaining language use (Bucholtz
1999), in seeing categories such as nerd girls and the influence of outside communities
of practice on the experience of pregnant women (Freed 1999), and how certain language
uses cannot be explained by the community practice framework (Holmes and Meyerhoff
1999). While the scope of the framework has been tested in various communities across
the world, scholars have also debated limitations of the framework (Davies 2005). Eckert
and Wenger (2005) have responded to the critique, Eckert and McConnell-Ginet (1999)
have cautioned against using the framework in situations where it might not apply, and
others have noted that defining exact practices should be a given for scholars using the
framework (King 2014). Scholars have also pointed to macro-concepts ‘such as taken for
granted presuppositions about appropriate cultural behaviour’ (Holmes 2018: 34) that can
influence interactions between people. An understanding of the full extent to which social
factors and cultural norms influence linguistic behaviour cannot be achieved without study-
ing local meaning and deciphering what influences minute speech and behavioural patterns
in a community.
A CofP is defined by learning and practices. It is not enough to say that a community of
people living together or occupying spaces where they meet and share practices constitutes
a CofP. These practices define how a community works in response to the world around
it and the parameters of community interaction. These parameters must be articulated by
members of the community for themselves and as a negotiation with the broader structures
within which they operate to understand the interactions and reactions of a CofP. To say that
a community is a CofP without explaining the practices that make it a CofP is not productive
to the framework. When it comes to defining motivations behind specific linguistic choices
like showing why exactly variation in language use might happen in some contexts and not
in others, it is important to highlight practices of a CofP. Ethnographic method is essential
in coming to know a community and its motivations. The CofP framework relies on finding
local meaning and, by default, propagates ethnographic methods. O’Reilly (2009: 3) writes:
‘Ethnography draws on a family of methods, involving direct and sustained contact with
human agents, within the context of their daily lives (and cultures), watching what happens,
listening to what is said, and asking questions’. Ethnographic methods assume significant
time spent with a community, and presenting a written ethnography is an exercise not just
in reporting what the researcher learned in the field but is a way to represent social life in its
complexity and nuance. How can one find meaning in a community without rigorous study
of practice in the community? Therefore, in my view, it is unlikely that one can study mean-
ing and practices in a community without collecting data with ethnographic methodology
in mind.

152
 Communities of practice and ethnography

Strathern (1991) points to the complexity of practice and suggests that any practice
within a community can be seen from perspectives that make it more and more complicated.
Practice is a process, and ethnography engages with it in ways that make the presentation
of practices true to whatever level possible. The complexity of practice makes the task of
a sociolinguist working with the communities of practice framework critical since no one
method of data collection is enough to produce results that can be considered sufficiently
transparent. In a way, recognising that only partial connections between practice, commu-
nities, and their context can be made is a powerful notion. As we address questions of
language use in context within communities, we can look for perspectives that come from
ethnographic fieldwork and address their meaning to members of the community. My work
with jananas included ethnographic methods such as participant observation, one-on-one
interviews, and focus group interviews. Working with the janana community using ethno-
graphic methods has shown me varied perspectives on the community which come from the
participants who give multidimensional perspectives in what they say about their lives and
how I saw their lives in Lucknow. This enhances the richness and diversity of the findings
and tells me that jananas are a CofP. Questions of meaning within particular practice also
need to be explained within the specific context of broader social configurations like patri-
archy, casteism, racism, sexuality, or class. Our job as sociolinguists has to be understood as
negotiators of these four separate spheres: ethnography, communities of practice, research
questions, and the place of a specific community of practice in its own global context (Nagar
2019). A discussion about ethnography and working with a specific community of practice
cannot be complete without also establishing the researcher’s role, positionality, and the
power dynamics their positionality creates within the community that they study. According
to Strathern (1991: 27):

I wish to suggest a third way of personifying the ethnographic experience, to draw a fig-
ure who seems to me more than one person, indeed more than a person. What happens
‘takes place’ because it happens somewhere, in the presence of others, because events
become interventions, the subjectivity of different persons the issue.

Strathern complicates the picture of ethnography. Representation, the ethnographer, and

writing an ethnography inform how one should look at and represent a community of prac-
tice and answer sociolinguistic questions. Local ethnographic findings within a CofP can be
indexical of wider social factors, power interests, and conflicts. When a researcher is work-
ing to understand language variation s/he cannot make connections between language use
and a community of practice without also addressing her own positionality, and issues of
practice and power that the community engages in to respond to the world – task that only
ethnography can answer.
In the next section, I discuss examples from my own research that highlight the com-
plexity of being a member of a community and how this is manifest in linguistic choices
members make.

Background and method: jananas and their lived experiences

Jananas live in transitional spaces. They are lower-middle-class to lower-class people with
uneven incomes and uneven prospects of income. Many jananas marry women and fulfil
responsibilities that come with being a man in their cultural context: providing financial
support to parents, siblings, nieces and nephews, children, wives, and sometimes siblings’

153
Ila Nagar

families. Since income is uneven for members of this community, supporting families finan-
cially is a constant struggle. Many jananas use the money they earn from sex work to sup-
port families. In spite of their admissions of masculine roles, jananas say that they are not
men, they are not women, and they are not hijras.1 In fact, they place themselves in all of
these categories and none of them, depending on their context. Being a janana means being
fluid with gender roles and subjectivities.
My interactions with jananas started in 2003 when I was a graduate student starting
a research project on language and gender. I knew nothing about this community. As I
was looking for a dissertation topic, a contact at a non-profit organisation in New Delhi
suggested that I visit Lucknow and meet with jananas since my interest was language. I
went to Lucknow from New Delhi without any information about what I was going to do.
I established contact with a non-profit organisation that worked in the HIV/AIDS sectors
for at-risk men and met jananas first through the non-profit organisation and then through
jananas I had initially met. My participant observation and interviews with jananas usually
happened in three spaces – the non-profit office, city parks where jananas ‘hung out’ and
some homeless jananas lived, and in spaces where they invited me, usually monuments
around the city, a quiet street, or, in some instances, their homes. The jananas I interviewed
presented a range within the lower socio-economic classes. The examples I present below
show how I used my ethnographic data to analyse jananas as a community of practice to
answer sociolinguistic questions. The examples come from two different points in my field-
work with jananas who were rather different from each other yet inhabited the same margin-
alised social space. As is often the case with ethnographic research, my research questions
changed over the years and with the amount of time I spent with jananas in Lucknow. I
started working with jananas in 2003 when my research question was primarily about the
Farasi: a secret code jananas use in situations that need discretion. By 2015, I was primarily
interested in the relationship between legal discourses and how something not necessarily
experienced by jananas on an everyday basis (specific laws concerning homosexuality)
affected janana subjectivity and ways in which this was reflected in language use.
While my questions changed over time, my methodology remained consistent. I started
noticing dimensions of a CofP in the janana community early in the process of fieldwork.
In the community, there was mutual engagement, joint enterprise, and a set of negotiable
resources that members of the community shared. Mutual engagement entails that partici-
pants interact with other participants on a regular basis. According to Wenger (1998: 85),
‘Over time, the joint pursuit of an enterprise results in a shared repertoire of joint resources
for negotiating meaning’, and joint enterprise is, ‘not just a stated shared goal but a negoti-
ated enterprise, involving the complex relationships of mutual accountability that become
part of the practice of the community’ (Holmes and Meyerhoff 1999: 175). For the janana
community, mutual engagement was establishing a certain kind of relationship with hijras
who could be perpetual sources of support or physical and verbal violence to the janana
community, or both. It was also establishing a way to engage with family; to hide their
janana status while maintaining all responsibilities of being a family man in some cases. It
was finding a way to be janana, and this translated into finding other jananas and being with
them to learn to be janana or mentoring younger jananas to be janana. Joint enterprise was
sharing in the marginalisation that came their way. Reactions to the marginalisation were
also shared. Negotiable resources included using gender marking in Hindi in a specific way,
giving each other janana names, specific swear words, clapping to signify group belong-
ing or anger, and using the special register, Farasi. The question for me was not whether
jananas were a community of practice but why it was relevant to my research aims to show

154
 Communities of practice and ethnography

that they were and how showing that they were a community of practice strengthened my
arguments and conclusions about this community. Showing that jananas were part of a CofP
that was organised in a certain way with dimensions that corresponded with what scholars
had suggested as essential for the functioning of a CofP was important because jananas
were not organised or recognised in an institutionalised way. They were not in a school,
they were not learning a trade, there was much fluidity in who was janana when, and there
was no single way in which jananas defined themselves. Yet, they were a CofP because
they shared practices, unique to their community. The mutual enterprises of accountability
to family, which for jananas can be a burden and a challenge, takes the form of hiding from
family, running away from one’s family, or negotiating family obligation and the jananas’
non-heteronormative desire. Accountability to family contradicts initiating and helping
young jananas to become janana. Living as a janana becomes a process that is deeply tied
to maintaining value systems initiated by the family that are contradicted by jananas’ own
desires. The fact that most jananas walk these thin and often dangerous lines makes their
community a community of practice. However, elaborating on why jananas are a CofP does
not necessarily achieve a research goal. The assumption of a community of practice is that it
can change and influence practices – it can create meaning that is relevant only to the com-
munity and as a response to the world (Eckert and McConnell-Ginet 2006). The idea that
identity is constructed discursively has been explored by many scholars of language, gen-
der, and sexuality (Bucholtz and Hall 2004, 2005; Eckert 2000; Mendoza-Denton 2008). In
the next section, I explore three ways in which jananas express their identity: gender mark-
ing in Hindi, defining the boundaries of gender and where jananas fit, and the parameters of
being a janana. My ethnographic work shows that jananas are a community of practice who
learn to use language and other means to define their interactions with the world.

Three cases: the analysis

In all the years that I interacted with jananas in Lucknow, Neerja was never too keen on
chatting with me, but I could always see her in the sidelines smiling at me or some other
jananas. I knew she was a janana, and she was friends with many of my informants. In
2006, I had been seeing Neerja in and around the non-profit office where she worked for sev-
eral years, and I asked her if I could interview her and ask questions about her life. She said
‘yes’ but did not come at the time of the interview. I asked her why, though she did not give
me a direct response. I understood her to be shy of the recorder or not wanting to be inter-
viewed. In 2013 when I returned to Lucknow after a hiatus of five years, Neerja approached
me to be interviewed. Neerja had had a rather typical experience for a janana in Lucknow.
She was from a lower-middle-class family, was literate and had a job at the non-profit office,
was admonished and eventually abandoned by her family because of her sexuality, had
friendships in the janana community (many of which were decades old), had a regular male
partner, and often encountered violence. My questions for Neerja centred on her interactions
with the law, with other jananas, and her experiences as a janana. Excerpt 1 below is from
an open-ended interview that started with a question about Neerja’s boyfriend. She told me
stories of heartbreak, cheating, and abandonment and how bad the past year had been. That
lead me to the question about how she was now. Here Neerja spoke about a deep sense of
loneliness she feels in spite of her friendships, access to sex, and making a life for herself.
While the excerpt reveals much about Neerja’s life and expectations, the last few lines of
this interview indicate a subjectivity that Neerja has carved for jananas – they are even
weaker than women, ‘ladies’, when it comes to matters of the heart. In the distinction that

155
Ila Nagar

Neerja makes, jananas are not part of a masculine identity nor a feminine identity. They
have their own place on the continuum.

Excerpt 1.
1. I: Thoda sa better feel kar rahe hain aap? I: Are you feeling a bit better now?
2. N: Haan, ab to pehle se thoda N: Yes, now I am a little
3. better feel kar rahen2 hai. Matlab apna hai. better. I mean it is mine,
4. Apni zindagi hai it is my life.
5. Hum kha rahen hain, I am eating.n
6. pi rahen hain, reh rahen hain. I am drinkingn, I am living.n
7. Fir bhi yeh hai ki man me Even so in my heart,
8. ek adhoorapan sa there is some emptiness
9. hamesha rehta hai ma’am that is always there.
10. Pata nahi kya? I don’t know what?
11. Aisa lagta hai kisi cheez ki It feels like I am looking for
12. hume talaash hai something,
13. jo nahi mili hai something I can’t find.
14. Aur jaane kya shayad wahi dhoondne And perhaps I am searching
15. ke liye hum sochte hain. and thinking about it.
16. Kotiyon me baithten hain I hang out with kotis (janana)
17. Giriyon se milten hain I meet with giriyas [boyfriends or
clients].
18. Ma’am, hota kya hai ki Ma’am, what happens is that
19. giriya aate hain. Mil jaate hain the giriyas come. We meet.
20. Khair wo to baat alag hai koi Although that is a different matter.
21. kabhi sex bhi ma’am ho jata hai. Sometimes I have sex too, ma’am.
22. Hota hi hai It happens.
23. Jaise ki ma’am poori kisi ke saath It is like, ma’am, I am not able to
24. ban ke nahi reh pate hai get along with anyone.
25. Jane kya koi kisi ke liye I don’t know why, but time
26. waqt khara nahi utarta hai does not fit with anyone.
27. Aur hum log bahut And people like me [us] are
28. naazuk hote hain, ma’am. very delicate, ma’am.
29. Ladies se bhi, ma’am. We are more delicate than ladies.
30. Phir meri soch hai, ladies I think that ladies
31. apne aap ko control kar lengi control themselves,
32. lekin hum log bilkul agar, but when something happens
33. hum log ko dhakka lage, when we get hurt
34. to hum to toot ke ekdum we break apart.
35. bilkul bikhar jaate hain We completely come apart
36. bilkul kamzor ho jaate hain we become very weak.

The excerpt from Neerja that I present above is an example of an average janana interaction
with themselves and with the world around them. Neerja’s feeling of being different from
men and women that she talks about in Lines 25–36 are part of her being a janana. The idea
that Neerja (or any other janana) is not a man or a woman and is emotionally weaker than

156
 Communities of practice and ethnography

women,3 a belief that many jananas hold is part of the practice of being a member of the
janana CofP. I learned from repeated interactions and time in cruising areas that jananas
across the age spectrum hold the view of emotional weakness of jananas or even women.
While it did not become a theme for my work, I learned that specific perceptions about being
janana were learned and circulated among jananas. Ethnographic methods support finding
these imperative details about any community of practice.
The second example I present is from an interview with Rajnidevi from 2004. I met
Rajnidevi in 2003 and continued contact with her until her passing in 2012. Rajnidevi
was a pacci4 janana who was about 50 when I first met her and was a lower-class janana
who spent considerable time in hijra households. Poverty was constant in Rajnidevi’s life
and defined much of her interactions. Rajnidevi did not have a regular job and her income
depended on what she was able to do for hijras or sex work. Rajnidevi engaged in sex work
and worked as an agent or pimp for other younger jananas who could not find clients on
their own. She was also often a senior member in janana spaces and was given respect due
to her senior status.
Excerpt 2 below is from an open-ended interview where I asked Rajnidevi questions
about her interactions with younger jananas. Rajnidevi’s response was centred on telling
jananas not to be janana. She says that the first advice she gives to jananas was to not
come into the profession, which was sex work. However, her advice was not just about sex
work, it was also about being janana in general and bettering themselves, specifically by
not being janana. Since being janana is not a choice for any janana, the response to this
type of discourse from seniors is usually nothing. Young jananas keep coming to the areas
where jananas congregate and keep up their interactions with other jananas and clients and
boyfriends. Jananas like Rajnidevi also know this and assume that anyone who wants to be
a janana will be one; they just give advice on the off-chance that things might change for a
janana. For Rajnidevi, after it is established that a young janana wants to stay a janana, the
learning and practice of being janana begins. This practice includes ways to navigate public
spaces, find clients, interact with hijras, and other modes of good janana behaviour. While
there are excerpts of interviews in this section that support my findings, no one interview or
cruising area visit or observation led me to arrive at my conclusions about the janana com-
munity. An aggregate of interactions using ethnographic methods can facilitate arriving at
conclusions about why a community of practice is a community of practice.
Rajnidevi also adds information that defines janana relationships with the janana com-
munity. She suggests that she does not interact too much with kade taal jananas. Kade taal
jananas are jananas who keep their janana status hidden from the world as much as they
can. They come to cruising areas but do not necessarily participate in hanging out with other
jananas or any other forms of social interactions with jananas or hijras. Kade taal jananas
do not have the same level of interactions with jananas like Rajnidevi. Within janana cir-
cles, jananas who are able to be kade taal successfully, that is, be janana and be a man and
do both contextually and without one persona overlapping with the other, are considered to
be living in the best of both worlds. While many jananas try to achieve this goal, it does not
come easy and many fail. Rajnidevi suggests a certain level of distaste about being janana,
shares information about what kind of knowledge she imparts to new jananas, and finally
suggests that if someone wants to be hidden as a kade taal janana and does not want to pro-
claim a janana identity, she lets them be. The self-deprecation that Rajnidevi shares in this
excerpt about being janana surfaces in many other jananas’ stories.

157
Ila Nagar

Excerpt 2.
1. I: Aap choti kotiyon ko kya salah dete hain I: What advice do you give to
younger kotis?
2. R: humare paas kotiyan aatif hain R: The kotis that come to me,
3. to pehle to hum kotiyon se yahi kehten hain first I say to the kotis,
4. ki beta is kooche main na aao, ‘Child, don’t come in this place.
5. is kooche main kuch rakha hai nahi, There is nothing here.
6. ye narak ki zindigi hai. This is the life of hell.
7. Agar humare samjhane se maan jaye If they understand what I say
8. to zyada better hai, then it is better.
9. agar samjhane se nahi mantif ho If you don’t understand,
10. to vo tumhare upar hai. then it is up to you.
11. Agar ana chahtif ho to aao If you want to come,
12. agar nahi ana chahtif, if you don’t want to come,
13. sambhalna chahtif ho to sambhal jao. if you can better yourself, better
yourself.’
14. Jo nahi sambhaltif hai The ones who don’t better
themselves
15. ana chahtif hain to phir who want to come in this, then
16. unko sari baatein batani padti hain. we have to explain everything to
them:
17. Inko ye kehte hain, ‘This is called this.
18. inko ye kehte hain. This is called that.
19. Inse hoshiyar rehna, Be careful of this,
20. police walon se hoshiyar rehna. be careful of the police.’
21. Har baatein unko samjhaya jata hai We have to explain everything to
them.
22. Bahut si kotiyan hain jo apna There are many kotis who
23. kade tal main rehtif hain, live in kade.
24. jananiyan mili, unse baat kiya, They meet jananis, talk to them
25. uske bad kehtif hain ki, ‘kade ho jao’ then they say, ‘Be kade.’
26. apna…phir hum My…then I
27. un kotiyon se milna bhi don’t even like
28. nahi ichchuk karaten to meet those kotis.
29. hain ki bhai tum apne ko The thing is, if you want to be
30. band roop main rehna chah rahif ho hidden form,
31. to band roop main raho na khulo then stay hidden form.
32. to zyada achcha hain That is better.
33. tumhara jaise dhanda chale However you can make your work,
34. waise achcha hai. that is good.

The third example is from a conversation with Imrana from 2004. I met Imrana very early on in
fieldwork and continue to interact with her to this day. The last time I met and interviewed her
was in 2015. Imrana is a kade taal janana who lived with her family which comprised of two
brothers and their families, five nieces and nephews, a mother, and one sister. Imrana took on
the responsibilities of a son in the household: providing financial help wherever possible and
being available to take care of the needs of the family. Imrana was in her late twenties when I

158
 Communities of practice and ethnography

met her and she had a story fit for a novel. She ran away from her home when she was still a
teenager and joined a hijra group. She danced with hijras, went around North India Hindu and
Islamic religious sites, worked in the Middle East, and now lives a simple life as a kade taal
janana who does not engage in sex work and mentors young entrants into the community. The
conversation below was between Imrana and me in the presence of another janana in the non-
profit office. The janana who was present during the conversation was Imrana’s friend, and
I knew her well. In many interviews I had heard jananas talk about ‘bigadna’, a Hindi verb
which means ‘to go bad’/’to rot’/’to be spoiled’. It is used in reference to food that has gone
rotten or people who have chosen paths that do not coincide with ‘proper’ behaviour. Jananas
use ‘bigadna’ to tease each other about sexual misconduct, comparing each other’s degree of
‘bigadna’; they use it to describe their coming into sex work (like Imrana does below), and as
a way to establish camaraderie. I was curious about why jananas used this word for each other
because of the negative meaning associated with the word. My question for Imrana was about
the choice of this particular word for specific janana choices. She did not answer this part of
my question but did explain her own process of ‘bigadna’. She has a janana heart, she says,
meaning desire for men and for women’s work, such as cooking, cleaning, other household
work, for professions like sewing, and for feminine products, such as makeup and clothing.
With a janana heart, Imrana found a janana friend and explored Lucknow by roaming around
and finding other jananas and men interested in jananas.

Excerpt 3.
1. Ila: aap log jaise kehte hain na, Ila: Like you guys say,
2. hum tab bigde, to ‘That is when I got spoiled.’
3. bigadna kyon kehte hain? Why do you say ‘spoiled’?
4. Imrana: bigadna ka matlab ye hua Imrana: Being spoiled means…
5. jaise hum apko bataien, Now let me tell you,
6. jaise hum chauda saal ki umar ke then, like I wasn 14 years old
7. hamara dil janana tha. my heart was janana.
8. humne samaj main ye nahi I had not seen
9. dekha tha ki dhanda kahan hota hai, where sex work is done,
10. janane kahan baithte hain, where jananas hang out.
11. matakne chatakne wale People who sway and move hips –
12. kahan baithte hain. where these people hang out.
13. Ab ye hamare mohalla ki taraf Now this one came to my
neighbourhood,
14. se guzre, janana matakta chatakta hua, a janana swaying and moving hips.
15. Humne inhe dekha, I saw him,
16. inse salam dua kari. I greeted him.
17. Humari inki dosti aa gayi, We became friends.
18. hamari bhi adat ban gayi I also got into the habit of
19. inke saath ghoomne ki. going out and roaming about.
20. Phir ghomte ghoomte hum bhi After roaming about for a while.
21. parko main jakar baithnen lagen, I startedn sittingn in parks.
22. hum inke saath ane lage jaanen lagen I startedn comingn and goingn with
this one.
23. baithnen uthnen lagen. I startedn hangingn out with him.
24. Humne dekha samaj. I saw the world around me.

159
Ila Nagar

25. Jab yahan se dekha dhanda hota hai I saw where sex work was done,
26. to humne bhi shuroo kar diya. so I also started.
27. Isko kehte hain hum This is called,
28. yahan se ab bigad gaye. ‘we got spoilt from here.’

While on the face of it the three examples I present in this section are parts of larger conver-
sations I had with Neerja, Rajnidevi, and Imrana, and do not give us a full picture into their
lives, they give us relevant pieces of information about how jananas see the world around
them. All three jananas in this section were jananas telling me about the consequences of
their sexual desires. Neerja and Imrana were kade taal jananas while Rajnidevi was a pacci
janana. Yet, there were other differences too. Neerja and Imrana, while they were both kade
taal, had very different family pressures. Neerja lived alone, had lived with a boyfriend at
one point and had no financial responsibility for her family. Imrana lived with her family,
had contributed significantly to the well-being of her family both financially and in terms
of family prestige, and had a very different outlook on what it meant to be janana. While
Imrana, Neerja, and Rajnidevi had uneven incomes and no regular sources of money, Imrana
felt financial pressures that were different from Neeraj’s, and Rajnidevi was often reliant on
others to provide for her food. For Imrana being a janana was an indulgence, for Neerja, it
was a life, and for Rajnidevi it was a life choice. Neerja’s example primarily points to cat-
egories of gender that Neerja has created and where she finds jananas. For Neerja, jananas
do not fit the male or the female category. They are something else, even more fragile than
women are. Rajnidevi’s example informs us about some relevant parameters in the janana
community of practice. We see that senior members engage new and younger members and
can actively engage in teaching them about being janana. Imrana’s example also gives us a
glimpse of how exactly members of this community come to understand membership and
become members. It is a matter of knowing someone who has or understands a janana heart
and is willing to show a younger janana the ropes of being a janana.
While the primary pieces of information from these examples are easy to extract, the
ethnographic task is identifying the key insights in these brief excerpts as members of the
janana CofP. The specifics that these examples present are not easy to parse if one does
not have a complete picture of the community. It is the task of the ethnographer through
participant observation, interviews, and interactions to decipher these categories. The CofP
framework, while based in practice and learning, cannot be utilised fully if issues of power,
hierarchy, and social structure within communities of practice are not uncovered. This is
why it is pertinent that an ethnographic approach be written into finding and commenting
on structures of language change or variation in communities of practice.

Conclusion
In this chapter, I have established what the communities of practice framework is meant to
do, I have shown why ethnographic methods are central to finding meaning within com-
munities of practice, and I have shown why the communities of practice framework is a
highly appropriate tool for me to understand meaning in the janana community. Yet, one
question that remains to be answered is, how does suggesting that meaning is created and
maintained within a community of practice help us to understand why community members
create the meaning that they do? Why is Neerja’s response to abuse a creation of a category
of gender that in her world is weaker than women in its response to abuse? Why is it that
jananas see themselves as people who engage in ‘bigadna’, a quality that is not celebrated

160
 Communities of practice and ethnography

outside the janana circles but can have some positive connotations for jananas? Why is it
that Rajnidevi needs a certain amount of openness from a janana before she lets them into
her life? Each of these boundaries that these jananas have created are responses to their
ways of dealing with the world. These can only be understood in the context of this CofP
and how it is placed in the broader context of Lucknow and in the sexuality continuum. This
CofP comes into focus only by engaging with it through ethnographic methods.

Future directions
Introducing an emphasis on local realities within an understanding of community-based
practices can help us flag blind spots in our interpretations of how language, gender, and
sexuality can be connected. Local practices can be hard to identify. A collaboration between
CofP and ethnographic methods along with an understanding of some of the challenges and
their remedies that methods can pose (see Angouri 2015, 2018; Holmes 2018; King 2014 for
a comment on CofPs and; Banerjea 2014; Long et.al. 2009; Nagar 2014 for a comment on
ethnography) can enhance a researcher’s understanding of their own data. Future directions
in work on CofPs and how they influence language change should see language variation as
a result of localised realities. Local realities can be situated in marginalisation and can help
us understand structures within communities or structures of which communities are a part.
Ethnographies conducted over periods of time are fundamental to these pursuits.

Notes
1 Hijras are a transgender presence in India who also have ritualistic and religious roles, for instance
the blessing of newborn children. For more on hijras see Reddy 2005, Hall and O’Donovan 1996.
2 Hindi has grammatical gender. The verb in Hindi agrees with the nominative argument. The gender,
person, and number is indicated on the verb morphologically in past and present perfect tenses, future
tense, past habitual, and past, present and future continuous. In case of compound verbs, gender and
number is indicated on the all delexicalised verbs in all tenses. Gender is also marked on some adjec-
tives, pronouns, and some postpositions. Jananas vary gender marking as their context changes. In the
examples, I use superscript m (for masculine), f (for feminine), and n (for neutral) to signify which
gender marking a janana is using. Neutral is a dialect feature of the dialect of Hindi spoken in Lucknow
where the second person plural conjugations, which are not marked for gender, are used in first person.
3 This sentiment is alien to us as a Western audience and speaks to the obvious sexism of the system of
which jananas are a part.
4 A pacci janana (as opposed to a kade taal janana) is more open about their status as a janana and
about their sexuality. While pacci jananas can live with family and support families, many of them
live alone or with boyfriends and other jananas. Pacci jananas engage in sex work more openly than
kade taal jananas.

Further reading
Eckert, P. (2012) ‘Three waves of variation study: the emergence of meaning in the study of
sociolinguistic variation’, Annual Review of Anthropology, 41, pp. 87–100.
Provides a history of sociolinguistic methods and explains why the third wave is important for
understanding linguistic variation.
King, B. (2014) ‘Tracing the emergence of a community of practice: beyond presupposition in
sociolinguistic research’, Language in Society, 43(1), pp. 61–81.
Provides an analysis of the pitfalls of using the communities of practice framework.

161
Ila Nagar

Related topics
Anthropological discourse analysis and the social ordering of gender ideology; gender, language
and elite ethnographies in UK political institutions; interactional sociolinguistics: foundations,
developments and applications to language, gender and sexuality; identity construction in gendered
workplaces; gender, stance and category-work in girls’ peer language practices.

References
Angouri, J. (2015) ‘Online communities and communities of practice’, in Georgakopoulou, A. and
Spilioti, T. (eds) The Routledge handbook of language and digital communication. London:
Routledge, pp. 323–338.
Angouri, J. (2018) Culture, discourse, and the workplace. London: Routledge Press.
Banerjea, N. (2014) ‘Critical urban collaborative ethnographies: articulating community with Sappho
for equality in Kolkata, India’, Gender, Place and Culture, 22, pp. 1058–1072.
Bucholtz, M. (1999) ‘“Why be normal?’’ Language and identity practices in a community of nerd
girls’, Language in Society, 28(2), pp. 203–223.
Bucholtz, M. and Hall, K. (2004) ‘Theorizing identity in language and sexuality research’, Language
in Society, 33(4), pp. 501–547.
Bucholtz, M. and Hall, K. (2005) ‘Identity and interaction: a sociocultural linguistic approach’,
Discourse Studies, 7(4–5), pp. 585–614.
Cohen, L. (2005) ‘The Kothi wars: AIDS cosmopolitanism and the morality of classification’, in
Adams, V. and Pigg, S. L. (eds) Sex in development: science, sexuality, and morality in global
perspective. Durham, NC: Duke University Press, pp. 269–303.
Davies, B. (2005) ‘Communities of practice: legitimacy not choice’, Journal of Sociolinguistics, 9(4),
pp. 557–581.
Eckert, P. (2000) Linguistic variation as social practice: the linguistic construction of identity in
Belten High. Malden, MA: Blackwell Publishing.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Think practically and look locally: language and gender as
community-based practice’, Annual Review of Anthropology, 21, pp. 461–490.
Eckert, P. and McConnell-Ginet, S. (1999) ‘New generalizations and explanations in language and
gender research’, Language in Society, 28(2), pp. 185–201.
Eckert, P. and McConnell-Ginet, S. (2007) ‘Putting communities of practice in their place’, Gender
and Language, 1(1), pp. 27–37.
Eckert, P. and Wenger, E. (2005) ‘What is the role of power in sociolinguistic variation?’, Journal of
Sociolinguistics, 9(4), pp. 582–589.
Freed, A. (1999) ‘Communities of Practice and Pregnant Women: Is There a Connection?’, Language
in Society, 28(2), pp. 257–271.
Hall, K. (2005) ‘Intertextual sexuality: parodies of class, identity, and desire in liminal Delhi’, Journal
of Linguistic Anthropology, 15(1), pp. 125–144.
Hall, K. and O’Donovan, V. (1996) ‘Shifting gender positions among Hindi speaking Hijras’,
in Bergvall, V., Bing, J., and Freed, A. (eds) Rethinking language and gender research: theory
practice. London: Longman, pp. 228–266.
Holmes, J. (2018) ‘Negotiating the culture order in New Zealand workplaces’, Language in Society,
47(1), pp. 33–56.
Holmes, J. and Meyerhoff, M. (1999) ‘The communities of practice: theories and methodologies in
language and gender research’, Language in Society, 28(2), pp. 173–183.
King, B. (2014) ‘Tracing the emergence of a community of practice: beyond presupposition in
sociolinguistic research’, Language in Society, 43(1), pp. 61–81.
Lave, J. and Wenger, E. (1991) Situated learning: legitimate peripheral participation. Cambridge:
Cambridge University Press.

162
 Communities of practice and ethnography

Long, N., Burner, E., and Mookherjee, N. (2009) ‘Discussion point: when informants lie’, Cambridge
Anthropology, 29, pp. 85–94.
Mendoza-Denton, N. (2008) Homegirls: language and cultural practice among Latina gang girls.
Malden, MA: Blackwell Publishing.
Moore, E. (2006) ‘‘‘You tell all the stories’’: Using narrative to explore hierarchy within a Community
of Practice’, Journal of Sociolinguistics, 10, pp. 611–640.
Nagar, I. (2019) Being Janana: language and sexuality in contemporary India. New York: Routledge.
Nagar, R. (2014) Muddying the waters: coauthoring feminisms across scholarship and activism.
Champaign: University of Illinois Press.
O’Reilly, K. (2009) Key concepts in ethnography. London: SAGE.
Reddy, G. (2005) With respect to sex. Chicago: University of Chicago Press.
Strathern, M. (1991) Partial connections. Lanham, MD: Rowman & Littlefield Publishing Group, Inc.
Wenger, E. (1998) Communities of practice: learning, meaning, and identity. Cambridge: Cambridge
University Press.
Zimman, L. and Hall, K. (2016) ‘Language, gender, and sexuality’, in Aronoff, M. (ed) Oxford
bibliographies in linguistics. New York: Oxford University Press.

163
 11
Digital ethnography in the study
of language, gender, and sexuality
Piia Varis

Introduction
This chapter focuses on a rather recent theoretical and methodological approach: digital
ethnography. The emergence of digital ethnography in sociology, media studies, and also,
somewhat later, in the study of language and discourse, has to do with the development
of the internet into an infrastructure for a mass form of communication, and the ensuing
need for researchers to develop new ways of studying the emerging digital means of com-
munication as well as adapt older approaches to the new digital environments. This chapter
describes digital ethnography as an approach to studying language, gender, and sexuality
in these new environments through a case study of a young female YouTuber described in
more detail below.
With its origins in anthropology, ethnography itself is not a new approach, of course,
nor is it new in the study of language and discourse; indeed, there is an extensive body of
research in linguistic ethnography (see e.g. Copland and Creese 2015; Copland, Shaw, and
Snell 2015). What kind of an approach exactly ethnography is, however, can sometimes be
difficult to decipher, especially for those with less experience of it. This is because it appears
in the literature in different guises: some define it as an approach, others as a toolbox of
methods, in the latter case mainly consisting of fieldwork (observation) and interviews.
The understanding of ethnography assumed in this chapter is the former: ethnography
in general, and consequently also digital ethnography, is an approach rather than a set of
methods. What it aims to produce is ‘thick descriptions’ (Geertz 1973) of communicative
conventions and practices: situated accounts of lived reality. This means that ethnography,
including digital ethnography, ‘does not, unlike many other approaches, try to reduce com-
plexity of social events by focusing a priori on a selected range of relevant features, but it
tries to describe and analyse the complexity of social events comprehensively’ (Blommaert
2007: 682, emphasis original).
To address situated reality in all its potential complexity, ethnographic fieldwork is in
essence a learning process where research is guided by experience gathered in the field; it
is a mode of discovery and learning through which the researcher aims to familiarise them-
selves with the local realities and (linguistic) practices in question (Blommaert and Dong

164
 Digital ethnography

2010; Velghe 2014) through long-term engagement in the field. In the words of Hymes
(1996: 13),

[Ethnography] is continuous with ordinary life. Much of what we seek to find out in
ethnography is knowledge that others already have. Our ability to learn ethnographi-
cally is an extension of what every human being must do, that is, learn the meanings,
norms, patterns of a way of life.

To this end, ethnography is methodologically flexible; it cannot be reduced to specific data

collection techniques but rather responds to whatever becomes relevant in the field.

Digital ethnography
In the early 2000s, ethnographic research took on many digital forms, including in the study
of language and discourse, and appeared, for instance, under the labels ‘cyberethnography’
(Robinson and Schulz 2009), ‘Internet ethnography’ (boyd 2008; Sade-Beck 2004), ‘ethnog-
raphy of virtual spaces’ (Burrel 2009), ‘ethnography on the Internet’ (Beaulieu 2004), ‘vir-
tual ethnography’ (Hine 2000), ‘discourse-centred online ethnography’ (Androutsopoulos
2008), and ‘digital ethnography’ (Hou 2018a; Maly 2018; Murthy 2008; Pink, Horst, Postill,
et al. 2016; Varis 2016). These different types of ethnography differ from each other in cer-
tain respects. For instance, some of them focus on online data only (online ethnography),
while others may include a combination of online and offline research (digital ethnography).
They also differ in the extent to which they subscribe to ethnography as an approach and a
specific contextualised way of producing knowledge (the stance assumed in this chapter),
or simply see it as a methodological toolbox, mostly consisting of fieldwork and interviews.
In any case, the emergence of these different ethnographies tells us that a distinct approach
to research on digital environments, although with some variation within it, has appeared.
The reason for the appearance of these studies is the changing communication environ-
ments mentioned above. It is undeniable that the infrastructures for our lived realities – our
communications and the fabric of our social lives – have changed. Miller (2011: 44, empha-
sis original), for instance, points to the idea that:

digital media products are as much process as object. (…) So where one views a tra-
ditional media object such as a film, a book, or a painting, these objects have a finite
‘object-like’ quality to them. Digital media objects, more often than not, break from this
and are usually in continual production, being in constant dialogue and transformation
with the audience and with other digital products and technologies.

The field in digital ethnography, in other words, is potentially in a constant state of change.
As it is an impossible task for the ethnographer to often be present online to witness in real
time the potentially around-the-clock interactions taking place, this means that in many
cases she is dealing with products rather than processes, and potentially, ‘modifiable and
editable as digital artefacts are, what remains visible is the end result of possibly countless
edits, changes and deletions’ (Varis 2016: 62). Interestingly though, for digital ethnogra-
phers this means also that time becomes compressed: it is possible to access months’ and
even years’ worth of digital data, for instance, on social media very quickly. This potentially
changes the nature of fieldwork, as the ethnographer is not necessarily present for the inter-
actions as they unfold (cf. Varis 2016).

165
Piia Varis

Apart from the processual nature of digital objects and thus also the nature of ‘data’, prac-
tices also change – as Jones, Chik, and Hafner (2015: 3) put it, ‘digital technologies, because
of the different configurations of modes and materialities they make available, both make
possible new kinds of social practices and alter the way people engage in old ones’. For exam-
ple, the way in which we compose and circulate self-portraits – now known as selfies – has
definitely changed with the emergence of smartphones equipped with front-facing cameras,
built-in filters, and social media (see e.g. Georgakopoulou 2016). At the same time, arguably,
we have new types of social relationships enabled by social media practices of ‘following’,
‘friending’, and ‘subscribing’, as well as new forms of sociality and translocal groupness, as
people from potentially all over the world come together in often loose and short-lived groups
on social media and discussion forums, for instance (Varis and Blommaert 2015).
It is these kinds of new digitally mediated phenomena that digital ethnography has
appeared to address. Labelling ethnography ‘digital’ thus has to do with the changing media
and communication landscape – with new kinds of objects and practices – in which people
find themselves and on which we do research.

A digital ethnographic approach to language, gender, and sexuality

Ethnography is not new in the study of language, gender, and sexuality either (e.g. Besnier
and Philips 2014; Mendoza-Denton 2008), and there are also interesting ethnographic stud-
ies (touching) on gender and sexuality in different digitally mediated environments such
as social media and games (e.g. boyd 2008; Senft 2008; Sundén and Sveningsson 2012).
However, digital ethnographic research focusing specifically on language, gender, and sexu-
ality is still scarce (see, however, e.g. Darwin 2017; Georgakopoulou 2016; Hou 2018a).
These existing studies have contributed to research on language, gender, and sexuality by,
for instance, studying otherwise hard-to-reach populations now visible through the com-
munities they form online. Darwin’s (2017) online ethnography for example focused on
discussion threads and selfies in a genderqueer community on Reddit, illuminating the kinds
of discursive strategies non-binary people use to construct gender. Georgakopoulou (2016),
on the other hand, studied young women’s self-presentation through selfies on Facebook.
Through an ethnographic approach, she was able to show how young women’s selfies are
not expressions of narcissism or ‘ideal selves’ as they are commonly viewed, but context-
specific and co-constructed presentations of self influenced by the affordances of social
media. In Georgakopoulou’s case, a digital ethnographic approach was useful as it allowed
for immersion and participation through fieldwork, and thus a deeper understanding of the
young women’s selfie culture.
The focus in the case reported on in this chapter is on analysing digital practices, as
proposed by Jones et al. (2015) as one of the new ways of addressing the types of discourse
and action emerging through digitally mediated interactions. Jones et al. (2015: 3) define
digital practices as:

‘assemblages’ of actions involving tools associated with digital technologies, which

have come to be recognised by specific groups of people as ways of attaining particu-
lar social goals, enacting particular social identities, and reproducing particular sets of
social relationships.

Their definition of ‘tools associated with digital technologies’ is ‘not limited to software and
websites, but includes hardware (physical objects) and semiotic tools (such as conventional

166
 Digital ethnography

ways of talking or writing that have grown up around digital media)’ (ibid.). An advantage
of focusing on digital practices is that it allows for a focus on social conventions as digitally
mediated: that is, the practice approach enables contextualising language and communica-
tion by seeing them as embedded in an assemblage of activities and mediating tools. This
is useful for a digital ethnographic study, as, unlike a more limited approach focusing on
language only, it enables accounting for the ways in which specific types of communication,
identities, and relationships come into being mediated through digital technologies. The dis-
cussion on digital practices here will focus on the case of the YouTube star Hannah Witton,
and will highlight specific themes arising from the analysis of her YouTube channel: gender
and (online) visibility, gender and intersectionality, platform literacy, and revealing (online)
as a political act, in particular in relation to feminist activism.

Camgirls: history and present

The case of Hannah Witton can be placed in an established line of inquiry focusing on
women streaming their lives online. This research has so far mainly taken place in media
studies, with Senft’s 2008 study of ‘camgirls’ the earliest example of ethnographic research
on the topic. In her seminal book Camgirls: Celebrity and Community in the Age of Social
Networks, Senft (2008: 1) presents a study of ‘one generation of camgirls and their viewers
from 2000 to 2004’. Senft’s study is an explicitly feminist one, and she described herself as
‘trying to examine webcamming as a form of women’s expression’ (Senft 2008: 78). She
defines camgirls as ‘women who broadcast themselves over the Web for the general public,
while trying to cultivate a measure of celebrity in the process’ (Senft 2008: 1).
While my case study here focuses on this very phenomenon, over the decade that has
passed since Senft’s research there have been changes both in the meaning of the label
‘camgirl’, and the practice of ‘camgirling’. More or less everyone whom I have spoken to
about my interest in the topic has the idea that camgirls are strictly about sexual practices.
It is of course the case that the internet has become an infrastructure for new forms of sex,
both for pleasure and work, offering a disembodied (in the sense of not involving physical
contact) context for sexual practices. Such digital media practices have also recently come
to broad mainstream attention through popular documentaries such as the Netflix series
‘Hot Girls Wanted: Turned On’ (2017), introducing e.g. camgirl Alice who performs sex
acts online, and Cam Girlz (2015), a documentary film on internet sex workers ‘who find
economic freedom, empowerment, intimacy and creative self expression from the comfort
of their own homes’.1 One broader frame for an investigation of the camgirl phenomenon is
indeed the sexual dimension; the notion ‘camgirl’ has become sexualised to an extent that
was not the case at the time of Senft’s research. Bleakley’s (2014: 893) more recent work
also defines ‘camgirls’ as ‘young women that operate their own webcams to communicate
with a broad audience online, often engaging with sexually explicit behaviour in real-time
in return for financial compensation’. However, here I rather focus on more ‘mundane’ types
of webcamming where young women broadcast their everyday lives and, while sexuality
becomes a topic in the content, the case discussed here is not about broadcasting sexual
practices per se.

Data collection and analysis

In focus here is YouTube, as that is where I found not only the most accessible material but
also the most watched material, giving some indication as to what characterises (popular)

167
Piia Varis

camgirling at this point in time – and also the type of material that speaks most clearly to
the kinds of conditions digital ethnographers work in nowadays. In practice, I have spent
countless hours watching videos and channels, at the beginning simply trying to get a gen-
eral idea of what was going on by typing in search words such as ‘camgirl’ on YouTube and
watching whatever YouTube offered me, following suggestions to create a general overview
of content. This of course also means that I found what an ordinary YouTube browser would
find; the system of ‘recommendations’ effectively filters out a lot of content, which means
that for ethnographers it is relatively easy to remain ignorant of what is going on elsewhere,
and in other genres, unless one takes specific steps to avoid that (such as modifying search
settings and preferences). My own fieldwork on camgirls has thus emulated the experience
of an ‘ordinary user’ (Bleakley 2014; Hou 2018a), which is an ethnographically relevant
approach as it does not prioritise specific content beforehand, and also enables finding out
what ordinary audiences find, watch, and engage with online. There is of course a necessary
qualification to be made to the ‘ordinary user’ experience: I am inevitably guided by my
research interest rather than the personal interest of somebody who is interested in camgirls
due to personal reasons. There is nothing ethnographically new about this, of course; while
the networked nature of digital media introduces at least potentially the issue of scope and
scale in ways that pre-digital ethnography did not have to struggle with (each site, depend-
ing on its nature, potentially points to numerous different other sites in scale that is new),
the issue of boundary construction in fieldwork is not new as such (cf. boyd 2008). That is,
each ethnographer will have to make choices as to where the boundaries of their field lie: in
practice this means decisions regarding which issues, objects, participants, and phenomena
become relevant to follow for one’s research.
In my case, observation meant resorting to the means available and used by anybody
with internet access (‘anybody’ of course excluding resource-poor users, my fieldwork
behaviour and probably also my findings thus resembling more the browsing behaviour
of well-equipped and highly internet-literate users, with excellent internet connections).
Similarly, Georgakopoulou (2016: 303) has described how ‘the researcher’s own immer-
sion and participation in social media culture with processes of catching up, sharing, and
real-time tracking and life-streaming, are recognised as a major part of the development
of ethnographic understandings’. Searching, subscribing to channels, browsing, watching,
and following recommendations provided by YouTube itself, are all digital ethnographic
practices involved in constructing the online field. This is how fieldwork also becomes a
digital practice as defined by Jones et al. (2015) above: data collection and processing is
an assemblage of materials and actions, where in the case of YouTube here the medium
itself contributes to the fieldwork by sorting information to the researcher. The ethnographic
staple of reflexivity is useful here, as following recommendations and suggestions leads
to a compilation of a particular kind of database, with one’s own actions of searching and
clicking in interaction with what is offered by the medium constructing a certain viewing
path and ‘field’.
In finding camgirls to follow, the affordances of YouTube thus led me from one recom-
mendation to another, and I also did some news media searching and screening to see what
kinds of young women appeared in articles about ‘most popular’ or ‘recommended’ chan-
nels to watch to supplement my idea of what kinds of content represented camgirling. In
general, there seemed to be striking similarities to a lot of the content I saw. Each camgirl
had her own specific way of doing things, of course, but at the same time much seemed
to repeat itself to the extent of being formulaic. These linguistic and semiotic patterns
in YouTube camgirling became the most interesting feature to focus on. These included

168
 Digital ethnography

aspects like apologising for not having followed the established schedule for posting, greet-
ing one’s followers at the beginning of each video, and requesting viewers to subscribe to
the channel, with instructions on how to do that, at the end of each video; segmenting videos
into an introductory ‘this is what I am going to do today’ part and a montage of moments
and events later that day; specific popular subgenres appearing in different channels such
as the shopping haul, product review series, etc. The most popular ones that I watched also
had a striking development of professionalisation, from sometimes very poor-quality early
videos to professionally shot and edited later ones, with sponsors and/or campaigns involv-
ing companies and products appearing along the way.

The case of Hannah Witton

One young female YouTuber whom I have specifically followed due to her explicit focus
on gender and sexuality in some of her videos, and whose broadcasting at the same time
serves as an illustration of typical broadcasting practices in the present-day mediascape, is
Hannah Witton (b. 1992). Witton’s channel is typical of present-day (YouTube) camgirling
and ‘micro-celebrity’. Marwick (2013: 114) defines micro-celebrity as ‘a state of being
famous to a niche group of people’ online which ‘requires creating a persona, producing
content, and strategically appealing to online fans by being “authentic”’. Apart from being
popular on YouTube, she has around 100,000 followers both on Twitter and Instagram, and
her own website hannahwitton.com describes Witton as ‘YouTuber, history grad, sex posi-
tive, feminist, Hufflepuff’. According to the Wikipedia page about her (with the existence of
the page being a further sign of her professionalism and a certain level of celebrity), she is

a British YouTuber, broadcaster, and author. Witton creates video blogs and informa-
tional content, mostly based around relationships, sex and sexual health, liberation and
welfare issues, literature, and travel. Witton’s debut book, Doing It, concentrating on
sex and relationships, was released on 6 April 2017.

Her YouTube channel2 has almost 400,000 subscribers, with playlists on topics ranging
from ‘Sex & Relationships’, ‘Drunk Advice’, and ‘Music’, to one dedicated to her book
(called #DoingItBook).
On 9th August 2016, Witton published the first video of the series ‘The Hormone
Diaries’ on her YouTube channel. I found this particular series interesting to focus on due
to its explicit focus on (the regulation of) female sexuality. While Jones (2016: 228) in
her research on camgirls found that women’s webcamming ‘has the potential to subvert
the hegemonic discourses of gender that constrain women’s sexuality because this work
allows for various forms of pleasure in the process’, Witton’s ‘hormone diaries’ offer a dif-
ferent perspective to the potential subversiveness of women’s streaming activities regarding
hegemonic discourses of gender and sexuality by critically discussing normative reproduc-
tive practices and their effects on the body.
In the first video in the series, titled ‘Why I’m Coming off the Pill’,3 Witton announces
that she is about to stop taking the pill. She has taken the decision after not having had her
period for seven years (hence the title ‘The Hormone Diaries’, which also already points to
a personal journaling approach to recounting personal, intimate experience), and the series
will describe her experiences along the way. This first video in the series has gathered more
than 200,000 views and close to 1,500 comments. In about seven minutes, in this first video,
Witton explains how she feels ‘like someone without a female reproductive system’ and

169
Piia Varis

expresses anxiety over the situation, yet ends, in a typical YouTuber gesture, by enthusiasti-
cally asking for comments and ‘thumbs-ups’ from the viewers if they are ‘excited’ about
the series. The most popular episodes in the series, Episode 9 ‘Trying a Menstrual Cup for
the First Time’4 (7th March 2017) and Episode 3 ‘FIRST PERIOD IN 7 YEARS’5 (20th
September 2016), have gathered more than 200,000 and around 270,000 views, respec-
tively. All the other episodes in the series (of which there are 14; checked 18th November
2017) have gathered views at least in the tens of thousands, attesting to her popularity and
visibility in general, and the fact that the series on the female body and contraception has
reached quite a large audience of viewers on YouTube.
Witton’s channel does not stand out as special in terms of the formulaic elements of
similar YouTube channels; for instance the end screen of her videos shows a link to her
website and Twitter handle, an attempt to engage more followers for all her social media
channels. As for content, broadly speaking Witton’s channel could be described as a form
of infotainment, apparent amongst her other content in ‘The Hormone Diaries’ series. With
a ‘girl-next-door’ approach, Witton seems easily relatable; she also does not shy away from
blunders and seeming silly, contributing to a sense of authenticity (cf. Marwick 2013) – in
the first video of ‘The Hormone Diaries’, for instance, the viewer sees her asking herself,
‘Why am I doing this?’, showing anxiety and vulnerability regarding the situation she finds
herself in.

Camgirling as feminist activism

Due to the kinds of themes Witton’s series addresses, it arguably does a kind of public ser-
vice, and her YouTube channel together with her other activities, such as publishing a book,
can be said to constitute a form of feminist activism. In ‘The Hormone Diaries’ series, after
a doctor’s appointment she, for instance, explains to her viewers the existing non-hormonal
forms of contraception to inform her viewers about alternatives to the pill, and reports on
research done on the kind of pill she had been taking (Episode 26). She also makes the point
that if there was a pill that was suspected to influence men’s sex drive there would have been
more of a reaction to the consumption of these pills – thus assuming a feminist perspective
and not only giving her readers ‘practical’ information, but also placing her own and many
of her viewers’ experiences in the broader societal context (cf. Mackenzie 2017). In the
fourth7 episode she interviews her mother about her experiences with contraception, and the
sixth8 one is about period sex, also mentioning UK government plans to try and ban period
sex from pornographic consumption. These videos similarly work not only as a form of
entertainment due to Witton’s humorous style or in perhaps establishing solidarity through
shared young woman’s experience of struggling with periods, hormones, and contraception
and living as a young woman in today’s Western society, but also enlighten the viewer with
factual information.
In today’s (social) media environment, the developments in Witton’s YouTube chan-
nel are not exceptional. Browsing her broadcasting through the YouTube archive from
the first videos to more recent ones, along the way sponsors appear, a book is launched; a
brand that works in relation to other brands comes into being. For instance in the seventh9
episode of ‘The Hormone Diaries’, Witton mentions that that particular video is spon-
sored by a LloydsPharmacy10 online doctor; she explains that the video is, at the request
of the company, about the morning-after pill, which the online doctor supplies. Witton’s
seemingly personal broadcasting of her life thus becomes entangled with commercial
interests.

170
 Digital ethnography

From a digital ethnographic point of view, what is important to highlight here about
Witton’s online presence is the kinds of infrastructures she has at her disposal for broadcast-
ing herself. That is, the environment in which especially those of us working with social
media do our research has changed quite dramatically over the past decade or two, with
the commercialisation of the internet and social media such as YouTube gaining ever more
prominence in everyday media consumption. Witton’s activities and the development of her
channel reflect general tendencies in channels competing for attention in the algorithmic
sorting environment of YouTube. This means, for instance, working through the algorithmic
popularity principle and monetising programmes, where, for instance, through YouTube’s
partnership programme and advertising practices certain broadcasters gain more promi-
nence. In terms of video content, this means that specific formulaic, increasingly profes-
sional types of content and commercial engagement seem to be able to attract more viewers
and rise to prominence (Hou 2018b). At the same time, certainly specific topics within
such professionalised channels become more popular: in a video entitled ‘Why Having
Big Boobs Sucks’11 (June 6, 2017) Witton discusses the changes in her breasts after her
hormone diary experiment. The video has garnered more than 600,0000 views and more
than 10,000 comments, several times more than the most popular videos in ‘The Hormone
Diaries’ series itself. Judging from the comments, the video, much more than the others,
seems to have attracted the attention of both female and male audiences, and there is also
blatant sexualisation of the topic, as well as Witton herself, in the comments. The colloquial
and explicit video title not only speaks in a relatable way to young female audiences but also
clearly attracts a different and broader viewership than, for example, the hormone diaries
videos addressing questions such as whether using the coil is a good form of contraception
(Episode 2, ‘Should I get the Coil’, 6th September 2016). On her channel page, what appears
under ‘Popular uploads’ is indeed the ‘Why Having Big Boobs Sucks’ video along with the
titles ‘NAKED IN PUBLIC!?’12 (more than 300,0000 views) and ‘10 MASTURBATION
HACKS’13 (more than 200,0000 views). Titles pointing to sexualisation and apparent trans-
gression (apparent only, to ensure not being censored by YouTube for sexual content, for
instance) thus work to bring viewers to the channel. At the same time, Witton concludes the
‘Why Having Big Boobs Sucks’ video with ‘We can’t expect ourselves to be body positive
all the time’, framing the discussion within a feminist discourse on the body. Discursively,
Witton’s broadcasting is therefore a successful mix catering to different audiences.
Simultaneously, thus, while clearly shaped by the commercial environment that is
YouTube, as mentioned, Witton’s channel is also a form of feminist activism. As a type of
discourse, her ‘Hormone Diaries’ series also reminds viewers of a 1970s feminist conscious-
ness-raising group, with experiences and knowledge being shared, injustices and problems
being pointed out, and solutions offered. This is also part of the broader media environment,
and perhaps an interesting addition to the way in which Gill (2007: 161) described the
present-day post-feminist media environment as arguably ‘much of what counts as femi-
nist debate in western countries today takes place in the media rather than outside of it’.
However, in this broader context what is interesting is the ‘siloing’ effect of media such
as YouTube: broadcasters such as Witton are popular, but on the other hand their activities
are nested in a commercial environment where the regulation by the companies hosting
their activities plays a role in what they are doing and how they run their broadcasting,
and explicit branding and sponsoring are part of the feminist messaging. Micro-celebrity
branding and conforming to certain commercial practices are part and parcel of ensuring
visibility in the public sphere. As Marwick (2013: 5) also found in her more recent study on
broadcasting oneself online:

171
Piia Varis

far from the revolutionary and progressive participation flaunted by entrepreneurs and
pundits, social media applications encourage people to compete for social benefits by
gaining visibility and attention. To boost social status, young professionals adopt self-
consciously constructed personas and market themselves, like brands or celebrities, to
an audience or fan base. These personas are highly edited, controlled, and monitored,
conforming to commercial ideals that dictate ‘safe-for-work’ self-presentation. The
technical mechanisms of social media reflect the values of where they were produced:
a culture dominated by commercial interest.

Consequently, scholars studying language and discourse in digital environments can benefit
from an engagement with approaches that focus not only exclusively on the language or
discourse on display online, but also the ways in which they are mediated by online environ-
ments – mediated discourse analysis and platform studies perhaps among them (e.g. Hou
2018a, 2018b; Massanari 2015; Norris and Jones 2005; Varis 2016, 2017a). What is ‘social
media discourse’, for instance, is, like in the case of camgirls broadcasting in commercial
environments, shaped by the increased commercialisation of the internet, the nesting of the
broadcasting of one’s mundane everyday life into commercial practices and branding, and
algorithmic sorting into popular and hence visible videos, as well as the sidelining of both
controversial and not so popular content. As van Dijck (2013) points out, social media envi-
ronments are not neutral intermediaries, but mediators in digital social life and relationships,
shaping the way in which these materialise.

Gender and online visibility

An important dimension to consider in relation to the issue of mediation mentioned above
has to do with the notion of visibility. There is a lot to investigate in terms of the visibility of
Witton’s as well as other young women’s self-presentation through digital media. In the field
of sociolinguistics, Hanell and Salö (2015) have already introduced what seems like a useful
notion of ‘orders of visibility’ to interrogate online discourse and the kinds of discourses and
practices that become visible in online environments (see also Bzura 2007 on girl cams and
blogs, and Brighenti 2007). The notion of visibility seems of utmost importance for scholars
working on gender and sexuality in online media, and no less for those doing that work from a
linguistic or discourse studies point of view. This relates to platform policies and algorithmic
regulation (e.g. Massanari 2015) that influence gender and sexuality – the point above by van
Dijck (2013) regarding the role of social media as mediators, instead of ‘neutral’ intermediar-
ies. Visibility of specific types of content, and camgirls, on social media has to do, first, of
course, with the power of social media companies to regulate content by censoring (recall
for instance the #freethenipple hashtag demanding as part of a broader campaign the end
to censoring female nipples on social media and elsewhere in the public sphere as just one
example of how gendered the regulation of content is). On a less immediately obvious and
visible level, the fact that camgirls such as Hannah Witton are visible to ordinary users and
ethnographers alike does of course have to do with the kind of content they produce and cir-
culate; the titling practices in Witton’s YouTube channel mentioned above are indicative of the
kind of discursive work going on in attracting audiences and the numbers of views attest to the
successfulness of this work. This kind of content makes sense in the context of being shaped
by the media environment that can be characterised as an attention economy and ruled by the
popularity principle, where specific formulaic camgirl discourse practices are rewarded, for
instance with sponsoring and advertising partnerships (cf. Hou 2018b).

172
 Digital ethnography

Gender, visibility, and intersectionality online

There is also another dimension related to gender and online visibility that becomes relevant
in the analysis of Witton’s camgirling: that of intersectionality. Keller’s (2012: 444) discus-
sion on virtual feminisms for instance reminds us that ‘structural inequalities continue to
prevent some girls from participating in blogging communities and other online practices’.
The more privileged will have the ‘material resources to blog, as well as the leisure time to
do so’ (ibid.), and as Keller (ibid.) points out, ‘In addition to these inequalities existing on a
global scale, race, class, and location inequalities continue to shape girls’ access to commu-
nication technologies in Canada, the United States, and other Western countries’ – and, we
should add, not to mention in ‘non-Western’ contexts (see also Jones 2015 on sex work in
a digital era for a similar conclusion regarding the importance of intersectional analysis). If
we are to produce sequels to Senft’s 2008 work in studying women’s online voices then ‘the
ways in which inequalities shape the practice of feminist blogging specifically must be an
important focus for future research, which will allow us to better understand whose voices
come to represent contemporary feminisms’ (ibid.).
It remains a question to be interrogated whether and how present-day social media due
to its principles ends up highlighting certain kinds of content, and perhaps certain kinds
of (branded) feminisms, or discursive mixes such as Witton’s which attract large numbers
of viewers with (seeming) transgression and daring, combined with informational content
and feminist discourse on the body. A further interesting point to ponder for linguists and
discourse analysts has to do with not necessarily focusing their analysis on the content of
online semiotic activity; as Koskela (2004: 210, emphasis original) remarks,

most home webcams are extremely mundane. (…) When looking at these webcams it
is difficult to see a slightest sign of resistance. However, the point that I want to make
is not that the actual pictures would show resistance. The question is not about political
activism in traditional sense but about revealing as a political act – intrinsically.

We might therefore ask whether popular channels such as Hannah Witton’s – as much as
they present a white, middle-class, well-educated perspective on young female life and fem-
inism in a commercial platform – have something intrinsically valuable to them in making
young female experiences visible in the public sphere. As platforms such as YouTube – that
filter out what they deem ‘offensive’ content – dominate a lot of today’s online traffic and
media consumption, personas that are ‘are highly edited, controlled, and monitored, con-
forming to commercial ideals that dictate “safe-for-work” self-presentation’, as described
by Marwick (2013: 5) above, may continue to be the dominant type of content. What this
means for representation of gender and sexuality will have to be investigated, preferably
through long-term ethnographic engagement that contextualises such content with past and
ongoing changes in the media environment.

Platform literacy
Finally, there is a broader issue that emerges from ethnographically analysing digital data
such as Hannah Witton’s YouTube channel: digital literacy, and specifically the issue of
platform literacy (Varis 2017b), which is something researchers need in order to make sense
of gender and sexuality in the present-day media environments. This means understanding
not only individual ‘micro-systems’ but also the online media ‘ecosystem’ (van Dijck 2013);
camgirls, just like other broadcasters, navigate an increasingly commercialised internet

173
Piia Varis

where discourse is shaped by the conditions in which users find themselves. For us research-
ers, it is of essence to understand how, for instance, the visibility and circulation of topics
pertaining to gender and sexuality are influenced by media affordances, and how media
ideologies – how people think about media and how they are designed (Gershon 2010) –
influence not only the performance of gender and sexuality, but also how these are contex-
tualised and put in further discursive trajectories. While we need to be careful in overstating
digital revolutions, at the same time we need to be aware of the genres and discursive condi-
tions of the new mediating environments. Digital ethnography is one useful approach for us
in doing that, and I join Keller (2012: 444) who concludes, based on her research on girls’
blogging and participation in public life, in ‘advocat[ing] for an ethnographic approach to
studying girls’ online practices, which has the ability to uncover the complexity of blogging
practices and place girls’ feminist blogging as part of broader changes in technology, social
activism, and feminism itself’.

Future directions
This chapter has described digital ethnography, with a specific focus on digital practices, as
an approach to studying digitally mediated discourse on gender and sexuality. This meant
viewing the discourse as mediated by the environments in question, helping illuminate the
reasons why particular discourses appear in the form they do in social media such as YouTube.
This entails attending to digital practices in which assemblages of actions, tools, and semiotic
resources become the object of analysis. When it comes to digitally mediated language and
discourse, research – including that focusing on gender and sexuality – so far has mainly
analysed only the discourse itself, rather than examining it in the context of the mediating
environments. However, researchers should look more specifically into the ways in which
the material infrastructures of digital media play a role in shaping semiotic material (see also
Jones 2016) as this would help further our understanding of how certain discourses on gender
and sexuality become more visible than others, and how specific subjects become the ones to
visibly voice them, and through which discursive means. The role of the mediating environ-
ments is crucial here, as visibility is the result of algorithmic social media systems offering
content for audiences. For us interested in discourses on gender and sexuality in the public
sphere this is an important line to pursue, and digital ethnography can help us understand how
these new mediated realities, as assemblages of human and non-human actors, work.

Notes
1 http://camgirlzthemovie.com/home.html
2 https://www.youtube.com/user/hannahgirasol
3 https://www.youtube.com/watch?v=5jx6sSpis9g
4 https://www.youtube.com/watch?v=fpBku_K_hzE
5 https://www.youtube.com/watch?v=637hUE6H1-A
6 https://www.youtube.com/watch?v=z3-bMOZFHUg
7 https://www.youtube.com/watch?v=i9Af6q05Jhw
8 https://www.youtube.com/watch?v=4ZQPBl_4vLA
9 https://www.youtube.com/watch?v=-9IIWgH_1jg
10 LloydsPharmacy is a British pharmacy company, offering e.g. prescriptions and health advice both
in-store and online.
11 https://www.youtube.com/watch?v=qg-yA78uke0
12 https://www.youtube.com/watch?v=w1yDoxvipVU
13 https://www.youtube.com/watch?v=K0gt-UaRS0U

174
 Digital ethnography

Further reading
Boellstorff, T., Nardi, B., Pearce, C., and Taylor, C. L. (2012) Ethnography and virtual worlds: a
handbook of method. Princeton, NJ: Princeton University Press.
This book is particularly useful for those who see themselves as beginners in ethnography: it
discusses myths and misunderstandings regarding ethnographic research, gives practical advice, and
while focusing on virtual worlds, has wider relevance.
boyd, d. (2008) Taken out of context: American teen sociality in networked publics. Doctoral
dissertation. University of California, Berkeley.
boyd’s doctoral dissertation about American teenagers and their use of social network sites, is
just one of her works that can be easily recommended as reading; while not a scholar of language or
discourse, her ethnographic research on digital culture is instructive.
Hine, C. (ed.) (2013) Virtual methods: issues in social research on the Internet. London: Bloomsbury.
Hine’s edited volume is a useful collection of case studies that serves as an introduction to different
methodological and ethical issues relevant to the study of online environments.
Page, R., Barton, D., Unger, J. W., and Zappavigna, M. (2014) Researching language and social
media. A student guide. Abingdon: Routledge.
This book is useful for students of language and social media in general, but also includes a very
instructive and practical chapter on ethnographic approaches, as well as e.g. online research ethics.
Varis, P. (2016) ‘Digital ethnography’, in Georgakopoulou, A. and Spilioti, T. (eds) The Routledge
handbook of language and digital communication. London: Routledge, pp. 55–68.
This handbook chapter focuses specifically on digital ethnography as an approach in the study of
language.

Related topics
Analysing gendered discourses online; gender, language, and elite ethnographies in uk political
institutions; using communities of practice and ethnography to answer sociolinguistic questions;
multimodal constructions of feminism; doing gender and sexuality intersectionally in multimodal
social media practices.

References
Androutsopoulos, J. (2008) ‘Potentials and limitations of discourse-centred online ethnography’,
Language@Internet, 5, art. 9. Available at: http://www.languageatinternet.org/articles/2008/1610/
androutsopoulos.pdf (Accessed: 6th December 2017).
Beaulieu, A. (2004) ‘Mediating ethnography: objectivity and the making of ethnographies of the
Internet’, Social Epistemology, 18(2–3), pp. 139–163.
Besnier, N. and Philips, S. U. (2014) ‘Ethnographic methods for language and gender research’, in
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) The handbook of language, gender, and sexuality,
2nd edn. Hoboken, NJ: John Wiley & Sons, pp. 123–140.
Bleakley, P. (2014) ‘“500 tokens to go private”: camgirls, cybersex and feminist entrepreneurship’,
Sexuality & Culture, 18(4), pp. 892–910.
Blommaert, J. (2007) ‘On scope and depth in linguistic ethnography’, Journal of Sociolinguistics,
11(5), pp. 682–688.
Blommaert, J. and Dong, J. (2010) Ethnographic fieldwork: a beginner’s guide. Bristol: Multilingual
Matters.
boyd, d. (2008) Taken out of context: American teen sociality in networked publics. Doctoral
dissertation. University of California, Berkeley.
Brighenti, A. (2007) ‘Visibility: a category for the social sciences’, Current Sociology, 55(3), pp. 323–342.
Burrel, J. (2009) ‘The field site as a network: a strategy for locating ethnographic research’, Field
Methods, 21(2), pp. 181–199.

175
Piia Varis

Bzura, K. (2007) I’m not who I was then, now: performing identity in girl cams and blogs. Master’s
thesis, University of South Florida. Available at: http://scholarcommons.usf.edu/cgi/viewcontent
.cgi?article=1650&context=etd (Accessed: 6th December 2017).
Copland, F. and Creese, A. (2015) Linguistic ethnography: collecting, analysing and presenting data.
London: Sage.
Copland, F., Shaw, S., and Snell, J. (eds) (2015) Linguistic ethnography: interdisciplinary explorations.
Houndmills: Palgrave Macmillan.
Darwin, H. (2017) ‘Doing gender beyond the binary: a virtual ethnography’, Symbolic Interaction,
40(3), pp. 317–334.
Geertz, C. (1973) ‘Thick description: toward an interpretive theory of culture’, in Geertz, C. (ed) The
interpretation of cultures: selected essays. New York: Basic Books, pp. 3–30.
Georgakopoulou, A. (2016) ‘From narrating the self to posting self(ies): a small stories approach to
selfies’, Open Linguistics, 2(1), pp. 300–317.
Gershon, I. (2010) Breakup 2.0: disconnecting over new media. Ithaca, NY: Cornell University Press.
Gill, R. (2007) ‘Postfeminist media culture: Elements of a sensibility’, European Journal of Cultural
Studies, 10(2), pp. 147–166.
Hanell, L. and Salö, L. (2015) ‘“That’s weird, my ob-gyn said the exact opposite!”: discourse and
knowledge in an online discussion forum thread for expecting parents’, Tilburg papers in culture
studies 125. Available at: https://www.tilburguniversity.edu/upload/dc6b8024-c200-4dbd-8bdb-
bd684aa636c0_TPCS_125_Hanell-Salo.pdf (Accessed: 6th December 2017).
Hine, C. (2000) Virtual ethnography. London: Sage.
Hou, M. (2018a) Social media celebrity: an investigation into the latest metamorphosis of fame. PhD
dissertation. Tilburg University.
Hou, M. (2018b) ‘Social media celebrity and the institutionalization of YouTube’, Convergence: The
International Journal of Research into New Media Technologies, 25(3), pp. 534–553.
Hymes, D. (1996) Ethnography, linguistics, narrative inequality: Toward an understanding of voice.
London: Taylor & Francis.
Jones, A. (2015) ‘Sex work in a digital era’, Sociology Compass, 9(7), pp. 558–570.
Jones, R. (2016) ‘The age of print literacy and “deep critical attention” is filled with war, genocide
and environmental devastation’, in: Simanowski, R. (ed) Digital humanities and digital media:
conversations on politics, culture, aesthetics and literacy. London: Open Humanities Press, pp.
228–246.
Jones, R. H., Chik, A., and Hafner, C. A. (2015) ‘Introduction: discourse analysis and digital practices’,
in Jones, R. H., Chik, A., and Hafner, C. A. (eds) Discourse and Digital practices. Doing discourse
analysis in the digital age. Abingdon: Routledge, pp. 1–17.
Keller, J. M. (2012) ‘Virtual feminisms’, Information, Communication & Society, 15(3), pp. 429–447.
Koskela, H. (2004) ‘Webcams, TV shows and mobile phones: empowering exhibitionism’, Surveillance
& Society, 2(2/3), pp. 199–215.
Mackenzie, J. (2017) ‘”Can we have a child exchange?” Constructing and subverting the “good
mother” through play in Mumsnet Talk’, Discourse & Society, 28(3), pp. 296–312.
Maly, I. (2018) ‘New media, new resistance and mass media: a digital ethnographic analysis of the Hart
Boven Hard movement in Belgium’, in Papaioannou, T. and Gupta, S. (eds) Media representations
of anti-austerity protests in the EU. Grievances, identities and agency. New York: Routledge, pp.
165–187.
Marwick, A. E. (2013) Status update: celebrity, publicity, and branding in the social media age. New
Haven: Yale University Press.
Massanari, A. (2015) ‘#Gamergate and the Fappening: how Reddit’s algorithm, governance, and
culture support toxic technocultures’, New Media & Society, 19(3), pp. 329–346.
Mendoza-Denton, N. (2008) Homegirls. Language and Cultural practice among Latina youth gangs.
Malden, MA: Blackwell.
Miller, V. (2011) Understanding digital culture. London: Sage.

176
 Digital ethnography

Murthy, D. (2008) ‘Digital ethnography: an examination of the use of new technologies for social
research’, Sociology, 42(5), pp. 837–855.
Norris, S. and Jones, R. H. (eds) (2005) Discourse in action: introducing mediated discourse analysis.
London: Routledge.
Pink, S., Horst, H., Postill, J., Hjorth, L., Lewis, T., and Tacchi, J. (2016) Digital ethnography:
principles and practice. London: Sage.
Robinson, L. and Schulz, J. (2009) ‘New avenues for sociological inquiry: evolving forms of
ethnographic practice’, Sociology, 43(4), pp. 685–698.
Sade-Beck, L. (2004) ‘Internet ethnography: online and offline’, International Journal of Qualitative
Methods, 3(2), pp. 45–51.
Senft, T. (2008) Camgirls: celebrity and community in the age of social networks. New York: Peter
Lang.
Sundén, J. and Sveningsson, M. (2012) Gender and sexuality in online game cultures: passionate
play. London: Routledge.
van Dijck, J. (2013) ‘‘‘You have one identity’’: performing the self on Facebook and LinkedIn’, Media,
Culture & Society, 35(2), pp. 199–215.
Varis, P. (2016) ‘Digital ethnography’, in Georgakopoulou, A. and Spilioti, T. (eds) The Routledge
handbook of language and digital communication. London: Routledge, pp. 55–68.
Varis, P. (2017a) ‘Superdiverse times and places: media, mobility, conjunctures and structures
of feeling’, in Arnaut, K., Karrebæk, M. S., Spotti, M., and Blommaert, J. (eds) Engaging
superdiversity: recombining spaces, times and language practices. Bristol: Multilingual Matters,
pp. 25–46.
Varis, P. (2017b) ‘On its 28th birthday, how democratic is the web? Or, why people need platform
literacy’, Diggit Magazine. Available at: https://www.diggitmagazine.com/column/its-28th-bir
thday-how-democratic-web (Accessed: 6th December 2017).
Varis, P. and Blommaert, J. (2015) ‘Conviviality and collectives on social media: virality, memes, and
new social structures’, Multilingual Margins: A Journal of Multilingualism from the Periphery,
2(1), pp. 31–45.
Velghe, F. (2014) “This is almost like writing”: mobile phones, learning and literacy in a South
African township. PhD dissertation. Tilburg University.

177
 Part III
Interactional sociolinguistic
approaches
 12
Interactional sociolinguistics
Foundations, developments, and applications
to language, gender, and sexuality

Cynthia Gordon and Deborah Tannen (Part III leads)

Introduction
There is general agreement that the field known as ‘interactional sociolinguistics’ traces
back to the theoretical and methodological approach developed in the late 1970s by
Gumperz and laid out in his book Discourse Strategies (1982a). In the decade immediately
following, work in this paradigm was done primarily by Gumperz and his students, or those
who had been trained by, or worked directly with, him or his students. In the years since,
however – a testament to the power and influence of the approach Gumperz pioneered – the
term ‘interactional sociolinguistics’ (IS) has come to include a broad range of qualitative
microanalyses of recorded, naturally occurring interactions conducted with attention to the
social context in which the interaction took place. Our goal in this introduction is to provide
a kind of genealogy of the field and overview of its development with brief summaries of
exemplary studies. Towards that end, we recap the key terms, concepts, and methods that
characterise Gumperz’ foundational work as well as some of the additions and elaborations
that have been adapted into the framework and have come to be associated with it.1 We con-
clude with a sampling of studies that draw on Gumperz’ and related approaches to examine
language and gender, and the smaller number of studies examining language and sexuality,
an area we hope will benefit from such investigations going forward.

John Gumperz’ conception of interactional sociolinguistics

Interactional sociolinguistics (IS) is a qualitative, interpretative approach to the analysis of
the language of real interaction in the context of social relationships. It is founded on the
convictions that language can only be studied in context and that it is constitutive of context.
In other words, context is not something pre-existing that language fits into but rather the
language of interaction creates context – as well as social relationships and identities. The
goal of IS, then, is no less than accounting for the communication of meaning through lan-
guage use. The word ‘use’ is crucial. IS views meaning not as inherent in words, but rather
as jointly created by speakers and listeners engaged in the act of using language to accom-
plish interactive goals. In order to understand and explicate this process, IS methodology

181
Cynthia Gordon & Deborah Tannen

involves an ethnographic component in the form of observation in naturally occurring con-

texts, often participant-observation; audio- or audiovisual-recording of interaction; detailed
linguistic transcription of recorded conversations; microanalysis of interaction as reflected
in the transcripts in light of the information gained through ethnography; and post-recording
‘playback’ interviews with participants and others from similar sociocultural backgrounds.
The first publication in which Gumperz laid out his approach that later became known as
interactional sociolinguistics is a paper entitled ‘Sociocultural knowledge in conversational
inference’, which he gave at the 1977 Georgetown University Round Table on Languages
and Linguistics (GURT) and was published in the collection of papers from that conference
(Gumperz 1977). It provides a window onto the roots and foundations of IS. (This paper, in
slightly adapted form, appears as Chapter 7 in Discourse Strategies.)2
It is worth noting that the term ‘interactional sociolinguistics’ does not appear in
Discourse Strategies, even though that book is the first in the Cambridge series entitled
‘Studies in Interactional Sociolinguistics’, and Gumperz was a founding co-editor of the
series. Gumperz (1977, 1982a) initially referred to his approach as a ‘theory of conver-
sational inference’, but he later (Gumperz 2001) uses ‘interactional sociolinguistics’ to
identify his approach, possibly to distinguish it from Labovian variation analysis, the other
dominant sociolinguistic theory and method, and from the other dominant approach to ana-
lysing recorded conversational interaction pioneered by Sacks and Schegloff that has come
to be known as conversation analysis (CA).
In contrast to these approaches, IS research has long had an expansive quality to it,
as Schiffrin (1994: 97) notes: ‘The approach to discourse that I am calling “interactional
sociolinguistics” has the most diverse disciplinary origins’ among those she discusses. She
traces IS to both Gumperz and Goffman, whom Gumperz frequently mentioned as one of
several scholars who influenced him and whom he called a ‘sociological predecessor’ of
his work (Gumperz 2001: 216). Goffman’s many books were transformational in providing
close analysis of behaviour in everyday life, beginning with his groundbreaking 1956 The
Presentation of Self in Everyday Life. However, it was not until his 1981 Forms of Talk that
he turned attention to language. By the time Schiffrin wrote Approaches to Discourse, many
of those who had adopted Gumperz’ interpretive theories and methods also made ample
use of Goffman’s notions of footing, alignment, and other aspects of framing, as Schiffrin
herself did (see for example Schiffrin 1993).

Roots and foundational concepts of Gumperz’ interactional sociolinguistics

In the introduction to his 1977 essay ‘Sociocultural knowledge in conversational inference’,
Gumperz identifies three ‘major research traditions that have dealt with social factors in
speech’ (1977: 192), acknowledging what he sees as the contributions of each to under-
standing language in interaction, then pointing out shortcomings that his own approach will
address. These traditions are (1) the ethnography of speaking, (2) linguistic pragmatics, and
(3) ethnomethodology, or the sociology of verbal interaction. We will recap what he says
about these approaches as a way of introducing and explaining aspects that Gumperz him-
self regarded as key to his theory of IS, and to provide insight into some of the roots of and
influences on his theory. (There are others that he mentions in later essays and interviews.)
Gumperz traces the first of these three traditions, ‘the ethnography of speaking’, to the
work of Hymes (1962), and to their own joint collection (Gumperz and Hymes 1972), whose
title and subtitle, Directions in Sociolinguistics: The Ethnography of Communication, imply

182
 Interactional sociolinguistics

that the two refer to a single field. Shortly after this volume’s appearance, the two scholars
seem to have divided the field between them: Hymes went on to lead the field ‘ethnography
of speaking’ while Gumperz went in the direction of qualitative ‘sociolinguistics’. Their col-
laboration can be traced further back, to the 1964 special issue of American Anthropologist
they co-edited, which includes papers by many of the same scholars included in their 1972
collection. In his 1977 GURT essay, Gumperz credits ‘ethnography of speaking’ with using
‘anthropological methods of interviewing and participant observation’ to describe cultur-
ally specific speech events. (In later writing, he traces his notion of ‘speech activity’ to that
of ‘speech event’.) He argues, however, that such descriptive methods do not adequately
account for ‘how people integrate social knowledge in interaction’ in ordinary, everyday
conversations (1977: 193). That will become one of the main goals of IS.
‘Linguistic pragmatics’, the second research tradition that Gumperz sees as predating his
own, is, in his view, ‘an effort to give linguistic substance to the philosophers’ notions of
speech act and verbal games’ and ‘motivated primarily by a concern with abstract linguistic
theory’ (1977: 194). He cites Fillmore (1977), Gordon and Lakoff (1973), and Halliday and
Hasan (1976) as linguists seeking to build on the work of philosophers such as Wittgenstein,
Austin, Grice, and Searle to go beyond simple grammatical analysis to account for how
speakers use words to accomplish speech acts, but concludes that this approach is limited
in focusing only on the speaker, failing to account for the fact that ‘conversing is rather like
collaborating in the production of a play, where each person’s contribution is constrained by
what the others can do and what the audience will accept’ (Gumperz 1977: 194). Gumperz
thereby highlights another key element of IS: that meaning does not reside solely in the
words spoken but is jointly created by speaker and hearer. This insight, the significance of
which cannot be overstated, is inherent in Gumperz’ notions of ‘conversational inference’
and ‘thematic progression’. By using the term ‘conversational inference’ to refer to his own
approach, Gumperz emphasised that listeners must be able not only to decode the words
spoken but also to identify what the speaker intends by saying them. Even that is insufficient
on its own. ‘Thematic progression’ refers to the need for listeners to also be able to predict
what the speaker is likely to say next. Thus, for example, a native speaker, by recognising
listing intonation, will know, on hearing only the first word or phrase, that it will be followed
by one or more additional words or phrases – a prediction necessary to fully understand the
meaning of the first words spoken.
The third research tradition Gumperz discusses, ‘ethnomethodology’, includes what is
now commonly referred to as ‘conversation analysis’ or CA. He credits this approach with
having ‘gone a long way toward producing a theory which treats conversation as a coopera-
tive endeavor, subject to systematic constraints’ (1977: 197) and therefore sees its findings
as ‘basic to the study of conversational inference’ (1977: 195). He cites Garfinkel (1967,
1972) for the key insight that ‘social knowledge is revealed in the process of interaction
itself’ (1977: 195). He credits ‘Harvey Sacks and his collaborators’ (specifically Garfinkel,
Schegloff, Jefferson, and Turner) as ‘the first to focus systematically on conversation as
the simplest instance of a naturally organized activity, and to the process of conversational
cooperation as a thing in itself …’ (1977: 195–196). He goes on to say, however, that this
tradition cannot account for many of the essential elements of conversational interaction
that his own approach will: ‘how speakers use verbal skills to create contextual conditions
that reflect particular culturally realistic scenes’; how ‘speakers’ grammatical and phono-
logical knowledge’ is ‘employed in carrying out these strategies’; and how they are ‘able to
recognise culturally possible lines of reasoning’ necessary to follow thematic progression

183
Cynthia Gordon & Deborah Tannen

(1977: 197). Gumperz is thus emphasising the need to bring linguistic and anthropological
perspectives to the work of these sociologists.
Yet another key element of IS is in focus in the first sentence of Discourse Strategies:
‘This book seeks to develop interpretive sociolinguistic approaches to the analysis of real
time processes in face to face encounters’ (1982a: vii). Of significance here is not only the
term ‘real time’ – IS always analyses naturally occurring recorded conversation – but also
the term ‘interpretive’. The significance of this aspect of IS cannot be overestimated. The
interpretive nature of IS analysis is essential because meaning in interaction is emergent,
and listening – that is, gleaning meaning from language – is itself an interpretive process. To
critics, this is the soft underbelly of IS: interpretation cannot be proved. Also problematic is
the difficulty, if not impossibility, of teaching interpretation. Tannen recalls a conversation
in which Labov (personal communication) remarked that anyone who learns his method can
apply it successfully, but the type of analysis developed by Gumperz, like that of Goffman,
depends for its insight and usefulness on the analyst’s perspicacity, which cannot be taught.
In addition to its roots in and connections to sociology, IS also sits at the intersection of
linguistics and anthropology, as does its founder’s own career. Gumperz was on the fac-
ulty of the anthropology department at the University of California, Berkeley, though he
received his PhD in linguistics at the University of Michigan. Many of those whose work
is characterised as IS, including many who trained with Gumperz, are linguistic anthro-
pologists. Indeed, the terms ‘linguistic anthropologist’ and ‘(interactional) sociolinguist’ are
sometimes used interchangeably to refer to the same scholars, and both are used by some
scholars to refer to themselves.

Social justice as an aim of interactional sociolinguistics

Like much early sociolinguistic work, Gumperz’ was founded on the conviction that uncov-
ering the role of linguistic phenomena in face-to-face interaction would contribute to the
cause of social justice. Gumperz developed his approach not only as a way of explain-
ing the interpretive nature of talk but also to address the consequences of real-life misun-
derstandings and misjudgements of abilities and intentions, that can lead to injustice and
discrimination, in particular against ethnic minorities. This is evident in the research sites
where he developed the theories and methods that became IS, where he focused on gate-
keeping encounters, a locus and concept originated by Erickson (1975). In The Counselor
as Gatekeeper, Erickson and Shultz (1982: xi) define gatekeeping encounters as situations
in which ‘two persons meet, usually as strangers, with one of them having authority to
make decisions that affect the other’s future’. This dynamic, and the character of Gumperz’
research site, are illustrated by the photograph on the cover of Language and Social Identity,
the collection of papers by Gumperz and his students and collaborators that was a compan-
ion volume to Discourse Strategies: a Southeast Asian woman wearing a sari sits at the
end of a desk looking up at a white British man sitting behind the desk wearing suit and
tie, holding a pen poised over paper in a notebook or folder. It is evident that she is apply-
ing or asking for something, and that he plays a role in determining whether or not she will
get it. The scene represents the essence of the research Gumperz conducted in London on
multicultural gatekeeping encounters, in which he shows that culturally influenced ways of
using language to signal meaning and intentions result in Southeast Asian speakers being
underestimated and misjudged by British gatekeepers, with the result that Asians often don’t
get benefits they are entitled to or jobs for which they are qualified.

184
 Interactional sociolinguistics

Gumperz (1982a: 175–179) illustrates this process in an exchange between a Pakistani

teacher who has been unable to find permanent work in London and a white British counsel-
lor whose job it is to help him. Gumperz shows that the audio-recorded interaction breaks
down in part because of different expectations about what needs to be said in the interview/
counselling session as well as different expectations regarding uses and interpretations of
prosodic and paralinguistic features like rhythm and intonation. Such features – along with
others that are linguistic (e.g. lexical choice) or extra-linguistic (e.g. eye gaze) – constitute
what Gumperz calls ‘contextualization cues’. These are all the ways speakers indicate and
listeners interpret how speakers mean what they say. Expectations and assumptions about
how to use these signals constitute the ‘sociocultural knowledge’ of the title of Gumperz’
original essay, while listeners’ interpretations of meaning and intentions, based on their
own sociocultural knowledge, are the title’s ‘conversational inference’. In the interview/
counselling session, for example, after a brief discussion about audio-recording, the staff
counsellor uses a rise in pitch to mark a shift in focus to the session proper, thus initiating the
speech activity, providing counselling. Instead of engaging in this activity – the reason he is
there – the teacher produces talk that overlaps with the counsellor’s, which suggests that he
either fails to recognise, or disagrees with, the shift in speech activity, thereby missing her
invitation to begin the counselling he came for. This seemingly trivial mismatch is just one
of a mounting series, reflecting the inability not only to agree on how to initiate counselling
but also to establish rhythmic synchrony and smooth turn-taking. The counsellor repeatedly
uses indirectness to encourage the teacher to state his problem, without success. Gumperz
draws on his own knowledge of Southeast Asian contextualization cues (he did fieldwork in
India and co-authored a grammar of conversational Hindi–Urdu), as well as on the results
of post-recording playback with other Pakistanis, to argue that the teacher’s rhythm and
intonation are signalling that he is seeking the counsellor’s acknowledgement of ‘the gravity
of his situation before he goes on to give more detail’ (1982a: 178). The result of these and
other mismatches in uses and expectations of signalling is that the teacher does not receive
needed support and the counsellor does not effectively do her job.
Such work by Gumperz, as well as by his students, starts from the belief that differ-
ences in how one uses and interprets contextualization cues contribute to, or aggravate,
discrimination, social inequality, and cross-cultural stereotyping – and that by uncovering
and explaining cultural differences and educating people about them, some such misunder-
standings can be avoided or overcome, thereby decreasing the likelihood of unintentional
discrimination or denial of resources. Of course, claiming that such differences contribute
to social injustice does not imply that they are the only contributors. Clearly, discrimina-
tion, prejudice, and economic, as well as social, inequality play enormous roles. Nor does
the hope that raising awareness about the role of such differences will be a starting point
to overcome them and help undo the injustices to which they contribute imply that other
contributing factors will thereby be miraculously erased.

Expansions and additions to interactional sociolinguistic theory

Gumperz’ emphasis on the crucial role of contextualization cues in expressing and interpret-
ing meaning has implications for linguistic theory. By bringing them into analytic focus he
shows that features of language which linguists previously dismissed as ‘marginal’ should
be seen as ‘core’. Signalling how speakers mean what they say, what they think they are
doing when they speak, is every bit as essential, for speakers and hearers, as knowing the

185
Cynthia Gordon & Deborah Tannen

grammar and lexicon of a language. In that sense, interpreting contextualization cues that
signal speech activity is akin to identifying what anthropologist Bateson ([1955]1972)
referred to as ‘frame’, so it is not surprising that one of the most significant ways in which
Gumperz’ IS has been expanded and enriched is the theoretical concept of framing. Just as
Gumperz argued that meaning cannot be gleaned in interaction except by reference to fea-
tures previously dismissed by linguists as ‘marginal’, in Bateson’s view, no utterance or act
can be understood except by reference to a ‘metamessage’ that identifies the frame – that is,
the nature of the interaction – such as whether a monkey biting another monkey is fighting
or playing. Merging Bateson’s and Gumperz’ terminology, one can say that contextualiza-
tion cues – the way the monkey executes the bite, or the way a person utters an insult – can
be understood as sending metamessages that guide interpretation of messages, so a mon-
key knows that another monkey is playing and a person knows that a friend is teasing. As
noted above, Bateson’s concept of framing was elaborated and expanded by Goffman (1974,
1981), who was on the faculty at Berkeley at the same time as Gumperz for a number of
overlapping years (1960–1968). Gumperz’ notion of speech activity is closely related to
that of frame; in later years, Gumperz made this connection explicit: ‘Contextualization
cues, along with other indexical signs, serve to retrieve the frames (in Goffman’s sense of
the term) that channel the interpretive process’ (Prevignano and di Luzio 2003: 10). Further,
they send metamessages regarding the relationships between participants, which relates to
Goffman’s (1981) notions of footing and alignment.
A frequently cited IS application and elaboration of frames theory is Tannen and Wallat’s
([1987]1993) analysis of a paediatric examination/interview. The authors introduce a dis-
tinction between two types of frame: interactive frames and knowledge schemas. The paper
demonstrates the power of frames theory to account for the discourse of a paediatrician who
is examining a child in the presence of the child’s mother. The doctor uses three distinct
registers which identify her addressees: she speaks to the child in ‘baby talk’ or ‘motherese’;
she addresses the mother in a conversational register; and she uses a monotonic ‘reporting
register’ to narrate the findings of her examination, presumably for the benefit of paediatric
residents who will later view the videotape of the interaction. Tannen and Wallat demon-
strate that while the notion of register helps identify whom the doctor is addressing, the
notion of ‘frame’ is necessary to account for and distinguish what the doctor is doing when
she speaks. For example, when she asks the mother for information relevant to the child’s
medical condition, her discourse is part of her examination. But when she answers the moth-
er’s question about an unrelated concern, the doctor must shift frames from examining the
child to consulting with the mother: a cognitive burden that disrupts the examination, as the
doctor attested during playback. The concept of knowledge schema explains why one such
interruption occurred: the mother misinterpreted the ‘noisy’ breathing the child produced
during the doctor’s examination as indicating difficulty breathing, and this required the doc-
tor to put the examination on hold to explain that the breathing was normal for a child with
cerebral palsy. In other words, it was the mismatch in knowledge schemas about cerebral
palsy that occasioned the sudden switch in frames.
The Tannen and Wallat study is not classic IS, insofar as sociocultural differences do
not figure in it: both doctor and mother are white middle-class Americans. The question
of whether ways of speaking are regarded as ‘sociocultural’ or individual is the focus of a
paper co-authored by Gumperz and Tannen (1979) entitled ‘Individual and social differences
in language use’. Analysing examples of interaction in which miscommunication occurs,
the authors posit an implicational hierarchy of levels on which signalling differences can
occur, with level-1 differences, such as confusion about the referent ‘here’, characterised as

186
 Interactional sociolinguistics

individual and level-4 differences characterised as social, as when an Indian English speaker
using prosody and amplitude – speaking emphatically and loudly – to build up to his point
is mistakenly heard by an American as having already made the point and is interrupted.
Interestingly, an example of level-2 differences, uses of indirectness, in this paper illustrates
individual differences, because the speakers are both white East Coast urban professional
men. However, in Conversational Style, Tannen ([1984]2005) cites the same example in a
study extending Gumperz’ notion of social differences to Americans of different regional
and ethnic backgrounds. She argues there that the men’s contrasting uses of indirectness
can be traced to their differing regional and ethnic backgrounds: one is Irish Catholic raised
in Boston, the other East European Jewish raised in New York City. That a single example
could in one study represent individual differences and in the other sociocultural differences
highlights the difficulty of specifying the boundaries of these categories.
Tannen’s notion of ‘conversational style’ expands on and adapts Gumperz’ approach in
another way, too: she folds Lakoff’s (1973) notion of communicative style into Gumperz’ of
conversational inference, thereby accounting not only for the patterns by which speakers signal
and listeners interpret how what is said is meant, but also explaining the ‘logic’ underlying and
driving the stylistic choices associated with each style. Tannen posits, following Lakoff, that
all speakers balance the needs to show involvement and to not impose, but some cultural styles
place more emphasis on showing involvement (hence Tannen calls this style ‘high involve-
ment’) while others place more on not imposing (hence ‘high considerateness’). Seeing con-
versational style features as reflecting these overriding emphases – different ways of observing
similar values and goals – shows that they are not random but patterned. Thus high-involve-
ment style speakers might stand closer, speak more loudly, leave shorter pauses between turns
and use ‘cooperative overlap’ – talking along to show enthusiastic listenership. Like Gumperz,
Tannen assumes and explains the validity of differing styles. She argues that styles are influ-
enced by a range of factors – not just broad cultural or ethnic identities but also regional back-
ground, religion, class, age, profession, gender, sexuality, and many other influences.
A liability associated with all research that addresses cultural patterns is the risk of being
seen as ‘stereotyping’. Does Tannen’s finding that speakers of New York Jewish background
were more likely to stand closer, speak more loudly, and talk along to show enthusiastic
listenership amount to stereotyping New York Jews as aggressive? Our strong conviction
is that the term ‘stereotype’ must be used with caution, and must not be confused with or
applied to a pattern observed. A ‘stereotype’ is a pre-existing assumption based on hearsay
that has been formed before encountering a person or interaction. A research finding based
on observation is the opposite of a stereotype. A research finding which identifies a pattern
that bears a resemblance to a stereotype is not a stereotype. The accusation that it reinforces
a stereotype, on the other hand, must be taken seriously. In our view, Tannen’s observations
of the patterns she describes can help explain and overcome the stereotype. For example,
Tannen showed that the use of cooperative overlap – speaking along to show enthusiasm –
by speakers of East European Jewish background did not constitute interruption in interac-
tions with those who shared a high-involvement style, because they did not stop if they had
more to say. Cooperative overlap only led to interruption when it was mistaken as an attempt
to take the floor. The interruption, in other words, was not the sole doing of the one who
began speaking; rather, it was also the doing of the speaker who stopped mid-turn. This is a
key point on many levels: first, the effect of a way of speaking in cross-style interaction may
result not from a speaker’s intention but rather from style difference between two speakers.
Second, it attests to the IS principle that anything that happens in interaction always results
from the interaction of all participants.

187
Cynthia Gordon & Deborah Tannen

Gumperz’ notion of conversational involvement is also elaborated in Tannen’s Talking

Voices ([1989]2007), where she argues that everyday conversation is made up of the same
linguistic strategies that are artfully shaped and elaborated in literary discourse: repeti-
tion, dialogue, and details that create imagery. As Tannen explains in an introduction to
the second edition (2007), her analysis of repetition, which makes up the first and by far
the longest analytic chapter in the book, is synonymous with the phenomenon that is now
routinely referred to as ‘intertextuality’ and is playing an increasingly central role in IS.
Gordon (2009) further develops this focus by demonstrating the interrelation between inter-
textuality and framing in her study of the discourse of three families, each consisting of a
mother, a father, and a child under age five. In this and related studies (Gordon 2004, 2006),
she explores how and why family members repeat one another’s words in everyday talk
as well as the interactive effects of those repetitions. Importantly, Gordon (2009) demon-
strates the function of repetition in creating and managing various forms of what Goffman
(1974) refers to as ‘laminations’ of frames. In addition to showing the complex means by
which speakers create different configurations of frames, she also shows how intertextuality
interplays with framing to create the shared meanings that give each family its distinctive
identity. For example, she demonstrates how a mother blends parenting and play frames
in how she encourages her nearly three-year-old daughter to, for example, get into her car
seat, speak politely, and get ready for naptime: the mother issues directives by using char-
acter names and other material from a storybook that she often reads to the little girl, thus
accomplishing both parenting and play. Gordon also suggests that interactions such as these
construct aspects of the family’s identity (e.g. as centring around literacy). In the next sec-
tion, we describe how select studies in IS have explored discourse as related to identities
pertaining to gender and sexuality.
The theoretical framework of IS has thus expanded and evolved in the years since
Gumperz first devised what he called his theory of conversational inference, and the
approach has been applied to ever-expanding domains of interaction. With this overview
of IS as a foundation, we turn now to the topic of this Handbook, describing how IS has
been used to explore how identities related to gender and sexuality are represented and
constructed in discourse.

Language, gender, and sexuality

Cross-gender communication as cross-cultural communication
A particularly influential as well as early application of IS to gender and discourse is the
chapter by Maltz and Borker in Language and Social Identity (1982). Spending time with
Gumperz at the University of California, Berkeley during the key years when he was devel-
oping IS, the authors, both anthropologists, concluded that Gumperz’ framework of cross-
cultural miscommunication could account for a range of findings in the then-nascent but
about to burgeon field of gender and language. They recap patterns identified in the exist-
ing literature regarding how American girls and boys, and women and men, tend to use
language. They then cite studies, such as those by Goodwin (1980) and Lever (1976), of
children’s play to support the view that boys and girls grow up in what can be seen as dif-
ferent cultures in the sense that they tend to play in sex-separate groups and to be treated
differently. Maltz and Borker then suggest that the differences in how women and men
use language can be traced to the gender-inflected sociolinguistic subcultures in which
they are socialised. In other words, they learn to use and interpret contextualization cues

188
 Interactional sociolinguistics

differently, and differently conceptualise certain speech activities. Brief, simple examples
are Hirschman’s ([1973]1994) observation that women tend to use more minimal responses,
especially ‘mm hmm’, and Fishman’s (1978: 402) that women are more likely to utter such
feedback throughout another’s talk rather than at a turn’s end. Maltz and Borker hypothesise
that women often use these minimal responses to indicate that they’re listening, whereas
men may use them to indicate agreement, or at least, ‘I follow your argument so far’.
Therefore, the authors argue, women may use more of them because they are listening more
often than men are agreeing. Furthermore, a man may get the impression that a woman has
been agreeing when she has simply been listening, and a woman may get the impression that
a man isn’t listening when he has been listening but doesn’t agree. They note, moreover, that
these impressions fit in with the larger complaints that members of each subculture tend to
make of the other: women often complain that men do not listen, and men often complain
that women are unpredictable. They thus apply Gumperz’ insight that ill effects in interac-
tion may result from the differing expectations and uses of contextualization cues rather
than from ill intentions. It is important to emphasise here that this does not imply that no one
ever has ill intentions, only that the impression of ill intentions in some instances may be
the unintended result of these differences. This brief article was significant in demonstrating
the applicability of IS theory and method to contexts beyond those Gumperz addressed, and
was influential in its specific extension of IS to language and gender.
Maltz and Borker’s article became the basis for Tannen’s application of Gumperz’ frame-
work to communication between women and men in her general-audience book You Just
Don’t Understand (1990). Around the time that Tannen was working on it, she took part in
a study of children’s dyadic conversations with their best friends that was spearheaded by
Bruce Dorval, who video-recorded the children’s conversations in his office. Tannen noticed
that at every age, the girls looked directly at each other and maintained that face-to-face
gaze whereas at every age the boys sat at angles or parallel and looked around the room.
These observations, recapped in a chapter of You Just Don’t Understand, correspond to
Maltz and Borker’s interpretation of Hirschman’s and Fishman’s findings: women often get
the impression that men are not listening because they are not maintaining gaze, and men
may feel wrongly accused of not listening when they were. (See Tannen 1994a for analyses,
based on these and other findings, written for academic audiences.)
Tannen’s (1994b) next book, Talking from 9 to 5, applies IS method as well as theory in
that it is based on extended ethnographic observation and the recording of naturally occur-
ring interaction. Tannen worked with two large corporations, one in California and one
in New York State, to identify four or five managers who carried or wore audio recorders
and recorded everything they felt comfortable recording at work for a week. After they
had recorded, she shadowed them and interviewed their peers, superiors, and subordinates.
Tannen documents how the women in her study are caught in what Lakoff (1975) identified
as a ‘double bind’ confronting all women in positions of authority. (The concept of double
bind, which traces to Bateson, is a no-win situation in which someone is faced with two
requirements, but anything they do to fulfil one violates the other. For women in authority,
anything they do to fulfil the expectations that a good woman is, for example, diffident and
self-effacing, violates the expectations that a person in authority be confident and asser-
tive. If they talk in ways expected of women, they are liked but seen as lacking confidence
and even competence. If they talk in ways expected of people in authority, they are seen as
too aggressive.) A condensation of this book appears as an article in the Harvard Business
Review (Tannen 1995). For further discussion of this double bind see also Appleby, this
volume.

189
Cynthia Gordon & Deborah Tannen

Gender and talk at home and at work

As the preceding sections illustrate, there are many language and gender studies that take
an IS approach to examine interactions at home and in the workplace. (Workplace studies
have been particularly numerous, including those by Angouri, Marra, and Dawson, and
by Schnurr and Omar, this volume, and the large body of work by Holmes and her col-
laborators (e.g. Holmes and Marra 2004).) A case study by Kendall (2003) is unique and
particularly revealing in comparing a woman’s discourse across these two domains. Using
a framing approach, Kendall explores the contrasting ways that the woman, Elaine, con-
structs authority at home, as a parent to her ten-year old daughter, and at work, as the
manager of two women subordinates, where Kendall documents how the woman manages
the double bind. Drawing on Goffman’s (1967: 83) notions of deference and demeanour as
well as Davies and Harré’s (1990) positioning theory, Kendall shows that Elaine creates a
‘benevolent demeanour of authority’ at work by issuing directives to her subordinates in
face-saving ways, positioning them as equals who will engage in a joint activity, such as by
using ‘let’s’. Strikingly, Kendall notes that Elaine ‘draws on mitigating strategies that evoke
the qualities associated with sociocultural concepts of “mother”; however, she does not use
these strategies to the same extent to “do” her identity as a mother’ (p. 13). At home, Elaine
creates a demeanour of ‘explicit authority’ by, for example, issuing unmitigated directives
in imperative form to her daughter during dinnertime, such as, ‘Just spoon in that, and stir it
around’ (p. 608). In a related study, Kendall (2008) zooms in on the dinnertime interactions,
demonstrating how Elaine and her husband create gendered identities through the number
and kinds of positions they take up in different frames. Elaine accomplishes many tasks,
such as monitoring their daughter’s etiquette and behaviour, and managing dinner prepara-
tion and clean-up, while her husband mainly positions himself as family comedian. Through
this unequal division of labour, as well as the gendered nature of their different tasks, the
adults create gendered parental identities. Simultaneously, Kendall shows, they construct
other social identities (e.g. authority on cooking, child’s teacher); she thus demonstrates that
gendered identities are multifaceted, and intersect with other aspects of identity.
Kendall also analysed conversations between parents as part of the larger study of family
interaction of which Gordon, cited above, is also a part. (See Tannen, Kendall, and Gordon
2007 for a detailed description.) The design of this study is modelled on the one Tannen
developed for the research behind Talking from 9 to 5. Parents in dual-income, roughly
middle-class American families (all of whom were white and had opposite-sex partners)
carried or wore digital audio recorders over the course of approximately one week, record-
ing as much as they felt comfortable doing at work and at home. Post-recording, they were
shadowed by a member of the research team for at least one day (Kendall and Gordon both
observed), and participated in playback with the researchers. Kendall (2007) used these
recordings and transcriptions to explore how members of two couples create gendered posi-
tions in conflicting discourses (or ideologies) of work and family: on one hand, the tradi-
tional ideology by which women are primary caregivers and men are breadwinners and
on the other, the feminist discourse by which women and men share equally the roles of
primary caregiver and breadwinner. The couples whose discourse she analysed espoused the
feminist discourse: it was important to them that both worked and both spent time with their
child as primary caregiver. However, their everyday interactions and conversations often
positioned them within the traditional discourse. For example, one mother is positioned as
primary caregiver when she overrules directives issued by her husband to their child during
potty-training; elsewhere, in a conversation with her friends, she describes her work outside

190
 Interactional sociolinguistics

the home as a chance to be ‘be stimulated’ and her husband’s as necessary for the family ‘to
survive’, thus positioning him as breadwinner.
Gordon has also explored gender in the discourse of these families. For example,
extending theorising by Ochs (1993), Gordon (2007) examines the acts and stances the
same mother performs in an interaction wherein her brother (the child’s uncle) explains
to the couple how their daughter misbehaved while he babysat her. Gordon shows that
the mother performs acts and takes up stances that are socioculturally linked to the iden-
tities of ‘parent’, ‘woman’, and particularly ‘mother’, by requesting details about her
daughter’s day; providing details about her child’s life in response to the report; assessing
her daughter’s and her brother’s depicted behaviours; and accounting for her daughter’s
misbehaviour.
The topic of language and gender has also been explored by scholars drawing on IS
in other contexts, both personal/social and workplace/institutional, and in various cul-
tures. We mention only a sampling here. Language and gender has been explored in such
interpersonal interactions as talk among adult women friends in Greece (Georgakopoulou
2005) and Germany (Günthner 1997; Kotthoff 2000); among college fraternity men in
the US (Kiesling 2001); among US American and Chinese preschool children (Kyratzis
and Guo 2001); between members of a mixed-gender pair of Japanese college friends
(Itakura 2015); and between a pair of preschool-aged British girls engaged in pretend
play (Cook-Gumperz 1995). Workplace/institutional contexts wherein language and gen-
der are examined include Brazilian all-female police stations and feminist crisis cen-
tres (Ostermann 2003); psychiatric interviews in Brazil (Ribeiro 2002); discourse among
men and women in the US Air Force (Disler 2008); employment interviews of female
engineering students in New Zealand (Reissner-Roubicek 2012); German-language coun-
selling/advising conversations among German advisors and Chinese research scientists
seeking guidance (Günthner 1992); interactions among employees at an American radio
network (Kendall 2004); conversations among members of an all-women management
team in the UK (Baxter 2014); everyday discourse in diverse types of New Zealand work-
places (e.g. Holmes 2006); and in multinational companies situated in Europe (Angouri
2018). Gendered styles of language use and identity construction have also been consid-
ered through the lens of IS in contexts beyond work and home, though less frequently,
including in news, entertainment, and social media contexts, such as in televised debates
(Kotthoff 1997), reality TV (Gordon 2015), online discussion boards (Gordon and İkizoğlu
2017), and online product reviews (Vásquez and China 2019). As indicated by these stud-
ies, as well as those briefly summarised above and many we have not specifically cited, IS
has proved a frequent and fruitful source of insight into gendered patterns of interaction.
Language and sexuality is a newer field and therefore includes fewer IS studies. Hence
the next and last section is brief, but will give a sense of work that has been done and of
work that might be done in the future.

Language and sexuality

There is a limited body of research in interactional sociolinguistics that focuses on sexu-
ality, as compared to gender. Kiesling’s (2001) work on fraternity men’s discourse, and
Georgakopoulou’s (2005) on the discourse of women friends both address heterosexual-
ity as relevant to the construction of gendered identities in their data. Wang (2020) draws
on IS to examine the construction of racialised sexuality in online text-based communica-
tion. Seals’ (this volume) analysis of a stand-up routine also addresses sexuality. To date,

191
Cynthia Gordon & Deborah Tannen

however, more research on language and sexuality has been conducted by scholars tak-
ing kindred context-bound approaches to the analysis of interaction, such as Bucholtz and
Hall’s (2005) sociocultural linguistic approach to the study of identity construction. For
instance, Wagner (2010) examines the discourse of lesbian families in the US; Hall (e.g.
Hall and O’Donovan 1996) explores language use by Hindi-speaking hijras (identified by
anthropologists as members of a ‘third gender’) in northern India, and Gaudio (2009) stud-
ies the discourse of feminine men in a Hausa-speaking Islamic city of northern Nigeria.
Collections have emerged that use qualitative sociolinguistic and anthropological linguis-
tic perspectives to investigate connections between gender and sexuality in language use
among people who self-identify as queer, transgender, or non-binary (e.g. Zimman, Davis,
and Raclaw 2014). We look forward to future studies applying theories of conversational
inference, discourse strategies, and framing and positioning, as developed in IS, to the study
of language and sexuality.

Conclusion
We have tried in this essay to provide an overview of ‘interactional sociolinguistics’ from
its inception in the pioneering work of Gumperz to the rich and expansive body of work
now commonly referred to by the term, as it is now used to characterise a wide range of
context-sensitive microanalysis of interaction. In revisiting the roots and development of
the approach, we have highlighted not only how it has blossomed over the years, but also
the potential and promise it holds for scholars who seek a qualitative, interpretive method to
explore language, gender, and sexuality.

Notes
1 Each author has written a lengthy essay focusing on IS which includes information relevant to this
joint essay. Gordon’s (2011) is an introduction to Gumperz’s academic biography, the motivations
that contributed to his development of IS, and key research trajectories in the areas of his research.
Tannen’s (2004) introduces the terms and concepts that characterise the theories and methods of IS,
then presents sample analyses to illustrate their application.
2 Tannen actually helped Gumperz write this paper, as well as two others that he drew on for Discourse
Strategies. When she was a graduate student in linguistics at the University of California, Berkeley,
Tannen was hired by Gumperz full time for a term to help him write this paper and another entitled
‘The conversational analysis of interethnic communication’ (Gumperz 1978). She later helped him
write ‘The sociolinguistic basis of speech act theory’ (Gumperz 1981), parts of which he adapted
for inclusion in Chapter 8. Finally, Chapter 6, ‘Contextualization conventions’, draws heavily from
the paper Gumperz and Tannen (1979) co-authored entitled ‘Individual and social differences in
language use’. The collaboration is acknowledged in the notes to each of these publications.

References
Angouri, J. (2018) Culture, discourse and the workplace. Abingdon: Routledge.
Bateson, G. ([1955]1972) ‘A theory of play and fantasy’, in Steps to an ecology of mind. New York:
Ballantine, pp. 177–193.
Baxter, J. (2014) ‘“If you had only listened carefully…”: the discursive construction of emerging
leadership in a UK all-women management team’, Discourse & Communication, 8(1), pp. 23–39.
Bucholtz, M. and Hall, K. (2005) ‘Identity and interaction: a sociocultural linguistic approach’,
Discourse Studies, 7(4–5), pp. 585–614.

192
 Interactional sociolinguistics

Cook-Gumperz, J. (1995) ‘Reproducing the discourse of mothering: how gendered talk makes
gendered lives’, in Hall, K. and Bucholtz, M. (eds) Gender articulated: language and the socially
constructed self. New York: Routledge, pp. 401–419.
Davies, B. and Harré, R. (1990) ‘Positioning: the discursive production of selves’, Journal for the
Theory of Social Behaviour, 20(1), pp. 43–63.
Disler, E. A. (2008) Language and gender in the military: honorifics, narrative, and ideology in air
force talk. Amherst, NY: Cambria Press.
Erickson, F. (1975) ‘Gatekeeping and the melting pot: interaction in counselling encounters’, Harvard
Educational Review, 45(1), pp. 44–70.
Erickson, F. and Shultz, J. (1982) The counselor as gatekeeper: social interaction in interviews. New
York: Academic Press.
Fillmore, C. (1977) ‘The case for case reopened’, in Cole, P. and Sadock, J. (eds) Syntax and semantics,
vol. 8. New York: Academic Press, pp. 59–81.
Fishman, P. (1978) ‘Interaction: the work women do’, Social Problems, 25(4), pp. 397–406.
Garfinkel, H. (1967) Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Garfinkel, H. (1972) ‘Studies of the routine grounds of everyday activities’, in Sudnow, D. (ed) Studies
in social interaction. New York: Free Press, pp. 1–30.
Gaudio, R. P. (2009) Allah made US: sexual outlaws in an Islamic African city. Chichester: John
Wiley & Sons.
Georgakopoulou, A. (2005) ‘Styling men and masculinities: interaction and identity aspects at work’,
Language in Society, 34(2), pp. 163–184.
Goffman, E. (1956) The presentation of self in everyday life. New York: Doubleday-Anchor.
Goffman, E. (1967) Interaction ritual: essays on face-to-face behavior. New York: Pantheon Books.
Goffman, E. (1974) Frame analysis. New York: Harper and Rowe.
Goffman, E. (1981) Forms of talk. Philadelphia: University of Pennsylvania Press.
Goodwin, M. H. (1980) ‘Directive-response speech sequences in girls’ and boys’ task activities’, in
McConnell-Ginet, S., Borker, R., and Furman, N. (eds) Women and language in literature and
society. New York: Praeger, pp. 157–173.
Gordon, C. (2004) ‘“Al Gore’s our guy”: linguistically constructing a family political identity’,
Discourse & Society, 15(4), pp. 607–631. Reprinted in Tannen, Kendall, and Gordon (eds) pp.
233–262.
Gordon, C. (2006) ‘Reshaping prior text, reshaping identities’, Text & Talk - An Interdisciplinary
Journal of Language, Discourse Communication Studies, 26(4/5), pp. 545–571.
Gordon, C. (2007) ‘“I just feel horribly embarrassed when she does that”: constituting a mother’s
identity’, in Tannen, D., Kendall, S., and Gordon, C. (eds) Family talk: discourse and identity in
four American families. Oxford: Oxford University Press, pp. 71–102.
Gordon, C. (2009) Making meanings, creating family: intertextuality and framing in family interaction.
New York: Oxford University Press.
Gordon, C. (2011) ‘Gumperz and interactional sociolinguistics’, in Wodak, R., Kerswill, P. E., and
Johnstone, B. (eds) The SAGE handbook of sociolinguistics. Los Angeles: Sage, pp. 67–84.
Gordon, C. (2015) ‘“We were introduced to foods I never even heard of”: parents as consumers
on reality television’, in Demo, A. T., Borda, J. L., and Kroløkke, C. H. (eds) The motherhood
business: consumption, communication and privilege. Tuscaloosa, AL: University of Alabama
Press, pp. 95–120.
Gordon, C. and İkizoğlu, D. (2017) ‘“Asking for another’’ online: membership categorization and
identity construction on a food and nutrition discussion board’, Discourse Studies, 19(3), pp.
253–271.
Gordon, D. and Lakoff, G. (1973) ‘Conversational postulates’, in Cole, P. and Morgan, J. (eds) Syntax
and semantics, vol. 3. New York: Academic Press, pp. 63–84.
Gumperz, J. J. (1977) ‘Sociocultural knowledge in conversational inference’, in Saville-Troike,
M. (ed) Linguistics and anthropology, Georgetown University round table on languages and
linguistics 1977. Washington, DC: Georgetown University Press, pp. 191–211.

193
Cynthia Gordon & Deborah Tannen

Gumperz, J. J. (1978) ‘The conversational analysis of interethnic communication’, in Lamar Ross, E.

(ed) Interethnic communication (southern anthropological society proceedings), vol. 12. Athens,
GA: University of Georgia Press, pp. 13–31.
Gumperz, J. (1981) ‘The sociolinguistic basis of speech act theory’, in Boyd, J. and Ferrara, S. (eds)
Speech act ten years after. Milan: Versus, pp. 101–121.
Gumperz, J. J. (1982a) Discourse strategies. Cambridge: Cambridge University Press.
Gumperz, J. J. (ed) (1982b) Language and social identity. Cambridge: Cambridge University Press.
Gumperz, J. J. (2001) ‘Interactional sociolinguistics: a personal perspective’, in Schiffrin, D., Tannen,
D., and Hamilton, H. (eds) The handbook of discourse analysis. Malden, MA: Blackwell, pp.
215–228.
Gumperz, J. J. and Hymes, D. (eds) (1972) Directions in sociolinguistics: the ethnography of
communication. New York: Holt, Rinehart, Winston.
Gumperz, J. J. and Tannen, D. (1979) ‘Individual and social differences in language use’, in Fillmore,
C., Kempler, D., and Wang, W. S.-Y. (eds) Individual differences in language ability and language
behavior. New York: Academic Press, pp. 305–324.
Günthner, S. (1992) ‘The construction of gendered discourse in Chinese-German interactions’,
Discourse & Society, 3(2), pp. 167–191.
Günthner, S. (1997) ‘Complaint stories. Constructing emotional reciprocity among women’, in
Kotthoff, H. and Wodak, R. (eds) Communicating gender in context. Amsterdam: John Benjamins,
pp. 179–218.
Hall, K. and O’Donovan, V. (1996) ‘Shifting gender positions among Hindi-speaking hijras’, in
Bergvall, V. L., Bing, J. M., and Freed, A. F. (eds) Rethinking language and gender research:
theory and practice. New York: Longman, pp. 228–266.
Halliday, M. A. K. and Hasan, R. (1976) Cohesion in English. London: Longman.
Hirschman, L. ([1973]1994) ‘female-male differences in conversational interaction’, Language in
Society, 23(3), pp. 427–442.
Holmes, J. (2006) Gendered talk at work. Malden, MA: Blackwell.
Holmes, J. and Marra, M. (2004) ‘Relational practice in the workplace: women’s talk or gendered
discourse?’, Language in Society, 33(3), pp. 377–398.
Hymes, D. (1962) ‘The ethnography of speaking’, in Gladwin, T. and Sturtevant, W. C. (eds)
Anthropology and human behavior. Washington, DC: Anthropological Society of Washington, pp.
13–53.
Itakura, H. (2015) ‘Constructing Japanese men’s multidimensional identities: a case study of mixed-
gender talk’, Pragmatics. Quarterly Publication of the International Pragmatics Association,
25(2), pp. 179–203.
Kendall, S. (2003) ‘Creating gendered demeanors of authority at work and at home’, in Holmes, J.
and Meyerhoff, M. (eds) The handbook of language and gender. Malden, MA: Blackwell, pp.
600–623.
Kendall, S. (2004) ‘Framing authority: gender, face, and mitigation at a radio network’, Discourse &
Society, 15(1), pp. 55–79.
Kendall, S. (2007) ‘Father as breadwinner, mother as worker: gendered positions in feminist and
traditional discourses of work and family’, in Tannen, D., Kendall, S., and Gordon, C. (eds)
Family talk: discourse and identity in four American families. Oxford: Oxford University Press,
pp. 123–163.
Kendall, S. (2008) ‘The balancing act: framing gendered and parental identities at dinnertime’,
Language in Society, 37(4), pp. 539–569.
Kiesling, S. F. (2001) ‘“Now I gotta watch what I say”: shifting constructions of masculinity in
discourse’, Journal of Linguistic Anthropology, 11(2), pp. 1–24.
Kotthoff, H. (1997) ‘The interactional achievement of expert status: creating asymmetries by “Teaching
conversational lectures” in TV discussions’, in Kotthoff, H. and Wodak, R. (eds) Communicating
gender in context. Amsterdam: John Benjamins, pp. 139–178.

194
 Interactional sociolinguistics

Kotthoff, H. (2000) ‘Gender and joking: on the complexities of women’s image politics in humorous
narratives’, Journal of Pragmatics, 32(1), pp. 55–80.
Kyratzis, A. and Guo, J. (2001) ‘Preschool girls’ and boys’ verbal conflict strategies in the United
States and China,’, Research on Language & Social Interaction, 34(1), pp. 45–74.
Lakoff, R. (1973) ‘The logic of politeness, or minding your p’s and q’s’, in Corum, C., Smith-Clark, T.
C., and Weiser, A. (eds) Papers from the ninth regional meeting of the Chicago linguistics society.
Chicago: University of Chicago Department of Linguistics, pp. 292–305.
Lakoff, R. (1975) Language and woman’s place. New York: Harper and Row.
Lever, J. (1976) ‘Sex differences in the games children play’, Social Problems, 23, pp. 478–483.
Maltz, D. N. and Borker, R. A. (1982) ‘A cultural approach to male-female miscommunication’, in
Gumperz, J. J. (ed) Language and social identity. New York: Cambridge University Press, pp.
196–216.
Ochs, E. (1993) ‘Constructing social identity: a language socialization perspective’, Research on
Language & Social Interaction, 26(3), pp. 287–306.
Ostermann, A. C. (2003) ‘Localizing power and solidarity: pronoun alternation at an all-female police
station and a feminist crisis intervention center in Brazil’, Language in Society, 32(3), pp. 351–381.
Prevignano, C. L. and di Luzio, A. (2003) ‘A discussion with John J. Gumperz’, in. Erdmans, S. L.,
Prevignano, C. L., and Thibault, P. J. (eds) Language and interaction: discussions with John J.
Gumperz. Amsterdam: John Benjamins, pp. 7–29.
Reissner-Roubicek, S. (2012) ‘“The guys would like to have a lady”: the co-construction of gender and
professional identity in interviews between employers and female engineering students’, Pragmatics.
Quarterly Publication of the International Pragmatics Association, 22(2), pp. 231–245.
Ribeiro, B. T. (2002) ‘Erotic discourse strategies in powerless women: analysing psychiatric
interviews’, in Litosseliti, L. and Sunderland, J. (eds) Gender identity and discourse analysis.
Amsterdam: John Benjamins, pp. 193–219.
Schiffrin, D. (1993) ‘“Speaking for another” in sociolinguistic interviews: alignments, identities,
and frames’, in Tannen, D. (ed) Framing in discourse. New York: Oxford University Press, pp.
231–263.
Schiffrin, D. (1994) Approaches to discourse. Malden, MA: Blackwell.
Tannen, D. (1990) You just don’t understand: women and men in conversation. New York: Morrow.
Tannen, D. (1994a) Gender and discourse. Oxford: Oxford University Press.
Tannen, D. (1994b) Talking from 9 to 5: how women’s and men’s conversational styles affect who gets
heard, who gets credit, and what gets done at work. New York: William Morrow.
Tannen, D. (1995) ‘The power of talk: who gets heard and why’, Harvard Business Review, 73(5),
pp. 138–148.
Tannen, D. (2004) ‘Interactional sociolinguistics’, in Ammon, U., Dittmar, N., Mattheier, K. J., and
Trudgill, P. (eds) Sociolinguistics: an international handbook of the science of language and
society. Berlin: de Gruyter, pp. 76–88.
Tannen, D. ([1984]2005) Conversational style: analyzing talk among friends. New York: Oxford
University Press. (Originally published by Ablex.)
Tannen, D. ([1989]2007) Talking voices: repetition, dialogue, and imagery in conversational
discourse, 2nd edn. Cambridge: Cambridge University Press.
Tannen, D., Kendall, S., and Gordon, C. (eds) (2007) Family talk: discourse and identity in four
American families. New York: Oxford University Press.
Tannen, D. and Wallat, C. (1987) ‘Interactive frames and knowledge schemas in interaction: examples
from a medical examination/interview’. Social Psychology Quarterly 50(2), pp. 205–216. Reprinted
in Tannen, D. (ed) (1993) Framing in discourse. Oxford: Oxford University Press, pp. 57–76.
Vásquez, C. and China, A. S. (2019) ‘From “my manly husband…” to “sitting down to take a pee”:
the construction and deconstruction of gender in Amazon reviews’, in Bou-Franch, P. and Garcés-
Conejos Blitvich, P. (eds) Analyzing digital discourse: new insights and future directions. Cham:
Palgrave Macmillan, pp. 193–218.

195
Cynthia Gordon & Deborah Tannen

Wagner, S. (2010) ‘Bringing sexuality to the table: language, gender and power in seven lesbian
families’, Gender & Language, 4(1), pp. 33–72.
Wang, P. H. (2020) ‘Negotiating racialized sexuality through online stancetaking in text-based
communication’, in Farris, D. N., Compton, D. R., and Herrera, A. P. (eds) Gender, sexuality, and
race in the digital age. Cham: Springer, pp. 187–203.
Zimman, L., Davis, J., and Raclaw, J. (eds) (2014) Queer excursions: retheorizing binaries in language
gender, and sexuality. New York: Oxford University Press.

196
 13
Leadership and humour at work
Using interactional sociolinguistics to explore
the role of gender

Stephanie Schnurr and Nor Azikin Mohd Omar

Introduction
This chapter explores language and gender in the professional domain with a focus on
leadership discourse. Leadership is a particularly relevant topic for language and gender
research as the notion of leadership is characterised by a male bias (e.g. Martin Rojo and
Esteban 2003), and globally prevailing gender stereotypes and expectations – which assume
leaders to be white, male, and handsome (e.g. Clifton et al. 2019) – continue to have real-
life implications for women and men taking up leadership roles or aspiring to do so. In
our analysis of leadership discourse and gender, we focus on humour, which is one of the
discursive strategies frequently used by leaders to achieve their various objectives (e.g.
Holmes 2007; Schnurr 2009b). As an analytical and methodological approach we employ
interactional sociolinguistics (IS), which facilitates the identification of humour as part of
a fine-grained analysis of authentic interactional data, while at the same time taking into
account the context in which an encounter occurs. In what follows, we first briefly outline
some of the recent trends in research on leadership discourse and gender with a particular
emphasis on humour, before introducing IS in more detail. We discuss some of the benefits
of IS for the study of leadership and humour, and we analyse several examples of authentic
leadership discourse to illustrate how IS can be applied, and how it guides, supports, and
facilitates our understanding of the role of gender in this context. We end the chapter with a
brief discussion of our findings, some remarks for future research, and a list of recommenda-
tions for further reading.

Gender, leadership, and humour – a brief overview

Much has been written on leadership and gender, and researchers from different disciplines
have repeatedly pointed out that leadership is a gendered concept and that expectations
of what is considered to be a good leader and effective leadership are strongly affected
by gender stereotypes (e.g. Billing and Alvesson 2000; Holmes 2006; Schnurr 2008). Not
only are many of the attributes typically associated with leadership almost synonymous
with stereotypical views of masculinity, such as being authoritative, decisive, competitive,

197
Stephanie Schnurr & Nor Azikin Mohd Omar

goal-oriented, hard-nosed etc. (Holmes 2006: 34), but the mantra ‘think leader – think male’
(Koenig et al. 2011) seems to prevail despite research that celebrates other, ‘more femi-
nine’ ways of doing leadership (e.g. Baxter 2010; Chesterman et al. 2005). However, these
‘other’ ways of doing leadership, which centre around collaboration, cooperation, empathy,
and long-term thinking – attributes stereotypically associated with femininity – are often
only perceived as valuable when displayed by men (Baxter 2012; Singh and Vinnicombe
2002). Yet, they tend to be overlooked or regarded as something else (and less valuable)
than leadership when displayed by women (Fletcher 2001; Rutherford 2001; Sinclair 1998).
Thus, although considerable evidence gathered in numerous studies on leadership and gen-
der in different sociocultural contexts shows that no direct link exists between gender and
actual leadership performance, these stereotypes prevail, and men and women tend to be
perceived very differently to the extent that the same behaviour is often evaluated differently
(e.g. Fletcher 2004; Holmes 2006; Schnurr 2009b). Moreover, in spite of an increase in the
number of women in leadership positions worldwide (according to the latest OECD (2016)
report, there is a 20% increase on average of women in leadership positions1), still only a
very limited set of roles are available to women leaders, which heavily draw on and thus
reinforce gender stereotypes by categorising women leaders as either a mother, a seductress,
an iron maiden, or a pet (Baxter 2012, 2017; Kanter 1977). All these observations point
to the fact that in spite of recent positive developments in terms of increasing visibility of
women in more senior positions, and the often cited feminisation of leadership or ‘female
advantage’ of doing leadership in some industries (Eagly and Carli 2003; Fletcher 2004),
women remain disadvantaged – both in terms of accessing leadership positions and in terms
of how they are being perceived and evaluated for their leadership work.

Leadership and humour

Among the various strategies that have been described as indexing leadership, humour
is receiving a considerable amount of attention (e.g. Avolio et al. 1999; Collinson 2006;
Holmes 2007; Holmes and Marra 2006; Hoption et al. 2013; Rogerson-Revell 2011; Schnurr
2009a, 2009b; Schnurr and Chan 2009, 2011, 2017). This continued interest in this discur-
sive strategy is perhaps not surprising given its versatility and complexity. Humour has been
found to be a valuable tool to assist those in leadership positions to achieve several leader-
ship objectives, such as getting things done, motivating the team, solving conflicts and man-
aging disagreements, creating group-cohesiveness, and making criticisms more palatable,
etc. For example, conducting case studies of six leaders in three internet technology (IT)
companies in New Zealand, Schnurr (2009b) illustrates how humour may assist those in
leadership positions to achieve various transactional and relational objectives – sometimes
even simultaneously. Similarly, Mullany (2004) explores how meeting chairs use humour
when attempting to gain their subordinates’ compliance in managerial business meetings
recorded in two businesses in the UK. Rogerson-Revell (2011) analyses the role of humour
with regard to the leadership styles displayed by chairs of meetings in international business
meetings in large corporations in Hong Kong and the EU. Holmes and Marra (2006) also
illustrate the multiple functions that humour may perform with regard to leadership, and dis-
cuss the ways in which the leaders’ use of humour contributes to and reflects specific aspects
of their leadership styles. They argue that humour is particularly useful for the performance
of leadership as it enables the leaders to exercise both power and politeness – often at the
same time. They provide several examples to show that ‘[s]killful leaders recognise the

198
 Leadership and humour at work

transformational potential of humour, as well as its value in helping achieve transactional

objectives, and they exploit its varied functions to create team, smooth ruffled feathers and
generate creative energy’ (Holmes and Marra 2006: 133). They argue that the ways in which
the leaders (and their subordinates) use humour is a reflection (and at the same time active
contribution) to the culture of their workplaces (see also Schnurr 2009b).

Humour and gender

Humour is not only a valuable tool for leaders, but, like the concept of leadership, it is also
heavily affected by gender stereotypes. Among the most common gender stereotypes are the
claims that women often ‘don’t get a joke’, that they use and appreciate humour less than
men (Kotthoff 2006: 2), and that they prefer different types of humour (Tannen 1994). For
example, in an early study on a group of white, middle-class American mothers who regu-
larly meet, Jenkins (1985) claims that the humour used by these women is cooperative, sup-
portive, inclusive, self-healing, integrated, and spontaneous, while the humour produced by
men is more exclusive, challenging, pre-formulated, and self-aggrandising. Similar trends
were observed by Hay (2000) who analysed the humour of men and women in informal
friendship groups in New Zealand. Like Jenkins (1985), she also observed considerable
gender differences and found the women to employ humour more frequently than the men
to create solidarity, whereas men used it more often to ‘highlight similarities or capitalise
on shared experiences’ (Hay 2000: 734). Interestingly, she found no noticeable differences
in the extent to which the women and men in her sample used teasing. Tannen (1994: 72),
drawing on various studies, also argued that ‘[t]he types of humor men and women tend to
prefer differ’. While men predominantly use ‘razzing, teasing, and mock-hostile attacks’,
women mainly employed self-denigrating humour (Tannen 1994: 72–73).
Research on humour and gender in the workplace is relatively recent (for an exception
see Tannen 1994), and most of these more recent studies challenge the findings of earlier
research on gender stereotypes regarding the use of humour in friendship groups. For exam-
ple, Mullany (2004: 13) observed that in contrast to the stereotype of humourless women,
the female chairs in her study frequently used repressive humour (i.e. humour through which
the speaker exercises (often subtle) control) ‘as a mitigation strategy to attempt to gain the
compliance of their subordinates’, while the male chairs did not adopt this strategy at all.
Similarly, in a study of two female leaders in small factory outlets in Hong Kong, Ladegaard
(2012) reports evidence of the women regularly using jocular abuse to maintain their hier-
archical position in the workplace. These observations challenge earlier claims about the
supportive, collaborative, self-denigrating, and inclusive nature of women’s humour. For
instance, Holmes (2006: 40), analysed the occurrence of humour in over 20 meetings in
private and governmental organisations in New Zealand and found no evidence for gender
differences. She rather observed ‘that both women and men used conjoint humour support-
ively, to elaborate or expand a previous contribution, and contestively, to challenge anoth-
er’s humorous assertion’. However, despite these similarities in men’s and women’s use
of humour, Holmes also observed that in meetings where women were present, supportive
conjoint humour tended to occur more frequently, while contestive joined humour was more
typically used in meetings which involved men. All these studies thus point to inconclusive
evidence about gender differences, and, as Kotthoff (2006: 3) succinctly maintains, ‘there
are more similarities than differences in the humor of women and men’. These observations
are in line with Kendall’s statement that ‘[s]ituations in which women and men consciously

199
Stephanie Schnurr & Nor Azikin Mohd Omar

choose language options to create femininity or masculinity are rare. In contrast, women
and men do consciously choose language to best fulfill their roles as managers’ (2004: 76).
Our examples below illustrate precisely this.
Moving on from these initial studies which often compared how men and women use
humour in a particular context, more recent studies focus on how the use of humour is
gendered and how humour may be used by interlocutors to put gender on the agenda and
to make often hidden gender stereotypes visible. For example, Schnurr and Holmes (2009)
and Holmes and Schnurr (2014) investigated how masculinity and femininity are enacted,
reflected, manifested, and sometimes challenged in participants’ use of humour in the work-
place. Rather than comparing the discursive behaviour of men and women, they explore
some of ‘the ways in which gender and humor intersect in workplace interaction’ (Holmes
and Schnurr 2014: 165) and how both men and women utilise humour to ‘do gender’,
thereby negotiating their professional and their gender identities.
But despite these recent trends, which reject treating men and women as homogenous
groups, gender stereotypes about women’s use of humour, and their abilities to lead, seem
to prevail in workplaces around the world (e.g. Clifton et al. 2019; Holmes 2006). This
chapter explores this stereotypical perception by further discussing the complex relation-
ship between gender, leadership, and humour. Drawing on authentic workplace interactions
collected in a range of different workplaces in New Zealand, the UK and Malaysia, we
analyse some of the ways in which humour is used by those in leadership positions, in order
to critically explore the role of gender. However, in contrast to much previous research,
our particular interest in this chapter is methodological, and we aim to illustrate some of
the benefits of using IS to inquire into the complexities of leadership, humour, and gender.

Interactional sociolinguistics as an approach to analyse leadership,

humour, and gender
IS is one of the major theoretical approaches to discourse analysis (Holmes 2014: 177).
It was pioneered by the anthropologist Gumperz. IS is a well-established framework that
is frequently used in the field of sociolinguistics for inquiries into actual language use in
specific situations and sociocultural contexts. In contrast to some other discourse analytical
approaches, such as conversation analysis, it pays particular attention to the ways in which
societal and interactive forces merge in specific communicative practices (Clifton et al.
2019; Gumperz 2003; Vine et al. 2008: 345).
The main aim of IS is to discover how social meaning is negotiated in interactions.
This is usually achieved through ethnographic observations and interviews (Gumperz 2003;
Holmes 2014: 179), as well as by an analysis of audio- or video-recorded interactions
(Bailey 2008). Based on the analysts’ knowledge of the specific community (and its norms)
in which an interaction occurs, interlocutors’ relationships, and the context, analysts make
informed presuppositions about meaning (Gumperz 2003; Holmes 2014; Vine et al. 2008).
These interpretive assessments build on local or context-specific background knowledge,
i.e. ‘taken-for-granted background assumptions that underlie the negotiation of shared inter-
pretations’ (Gumperz 1999: 454).
Central to IS are the concepts of conversational inferencing and contextualisation cues
(Gumperz 2003), which capture the processes through which interlocutors signal and inter-
pret meaning in social interactions (Bailey 2008). Conversational inference refers to the
interpretive procedure by which interlocutors assess what is communicatively intended

200
 Leadership and humour at work

at any point in an exchange, and on which they rely to plan and produce their responses
(Gumperz 2003). An example of this is the sequential positioning of turns (Gumperz 1982:
131) – such as what comes before and after a humorous instance, i.e. what triggers the
humour and how it is responded to, which are important for an understanding of what makes
a particular utterance humorous and what functions it performs at the specific point in the
interaction where it occurs. Contextualisation cues, on the other hand, are those linguistic
and paralinguistic features that contribute to the signalling of contextual presuppositions.
For example, in order to mark and understand a particular utterance as humorous, inter-
locutors draw on specific verbal and nonverbal cues, such as laughter, a smile, and a raised
eyebrow, as well as a change in pitch, tone of voice, and speed of delivery (c.f. Hay 2001;
Schnurr and Chan 2011).
As we demonstrate in more detail in the next section, in actual analysis, IS draws on
and combines both a fine-grained analysis of interaction and the analysts’ knowledge and
interpretation of the specific context in which the interaction occurs. When using IS to
analyse a particular excerpt of an interaction, researchers look at what people say but also
at how they say it. More specifically, they note not only who makes a humorous comment,
gives an instruction, or makes an apology, but they also describe the discursive processes
through which these acts are realised. For example, what pronouns are used, whether laugh-
ter occurs (and where), whether there are any pauses, hesitations, restarts, and – importantly
for humour – in what tone of voice a particular remark was uttered. This information about
the micro-level of an interaction is then combined with the researcher’s knowledge of the
macro-level (e.g. about the relationship among participants, how long they have worked
together, and the context in which an utterance occurs). Through this combination insights
are gained into how meaning is constructed and negotiated among interlocutors as their
interaction unfolds (Holmes et al. 2011: 21).
Due to its ability to combine the micro and the macro in meaningful ways, IS is an excel-
lent framework for an analysis of gender, leadership, and humour. It provides the analysts
with the discursive tools to identify and describe the specific discursive processes through
which humour is produced and responded to in an interaction, while at the same time allow-
ing the analysts to relate these micro-level observations to the macro-level concepts of
gender and leadership. It thus facilitates the operationalisation of these relatively abstract
concepts, and provides a frame in which they can be identified and traced throughout an
interaction, as we illustrate in the next section.

Leadership, humour, and gender in action

For our analysis, we have chosen four instances of humour to illustrate some of the com-
plexities of leadership and gender. More specifically, we look at how humour is used to dis-
play various behaviours associated with leadership, such as getting things done, criticising
subordinates and telling them what to do, giving feedback, reinforcing solidarity, and build-
ing rapport. The examples come from different interactional contexts – including small and
large meetings, an email, and a range of related WhatsApp messages – and different socio-
cultural contexts – including New Zealand, the UK, and Malaysia. This diversity is delib-
erate as we want to show that the use of humour, and its relationship with leadership and
gender, is not a local phenomenon that is tied to a particular sociocultural context, but rather
that it can be (and indeed has been) observed in many different contexts. In line with recent
developments in language and gender research (see Eckert and Podesva, this volume), the

201
Stephanie Schnurr & Nor Azikin Mohd Omar

starting point of our analysis is not the question of whether there are differences in the ways
in which men and women use humour to do leadership, but rather whether gender is relevant
to the ongoing interaction,2 and if so, in what ways.

Getting things done

Example 1 (from Schnurr 2009b)3
Context: A small meeting of three colleagues in an IT department at a New Zealand
organisation. All participants are IT managers on the same hierarchical level within the
organisation, but Noel chairs the meeting and is in charge of this particular project.

1. Noel: you’re clearly the most important person

2. Isabelle: oh definitely
3. Patrick: cool do I get veto rights
4. Noel: [voc] well yes but you get to do all the work as [laughs]: well:
5. Patrick: oh (great what a move up)

This is a good example to show how humour can assist people to get things done, which is
one of the activities associated with leadership (e.g. Schnurr and Mak 2011). The relevant
humour in this extract occurs in the form of two witty one-liners (lines 1 and 4) that are
uttered by Noel towards his colleague Patrick to remind him of the responsibilities that
come with his special role in a project that they are working on. Noel’s repeated use of the
second singular personal pronoun ‘you’ illustrates this allocation of responsibilities. There
are several contextualisation cues that mark this brief exchange as humorous, including
tone of voice and laughter. The first witty one-liner, in which Patrick is singled out as ‘most
important’ (line 1), is responded to with agreement from Isabelle and is continued with a
witty remark by Patrick himself (line 3), who plays along with it by tongue-in-cheek asking
for ‘veto rights’ as a reward for his important role in the project. This humorous comment, in
turn, is responded to by Noel, who teasingly reminds his colleague that his special position
also comes with certain obligations and responsibilities, and that Patrick is in fact expected
‘to do all the work as well’ (line 4). The utterance-final laughter produced by Noel is the
only laughter that occurs in this brief exchange, and it could be interpreted as an attempt
to mitigate his transactional message (i.e. reminding his colleague Patrick of his duties and
thereby eventually ensuring that things are getting done). Looking at the ways in which
interlocutors skilfully pick up and expand the humour, we find further support in the form
of conversational inferencing that they conjointly interpret this instance as humorous. The
humorous sequence comes to an end with Patrick’s perhaps slightly ironic remark (line 5) in
which he acknowledges his additional responsibility and makes fun of it, while at the same
time acknowledging that he has understood the underlying serious message.
Addressing the aim of this chapter and exploring whether gender is (or is not) explicitly
relevant in the use of humour by those in leadership positions, there is no marker (or con-
textualisation cue) in the above excerpt that indicates that gender plays a role here. Although
Noel’s use of teasing could perhaps be associated with masculinity, the bonding style in
which it is delivered (see also Schnurr 2009a) could in turn be associated with femininity.
This example thus illustrates that while leaders often use humour, in contrast to stereotypi-
cal perceptions and expectations, gender is not necessarily an issue. Rather, there is plenty
of evidence in our data from different workplaces in New Zealand, the UK, and Malaysia of
witty one-liners used by those in leadership positions – both men and women – to get things

202
 Leadership and humour at work

done by reminding people of their responsibilities or making sure they are aware of what
is expected of them (see Schnurr 2009b for more examples). Sometimes they do this by
drawing on speech elements traditionally associated with masculinity – such as Noel’s teas-
ing – and at other times they display behaviours indexed for femininity –such as the overall
bonding style in which the teasing is delivered. Thus, by using humour (in whatever form
or style) to convey these potentially critical and possibly even face-threatening messages,
while at the same time mitigating the illocutionary force of the utterance (that is, ‘soften the
blow’) and reinforcing solidarity (e.g. Schnurr 2009b) as Noel does in the example above,
the speakers primarily do leadership, and gender takes a backseat (cf. Kendall 2004).
We discuss one more example here to show how humour may be used to get things done
by criticising subordinates and telling them what to do by overruling the group’s previous
decision. In the next example, the person in a leadership position is a woman, and some of
the behaviours she displays are traditionally associated with masculinity.

Criticising subordinates and telling them what to do

Example 24,5,6
Context: This exchange occurred during a WhatsApp7 interaction among a group of
academics at an institute of Higher Education in Malaysia. All members are academic
colleagues and Zana is the head of department. At this point in the interaction Zana
(who was absent during the previous 1,645 posts) joins the discussion which took
place over approximately four hours, which is partly due to a delayed response time
between posts.

1646) 14/02/2016, 10:17 pm - Zana: ellooo!!

1647) 14/02/2016, 10:17 pm - Zana: bru bole lyn wasap
just got the chance to look at wasap
1648) 14/02/2016, 10:17 pm - Zana: 140 mesej!! mmg soheh grup ni penuh kontroversi
140 messages!! it is certain that this grup is con-
troversial
1649) 14/02/2016, 10:18 pm - Zana: kita x leh dgr dri stdnts jer
we cannot hear from the stdnts only
1650) 14/02/2016, 10:18 pm - Zana: find out properly
1651) 14/02/2016, 10:19 pm - Zana: sometimes, thngs likeths can be blown out of
proportion!!
1652) 14/02/2016, 10:19 pm - Zana: be rationale people!!
1653) 14/02/2016, 10:49 pm - Wani:
1654) 14/02/2016, 10:53 pm - Zana: klu mmg betul citer ni, bru kita gi ganyang GB
ramai2
if this story is true, then we will ‘beat up’ the
headmaster together
1655) 14/02/2016, 10:54 pm - Wani: Dr. Eusof will investigate…
1656) 14/02/2016, 10:55 pm - Zana: sherlock Holmes
1657) 14/02/2016, 11:20 pm - Eusof:
1658) 15/02/2016, 6:48 am - Sarah: Heheheh

203
Stephanie Schnurr & Nor Azikin Mohd Omar

Like the previous example, this excerpt further illustrates how humour may be skilfully used to
do power, while at the same time creating solidarity among team members. Like in the exam-
ple above, gender is not explicitly made relevant here, although Zana, the female leader of the
team, employs a range of discursive features associated with masculinity when doing humour.
With the rather informal greeting ‘ellooo!!’ (line 1646) Zana, the head of department and
one of the most senior members of the team, joins the discussion of her team on WhatsApp.
Her greeting resembles that between friends rather than colleagues (especially with its infor-
mal spelling (dropping of initial letter ‘h’ and lengthened vowels throughout) and the two
exclamation marks at the end). These relatively informal and collegial behaviours could be
ascribed to a feminine speech style. However, Zana then swiftly and without much ado moves
on to her transactional agenda, namely to criticise the team for an unnecessarily long discus-
sion (cf. her remarks about the number of messages followed by two utterance-final exclama-
tion marks in line and her use of the term ‘controversial’ (line 1648)), and the team’s one-sided
gathering of information (i.e. taking students’ complaints at face value without consulting the
respective school for their view (line 1649)). What is noteworthy about Zana’s interactional
style here is its relative directness and hence potential face-threat to her subordinates, which
are behaviours stereotypically associated with masculinity (e.g. Holmes 2006). This directness
of her messages is reflected in several contextualisation cues, including her frequent use of
exclamation marks (lines 1651 and lines 1652), the absence of a personal pronoun (line 1650),
and her directive (line 1652). However, at the same time, it is somewhat mitigated by her
use of the inclusive pronoun ‘we’ (line 1650), her non-standard spellings (lines 1650, 1651),
and the use of two laughing emojis at the end of her seven consecutive messages (line 1652).
The emojis could mitigate the potential face-threat of her previous utterances but also act as
contextualisation cues which reframe the tone of the interaction as friendly rather than strictly
professional, and which open up a humorous context (Darics 2017) and perhaps even invite
the subsequent development of conjoint humour among the other team members.
In terms of conversational inferencing, it is important to note that Wani’s thumbs-up emojis
in line 1653 may signal not only that she has understood and taken on board Zana’s criticism,
but they also continue the humour, as does Zana’s tongue-in-cheek reply about beating up the
headmaster ‘if this story is true’. Using fantasy humour here to describe the team’s potential
(clearly non-serious) actions, she expresses solidarity with her subordinates and bonds with
them by making fun of the absent headmaster of the school that is causing the trouble (line
1654). In particular, her reference to a joint future action (‘together’) performs this bonding
function, which – just like criticising subordinates and telling people what to do – has also
been associated with leadership (e.g. Schnurr 2009b; Wodak et al. 2011). The humorous tone
of her posts is then picked up and continued by other members of the team. First, Wani teases
Eusof by referring to him as ‘Dr Eusof’ and portraying him as a detective who will ‘investi-
gate’ matters (line 1655). This fantasy humour is continued by Zana who mentions the famous
fictional detective ‘sherlock Holmes’ (line 1656) thereby implicitly making a link between
Eusof and this fictional character. Although he is the butt of the humour, Eusof responds posi-
tively (with an emoji that is laughing so hard it is crying (line 1657)). This in turn is responded
to by Sarah, who produces some orthographic laughter (line 1658). The humour comes to an
end and people discuss more serious matters (not shown here).
Using humour in this short exchange, then, enables Zana to exercise her power and
authority as the head of department and to be the one who is able to overturn a decision
made by the team in a way that reinforces solidarity and contributes to bonding among
interlocutors. She thereby gains her subordinates’ compliance rather than threatening their
faces and generating resistance. While these semiotic resources are claimed to have positive

204
 Leadership and humour at work

implications on the mood of a discussion, this example has shown that discursive strategies
such as emoji, informal spellings (‘grup’ for ‘group’), and abbreviation (‘thngs likeths’) are
also useful to attenuate the illocutionary force of potentially threatening speech acts like
directives and disagreement; at the same time they act as important contextualisation cues
and facilitate conversational inferencing for the researcher and participants to interpret this
exchange as humorous. Moreover, these strategies also allow Zana to exercise her author-
ity and establish a sense of solidarity with her subordinates. Thus, although IS analyses of
digital communication focus on different features than IS analyses of spoken interactions,
some of these semiotic resources are actually reminiscent of the characteristics of spoken
interaction. For example, emojis perform similar functions to facial expressions and tone
of voice, while non-standard spellings could be interpreted as non-standard pronunciation.
Looking at Example 2 in relation to Example 1, it is noteworthy that although they
occurred in different sociocultural contexts, different professional industries, and were
produced by different individuals in leadership positions – one male, one female – there
are some remarkable similarities with regard to the use of humour. Both leaders draw on
humour to combine transactional and relational behaviours, and both display behaviours
indexed for masculinity (such as directness, and the exercise of power) as well as behav-
iours typically associated with femininity (such as bonding, creating solidarity, and mitigat-
ing negatively affective speech acts).

Giving feedback
Example 3 (from Schnurr 2009b)
Context: At an IT company in New Zealand. Gerry is the mentor to several graduates
who have recently joined the company. At this meeting Gerry is giving feedback to a
group of graduates who are working on a particular task with which they are experienc-
ing some difficulties.

1. Gerry: [teasing tone of voice] : there’s a really easy answer to this

2. and I’m not gonna tell you what it is :
3. Hank: oh you [laughs]
4. Magnus: [laughs]

The relatively short humour in this example takes the form of teasing, which is a particularly
ambiguous type of humour (e.g. Alberts 1992; Schnurr 2009a). The teasing tone of voice in
which this short remark is uttered sets it up as playful and non-serious. This contextualisa-
tion cue is picked up by the audience as Hank’s and Magnus’ supportive responses indicate.
Hank plays along with Gerry’s teasing by pretending to be upset (line 3). He continues the
playful, mocking tone and produces some utterance-final laughter, which is then mirrored
by Magnus in the next utterance (line 4).
Using humour in this context enables Gerry to encourage his mentees and provide some (at
this stage largely moral rather than technical) support, while at the same time also reinforcing
his own status and authority as the expert. Moreover, the laughter that his humour triggers gen-
erates a welcome opportunity to release some tension and take a break – even if only momen-
tarily – from the difficult task (Glenn 2003). Like in the examples above, however, there is no
indication – in the form of a contextualisation cue or conversational inferencing – that points
to the explicit relevance of gender in this exchange. Rather, the perhaps more-biting style of
the teasing (compared to Noel’s more bonding style (cf. Example 1)) could be explained by

205
Stephanie Schnurr & Nor Azikin Mohd Omar

reference to the different workplace cultures in which these exchanges took place (see also
Schnurr 2009a). Comparing Examples 1, 2, and 3 in particular further illustrates that there is
considerable variation and diversity in which leaders – male or female – use humour to do lead-
ership. But there is clearly no preferred style by either men or women. The next example fur-
ther illustrates this by showcasing an example of humour that is constructed by largely drawing
on linguistic features indexed for femininity. This instance of humour occurred in an email.

Reinforcing solidarity and building rapport

Example 48

Context: Email between two colleagues at a higher education institution in the UK.

Hi Rebecca

Well done on getting this sorted. Can I just check that you have confirmed this allocation
with all available parties, i.e. LMM! (Sure you have….)
For my part I have phoned Alma and reallocated you [name of teaching module]. And
apologies for this SNAFU. As these things come in threes […] you can expect open [i.e.
‘one’] more gaffe from me before beginning next year!
All I can say is – it’s nothing personal!!!!
Cheers
Bentley

The humour in this comment arises from the use of the abbreviation ‘SNAFU’, which means
‘situation normal all fucked up’, and the subsequent self-denigrating and equally humorous
comments with which Bentley apologises for his mistakes. The light-hearted tone of this apol-
ogy is further intensified by his excessive use of exclamation marks and his emphasis at the end
of the email, that ‘it’s nothing personal!!!!’. These contextualisation cues mark this exchange
as humorous. This humour comes after a more serious and transactional beginning, in which
he asks Rebecca to confirm that she has performed some previously discussed work (‘Can I
just …’), and his subsequent expectation that she has indeed done so (‘Sure you have’). He
follows this up by outlining his own actions (‘For my part …’), and so the humour in the fol-
lowing apology seems more like an add-on, and not strictly speaking necessary to convey all
these transactional messages. Nevertheless, the humour plays several important functions here:
not only does it make Bentley’s apology more acceptable, but it also creates solidarity and
builds rapport between interlocutors. All these behaviours are important elements of leader-
ship as they combine transactional and relational behaviours to ensure that things are getting
done, while maintaining a positive and collegial relationship (e.g. Schnurr 2009b). These witty
remarks, which are delivered in a style that has often been described as feminine (especially
the self-denigrating comment and the apologetic tone) thus assist Bentley in his leadership per-
formance. But like in the examples discussed above, there are no contextualisation cues or evi-
dence of conversational inferencing that would mark his use of humour as explicitly gendered.

Discussion and conclusion

Our examples of authentic spoken and written interactions collected in a range of work-
places in different sociocultural contexts provide a snapshot of the wide diversity of the

206
 Leadership and humour at work

ways in which those in leadership positions use different types of humour to achieve their
various leadership aims. Our observations largely confirm findings of previous research
which shows that leaders regularly use humour to achieve their various objectives and to
perform a range of transactional, as well as relational, behaviours that are associated with
leadership, including getting things done, criticising subordinates and telling them what
to do, as well as creating solidarity and maintaining positive relationships (e.g. Holmes et
al. 2011; Schnurr 2009b). Our examples above have shown that male and female leaders
in different sociocultural contexts regularly use humour to assist them in their leadership
performance. They all sometimes use humour in ways that are traditionally associated with
masculinity or femininity, and we have ample evidence of women displaying behaviours
that are typically indexed for masculinity, and men engaging in behaviours often associated
with femininity (see also Mohd Omar 2019; Schnurr 2008, 2009b). These observations
about the use of humour by male and female leaders thus provide compelling evidence to
challenge prevailing stereotypes about gender and leadership, and also about gender and
humour. In contrast to these negative preconceptions which tend to portray women as not
using or not appreciating humour as much as men, we have described several instances in
which leaders – regardless of gender – skilfully use humour when performing a range of
activities associated with leadership.
In addition to adding evidence to earlier studies challenging gendered stereotypes (see
also Holmes 2006; Ladegaard 2012; Mullany 2007), one of the aims of this chapter was
to illustrate some of the benefits of using IS to analyse leadership, humour, and gender at
work. As our analyses above have illustrated, one of the advantages of such an approach is
that it equips the analysts with specific tools that facilitate the identification of humorous
instances and their interpretation by the immediate audience via contextualisation cues (e.g.
in the form of audible or orthographic laughter, emojis, a teasing tone of voice, excessive
use of punctuation devices, and non-standard spelling). Moreover, through conversational
inferencing and focusing on interlocutors’ reactions to the humour (i.e. what triggered the
humour and how it was responded to (e.g. whether it was successfully continued, ignored or
rejected) and by whom (e.g. the target of the humour, by several or only one other interlocu-
tor)), analysts can gain insights into the interactional dynamics, and can better understand
how interlocutors negotiate meaning throughout an exchange.
These observations about interlocutors’ behaviours on the micro-level of an interaction
(such as their use of and responses to humour) can then, in turn, be linked to macro-level
categories (such as leadership and gender), which enables researchers to identify and describe
some of the specific (discursive) strategies through which interlocutors do leadership. IS
thereby provides the tools and practices to generate empirical evidence in support of claims
made by recent constructivist conceptualisations of leadership and gender as dynamic perfor-
mances that are enacted and negotiated on the micro-level of interaction (e.g. Baxter 2014).
So, returning to the role of gender in the context of leadership and humour, in the examples
that we discussed in this chapter, there was very little evidence of gender being explicitly rel-
evant. The leaders in our data all used humour to achieve similar aims, regardless of gender,
position, and sociocultural context. However, this does not mean that gender is not important.
In fact, as we have discussed elsewhere, gender often remains relatively hidden but may come
to the fore in the form of humorous (and sometimes sarcastic) remarks to send up some of the
stereotypes that exist in the work domain (such as men not being able to multi-task (Schnurr
and Holmes 2009); women decorating their office to make it more ‘girly’ (Schnurr 2008);
and women being constantly worried about their looks (Holmes and Schnurr 2014)). Thus,
while gender – with a few exceptions – does not seem to be constructed as an issue per se on

207
Stephanie Schnurr & Nor Azikin Mohd Omar

the micro-level of actual everyday workplace interactions, negative stereotypes about gender,
leadership, and humour seem to prevail in many workplaces.
However, this persistence of gender stereotypes in the ways in which women leaders
are presented in the media or are being talked about in their workplaces is rather surprising
given the evidence provided in this and many other studies that challenge and reject such
derogatory claims (e.g. Holmes 2006; Ladegaard 2012; Mullany 2007). It is thus clearly
about time these old and inadequate stereotypes about gender, leadership, and humour are
being abolished and replaced by less discriminatory views, and we hope that this study
makes a small contribution to these efforts.

Future directions
As we have shown in our analyses, IS is a very useful approach to capture the complexi-
ties and dynamics of leadership, gender, and humour – both in spoken interactions as well
as written digital texts. However, we believe that more could be made of this versatility of
IS, and that researchers should in particular apply IS more systematically in their studies of
leadership, gender, and other social phenomena in digital and multi-modal contexts, such as
websites, Instagram, Snapchat etc. Applied to these complex and rapidly changing contexts,
in which spoken, written, and visual texts are intricately interwoven with each other to create
meaning, IS – perhaps when combined with another approach – has the potential to provide
further evidence of the complex ways in which gender is constructed, enacted, and negotiated.

Transcription conventions
[laughter] square brackets provide information about paralinguis-
tic features
(…..) round brackets indicate the transcriber’s best guess at
what was said
[teasing tone of voice] : …… : words between colons are uttered in a teasing tone of
voice

Notes
1 The Organisation for Economic Co-operation and Development (OECD). 2016. Background report
conference on improving women’s access to leadership 2016. https://www.oecd.org/daf/ca/OECD-
Women-Leadership-2016-Report.pdf
2 In exploring the relevance of gender we look for evidence where interlocutors orient to gender in
explicit or implicit ways, for example, by using specific pronouns when referring to allegedly gender-
neutral terms (like ‘manager’ – ‘he’, ‘nurse’ – ‘she’) or by mobilising overtly gendered identity cat-
egories (such as ‘bloke’, ‘girl’, ‘woman/female manager’ (as opposed to ‘manager’)). For a useful
discussion of the theory behind ‘making gender relevant’ see Stokoe and Smithson (2001).
3 We have italicised the relevant humour in all examples to facilitate an understanding and for ease of
reading. The transcription conventions are included at the end of the chapter.
4 Spelling, grammar, and capitalisation have been left as in the original, and we have tried to reflect
participants’ practices as much as possible in our translations. Only names have been anonymised.
5 This example also appears in Mohd Omar (2019).
6 In this excerpt participants use two different codes: Malay and English. We have translated the Malay
into English which appears below the original in the transcript.
7 WhatsApp is an instant messaging application that operates across most smartphones, tablets, and
desktops via a connection to the internet. It is claimed to be the most well-run instant messaging

208
 Leadership and humour at work

service available; it handles more instant messages in a day than the entire global short-messaging
system industry (Sutling et al. 2015). Its speediness in transmitting messages has enormously facili-
tated the way people communicate with each other (ibid.). For the purpose of this study, we have
extracted bite-sized chunks of text from a longer decision-making interaction of a committee team in
Malaysia.
8 Spelling, grammar, and capitalisation were left as in the original. Only names have been anonymised.

Further reading
Baxter, J. (2014) ‘Who wants to be the leader? The linguistic construction of emerging leadership in
differently gendered teams’, International Journal of Business Communication, 52(4), pp. 427–451.
This paper provides a good discussion of some of the advantages of IS for analysing and
understanding leadership in action.
Holmes, J. (2006) Gendered talk at work: constructing gender identity through workplace discourse.
Malden, MA: Wiley- Blackwell.
This research monograph contains two very good chapters on gender and leadership, and gender
and humour.
Mullany, L. (2004) ‘Gender, politeness and institutional power roles: humour as a tactic to gain
compliance in workplace business meetings’, Multilingua, 23, pp. 13–37.
This paper explores gender, leadership, and humour from the angle of politeness.
Rogerson-Revell, P. (2011) ‘Chairing international business meetings. Investigating humour and
leadership style in the workplace’, in Angouri, J. and Marra, M. (eds) Constructing identities at
work. Basingstoke: Palgrave.
This paper is a very interesting study of leadership and humour. Although its main focus is not
gender, it does briefly touch upon it.
Schnurr, S. (2009b) Leadership discourse at work: interactions of humour, gender and workplace
culture. Basingstoke: Palgrave Macmillan.
This research monograph specifically deals with leadership and humour using an IS approach. One
of the chapters is dedicated to the topic of gender.

Related topics
Interactional sociolinguistics: foundations, developments and applications to language, gender and
sexuality; identity construction in gendered workplaces; leadership language of Middle Eastern
women; interactional sociolinguistics in language and sexuality research: benefits and challenges;
language, gender, and the discursive production of women as leaders.

References
Alberts, J. K. (1992) ‘Teasing and sexual harassment: double-bind communication in the workplace’,
in Perry, L. Turner, L., and Sterk, H. (eds) Constructing and reconstructing gender: the links
among communication, language, and gender. Albany, NY: SUNY Press, pp. 185–196.
Avolio, B. J., Jane, M. H., and John, J. S. (1999) ‘A funny thing happened on the way to the bottom
line: humor as a moderator of leadership style effects’, Academy of Management Journal, 42(2),
pp. 219–227.
Bailey, B. (2008) ‘Interactional sociolinguistics’, in Donsbach, W. (ed) International encyclopaedia of
communication. New York: Blackwell, pp. 341–369.
Baxter, J. (2010) The language of female leadership. Basingstoke: Palgrave Macmillan.
Baxter, J. (2012) ‘Women of the corporation: a sociolinguistic perspective of senior women’s
leadership language in the UK’, Journal of Sociolinguistics, 16(1), pp. 81–107.
Baxter, J. (2014) ‘Who wants to be the leader? The linguistic construction of emerging leadership
in differently gendered teams’, International Journal of Business Communication, 52(4), pp.
427–451.
209
Stephanie Schnurr & Nor Azikin Mohd Omar

Baxter, J. (2017) ‘Resolving a gender and language problem in women’s leadership: consultancy
research in workplace discourse’, Discourse & Communication, 11(2), pp. 141–159.
Billing, Y. and Alvesson, M. (2000) ‘Questioning the notion of feminine leadership: a critical perspective
on the gender labelling of leadership’, Gender, Work & Organization, 7(3), pp. 144–157.
Chesterman, C., Ross-Smith, A., and Peters, M. (2005) ‘The gendered impact on organisations of a
critical mass of women in senior management, Policy & Society, 24(4), pp. 69–91.
Clifton, J., Schnurr, S., and Van de Mieroop, D. (2019) The language of leadership narratives: a social
practice perspective. Abingdon: Routledge.
Collinson, D. (2006) ‘Rethinking followership: a post-structuralist analysis of follower identities’, The
Leadership Quarterly, 17(2), pp. 179–189.
Darics, E. (2017) ‘E- leadership or “how to be boss in instant messaging?” the role of nonverbal
communication’, International Journal of Business Communication, 57(1), pp. 1–27.
Eagly, A. H. and Carli, L. L. (2003) ‘The female leadership advantage: an evaluation of the evidence’,
The Leadership Quarterly, 14(6), pp. 807–834.
Fletcher, J. K. (2001) Disappearing acts: gender, power, and relational practice at work. Cambridge,
MA: MIT Press.
Fletcher, J. K. (2004) ‘The paradox of postheroic leadership: an essay on gender, power, and
transformational change’, The Leadership Quarterly, 15(5), pp. 647–661.
Glenn, P. (2003) Laughter in interaction. Cambridge: Cambridge University Press.
Gumperz, J. (1982) Discourse strategies. Cambridge: Cambridge University Press.
Gumperz, J. (1999) ‘On interactional sociolinguistic method’, in Srikant, S. and Roberts, C. (eds.)
Talk, work and institutional order. discourse in medical, mediation and management settings.
Berlin: de Gruyter, pp. 453–472.
Gumperz, J. (2003) ‘Response essay’, in Eerdmans, S., Prevignano, C., and Thibalut, P. (eds) Language
and interaction: discussions with John J. Gumperz. Amsterdam: John Benjamins, pp. 105–126.
Hay, J. (2000) ‘Functions of humor in the conversations of men and women’, Journal of Pragmatics,
32(6), pp. 709–742.
Hay, J. (2001) ‘The pragmatics of humor support’, Humor – International Journal of Humor Research,
14(1), pp. 55–82.
Holmes, J. (2006) Gendered talk at work: constructing gender identity through workplace discourse.
Malden, MA: Wiley- Blackwell.
Holmes, J. (2014) ‘Doing discourse analysis in sociolinguistics’, in Holmes, J. and Hazen, K. (eds)
Research methods in sociolinguistics: a practical guide. West Sussex: John Wiley, pp. 177–193.
Holmes, J. and Marra, M. (2006) ‘Humor and leadership style’, Humor - International Journal of
Humor Research, 19(2), pp. 119–138.
Holmes, J., Marra, M., and Vine, B. (2011) Leadership, discourse, and ethnicity. Oxford: Oxford
University Press.
Holmes, J. and Schnurr, S. (2014) ‘Funny, feminine, and flirtatious: humor and gendered discourse
norms at work’, in Chiaro, D. and Baccolini, R. (eds) Gender and Humour. Interdisciplinary and
International Perspectives. New York: Routledge, pp. 165–181.
Holmes, J., Schnurr, S., and Marra, M. (2007) ‘Leadership and communication: discursive evidence of
a workplace culture change’, Discourse & Communication, 1(4), pp. 433–451.
Hoption, C., Barling, J., and Turner, N. (2013) ‘”It’s not you, it’s me”: transformational leadership
and self-deprecating humour’, Leadership & Organization Development Journal, 34(1), pp. 4–19.
Jenkins, L. and W. (1985) ‘Locating the language of gender experience’, Women & Performance: A
Journal of Feminist Theory, 2(1), pp. 5–20.
Kanter, R. and M. (1977) ‘Some effects of proportions on group life: skewed sex ratios and responses
to token women’, American Journal of Sociology, 82(5), pp. 965–990.
Kendall, S. (2004) ‘Framing authority: gender, face and mitigation at a radio network’, Discourse &
Society, 15(1), pp. 55–79.
Koenig, A., et al. (2011) ‘Are leader stereotypes masculine? A meta-analysis of three research
paradigms’, Psychological Bulletin, 137(4), pp. 616–642.

210
 Leadership and humour at work

Kotthoff, H. (2006) ‘Gender and humor: the state of the art’, Journal of Pragmatics, 38(1), pp. 4–25.
Ladegaard, H. (2012) ‘Rudeness as a discursive strategy in leadership discourse: culture, power and
gender in a Hong Kong workplace’, Journal of Pragmatics, 44(12), pp. 1661–1679.
Martin Rojo, L. and Gomez Esteban, C. (2003) ‘Discourse at work: when women take on the
role of manager’, in Weiss, G. and Wodak, R. (eds) Critical discourse analysis. Theory and
interdisciplinarity. New York: Palgrave Macmillan, pp. 241–271.
Mohd Omar, N. A. (2019) Leadership discourse and decision-making in a Malaysian workplace.
Unpublished PhD thesis. University of Warwick, UK.
Mullany, L. (2004) ‘Gender, politeness and institutional power roles: humour as a tactic to gain
compliance in workplace business meetings’, Multilingua - Journal of Cross-Cultural &
Interlanguage Communication, 23(1–2), pp. 13–37.
Mullany, L. (2007) Gendered discourse in the professional workplace. Basingstoke: Palgrave
Macmillan.
Rogerson-Revell, P. (2011) ‘Chairing international business meetings. Investigating humour and
leadership style in the workplace’, in Angouri, J. and Marra, M. (eds) Constructing identities at
work. Basingstoke: Palgrave, pp. 61–84.
Rutherford, S. (2001) ‘Organizational cultures, women managers and exclusion’, Women in
Management Review, 16(8), pp. 371–382.
Schnurr, S. (2008) ‘Surviving in a man’s world with a sense of humour. An analysis of women leaders’
use of humour at work’, Leadership, Special Issue on Communication and Leadership, 4(3), pp.
299–319.
Schnurr, S. (2009a) ‘Constructing leader identities through teasing at work’, Journal of Pragmatics,
41(6), pp. 1125–1138.
Schnurr, S. (2009b) Leadership discourse at work: interactions of humour, gender and workplace
culture. Basingstoke: Palgrave Macmillan.
Schnurr, S. and Chan, A. (2009) ‘Leadership discourse and politeness at work. A cross cultural case
study of New Zealand and Hong Kong’, Journal of Politeness Research, 5(2), pp. 131–157.
Schnurr, S. and Chan, A. (2011) ‘Exploring another side of co-leadership: negotiating professional
identities through face-work in disagreements’, Language in Society, 40(2), pp. 187–210.
Schnurr, S. and Chan, A. (2017) ‘Leadership and self-denigrating humour: an oxymoron?’, in
Meredith, M. and Warren, P. (eds) Festschrift for Janet Holmes. Wellington: Victoria University
Press, pp. 234–254.
Schnurr, S. and Holmes, J. (2009) ‘Using humor to do masculinity at work’, in Norrick, N. R. and
Chiaro, D. (eds) Humor in interaction. Amsterdam: John Benjamins, pp. 101–123.
Schnurr, S. and Mak, B. (2011) ‘Leadership in Hong Kong. Is gender really not an issue?’, Gender
and Language, 5(2), pp. 337–364.
Sinclair, A. (1998) Doing leadership differently. Victoria: Melbourne University Press.
Singh, V. and Vinnicombe, S. (2002) The 2002 female FTSE report: women directors moving forward.
Centre for Developing Women Business Leaders at the Cranfield School of Management.
Stokoe, E. and Smithson, J. (2001) ‘Making gender relevant: conversation analysis and gender
categories in interaction’, Discourse & Society, 12(2), pp. 243–269.
Stuling, K. et al. (2015) ‘Social media: how it ensures effective communication among project
team’, 14th International Conference on Applied Computer and Applied Computational Science
(ACACOS’15), Recent Advance in Computer Science, pp. 222–227.
Tannen, D. (1994) Talking from 9 to 5. Women and men at work: language, sex and power. Reading:
Virago.
The Organisation for Economic Co-operation and Development (OECD). 2016. Background report
conference on improving women’s access to leadership 2016. https://www.oecd.org/daf/ca/OECD-
Women-Leadership-2016-Report.pdf.
Vine, B. et al. (2008) ‘Exploring co-leadership talk through interactional sociolinguistics’, Leadership,
4(3), pp. 339–360.
Wodak, R., Winston, K., and Clarke, I. (2011) ‘“Getting people on board”’: discursive leadership for
consensus in teams’, Discourse & Society, 22(5), pp. 592 –644.

211
 14
More than builders in pink shirts
Identity construction in gendered workplaces
Jo Angouri, Meredith Marra, and Shelley Dawson

Introduction
As a theoretical stance, interactional sociolinguistics (IS) dominates research conducted
under the umbrella often labeled as ‘workplace discourse analysis’. Equally, research on
gender within linguistics is heavily influenced by an IS approach. This chapter addresses the
intersection of these two areas by outlining research and methods for investigating gender
and gender identity in the workplace context. In doing so we highlight the affordances of the
different datasets that are used by IS scholars, using examples from a range of workplaces
by way of illustration. Recently there has been increased discussion of the relationship
between agency and structure, as well as the impact of ideologies, especially the pre-existent
hegemonic orders that constrain the way we use language at work (namely unquestioned
power imbalances and hierarchies in society, including the ‘gender order’). After working
through the advancements in the field, we conclude with a discussion about how we can
operationalise these recent ideas in our data analysis.
In our account of the development of the field we describe how analysis of gender has
traversed from gender identity construction at work to workplaces as gendered settings,
always challenging a static conceptualisation of gender in favour of a dynamic and fluid
approach. We cover different industries and different professions, demonstrating through
naturally occurring examples the ways in which people negotiate their identities, not just
with their colleagues but also in the context of the shared workplace practices within which
they are operating. We argue that all workplaces are gendered, and that gender is indexed,
negotiated, and challenged in various ways according to context. We turn now to a brief
description of the field and outline how IS, as the core theoretical approach used by work-
place discourse analysis researchers, is adopted.

Workplace research and IS

IS is uniquely suited to capture discursive complexities. It provides the tools to access
how people use language to make communicative meaning ‘in the now’, connecting fine-
grained micro-practices to wider sociocultural contexts and social discourses (e.g. Eckert

212
 Identity construction in the workplace

and McConnell-Ginet 2013). This has clear implications for investigations of gender, where
fixed understandings of gender roles, behaviours, and identities are extremely slow to dis-
lodge from the collective consciousness. At the micro-level, IS allows for access to gender
performativity (Butler 1990), that is to the dynamic ways in which we negotiate both our
understandings and presentation of gender identities. At the same time, connections with
macro-contexts encourage researchers to look beyond the ‘doing’ in order to ‘see’ wider
structural forces at play. In this sense, an IS approach reveals the hidden aspects of gender
identity construction and contributes to destabilising notions of a ‘core’ gender identity (e.g.
Hall 2000).
This translation process takes place through the eyes of the researcher. IS’s close rela-
tionship with ethnography and anthropology has allowed for a systematic consideration of
the role of the researcher and the importance of reflexivity. The framework and associated
methodology provide a voice to the researcher to systematically discuss their own role in
the research process and in the framing of the findings (Angouri 2018) and a way to connect
with widespread ideologies visible in the lives of the interactants.
Gender has been said to be omnirelevant (e.g. Cameron 2009; Holmes 1995), which
makes it difficult to ‘pin down’ analytically. This is where IS is so valuable. Early investiga-
tion of gender in an IS tradition is influenced by the work of Tannen, a student herself of the
originator of IS, Gumperz. Tannen (1994) addressed the workplace as a setting in Talking
from 9 to 5, which like her most famous book (You Just Don’t Understand, 1990) was aimed
at a more general audience. In both cases she argued that just because there are recognisably
gendered ways of behaving, this does not mean that any behaviour is inherently inferior or
superior. Her work showed the complexities of distinguishing between gender/ed and other
categories, paving the way for an emphasis on the symbols, or indexicalities, of meaning
as well as the importance of intersectionality in our understanding of identity construction.
Understanding the enactment of gender identity involves a microanalytic approach and
typically makes use of naturally occurring data. In line with the wider discursive trend in
the social sciences, workplace discourse analysts argued for the importance of data which
showed what we actually do in our communication at work rather than what we think we do.
This premise motivates the research of workplace discourse scholars and means that since
the earliest days of the field of workplace discourse analysis (in the early 1980s/90s) data
regularly comprises recordings of authentic interactions between colleagues in workplace
settings. Self-reported data widely used by other social science disciplines, and notably in
business schools, has been criticised for failing to capture the nuance of identity work in
situated moments.
The data-collection techniques used by all three authors follows the philosophy and
flexible procedures developed by Holmes and Stubbe for the Wellington Language in the
Workplace project (see description in Holmes and Stubbe 2003a). After identifying suc-
cessful communicators (as judged by colleagues and industry), the process involves making
contact with chosen organisations, developing mutually agreed goals, and then handing over
the recording process to workplace volunteers who act as collaborators in recording an array
of everyday practices and contexts ranging from core events, such as meetings, to other
types of encounters, such as interviews.
To support the analysis of this material, and in line with the IS approach, we augment
these recordings with ethnographic observations, debriefs, member-checking, and, increas-
ingly, with interviews. Together this means we have an etic, analyst perspective as well
as access to an emic, insider perspective, providing information that pulls together wider
societal context to interpret individual behaviour and interactional negotiation (see also

213
Jo Angouri et al.

Marra and Lazzaro-Salazar 2018). This approach is widely used across workplace discourse
research, and although it is regularly adapted to match the logistical needs of the worksite
or industry (Vine and Marra 2017), it has served the purpose of gathering high-quality data
that participants are happy to provide.
Next we turn to the ways in which gender has been analysed using these datasets. We
begin with the core focus, gender identity, which is as relevant to current research as it was
in the earliest days of workplace discourse research some two to three decades ago.

Constructing gender identity at work

Identity is one of the main interests of the IS analyst, and gender is one of the identities
that has received the most attention within workplace discourse analysis. Holmes (2006),
for instance, provides a comprehensive, and intentionally non exhaustive, list of normative
gendered styles at work. Traits stereotypically associated with femininity include: provid-
ing supportive feedback, being conciliatory, contributing less in public, and being person/
process oriented. Amongst the features identified as normatively associated with masculin-
ity we find competitiveness, aggressiveness, and confrontational style, alongside being task/
outcome oriented. These and other gendered behaviours are not equated with biological sex;
they do however reflect, and perpetuate, the wider sociopolitical and ideological context
(macro-level) when they are claimed, projected, and resisted discursively in the situated
moment of interaction (micro-level).
This is how and where an IS approach becomes relevant for bridging the macro- and
micro-levels and unpacking gendered norms in the workplace, investigating the indexicality
of meaning which considers the correlation between linguistic/discursive features and social
identities. Angouri (2015: 45) argues that ‘it is through the utterances’ indexical properties
that people create relationships between the identity categories [i.e. social identities] and
position self and other’. This applies to the male/female dichotomy as well.
An area in which gender has been particularly highlighted in workplace talk is leader-
ship. Building on the ‘think leader, think male’ stereotype (Marra, Schnurr, and Holmes
2006), researchers have been eager to show that femininity and leadership can be co-pre-
sent identities within talk. Recent research demonstrates that leadership performances are
dependent on the characteristics of the local context. For instance, Mesinioti et al. (2020)
explore leadership enactment in medical emergencies, investigating how senior clinicians
‘do being’ the leader discursively. Their findings highlight the consistency of leadership
claims by senior clinicians. They show that leadership in that context is signalled, projected,
and resisted discursively, through directives and questions that are used in ways consistent
with the role/responsibility being claimed. Although the authors are immediately concerned
with the use of material space in the emergency room, their findings show the difficulty in
disentangling gender from other ways of enacting self and other in any professional setting
as well the importance of systematic and detailed analysis from multiple angles in a given
‘community of practice’.

Gendered communities of practice

By the 2000s, there was recognition that gender identity occurred within gendered contexts
– enacting femininity in a feminine workplace was different to constructing femininity in
a masculine context. An early example of this argument came from Holmes and Stubbe

214
 Identity construction in the workplace

(2003b) who contrasted an all-female and ‘feminine’ workplace in the form of a particular
government department with a factory team described as masculine in which the gender
identification of the team members included both women and men but where identity con-
struction was shaped by the more masculine norms of the group.
Their arguments made use of the Community of Practice (CofP) framework which
derives from the work of Lave and Wenger (1991) and Wenger (1998). Lave and Wenger
define a CofP as ‘a set of relations among persons, activity, and world, over time and in
relation with other tangential and overlapping communities of practice’ (1991: 98). In their
definition, CofPs are characterised by mutual engagement (typically enacted through regu-
lar interaction), joint negotiated enterprise (involving the complex relationships of recipro-
cal accountability), and a shared repertoire of negotiable resources accumulated over time.
The CofP framework has been widely applied in workplace studies, as many of the ideas it
employs are applicable to the workplace e.g.: the interaction among CofPs; the progression
from legitimate peripheral participation towards core membership; and the ways newcom-
ers learn from old-timers through apprenticeship schemes (Angouri 2018; King 2019).
The focus of sociolinguistic work has concentrated on the discourse practices that CofPs
develop over time, as CofPs ‘provide a way to study language use locally and to link the
micro-context to the broader socio-cultural environment’ (Angouri 2016: 314). CofPs
are also gendered, operating as sites of gendered activity (Paechter 2003); being part of a
CofP more often than not comes with gendered expectations. Membership is discursively
claimed and negotiated through adherence to gender norms that meet these expectations.
The construction of an identity in line with hegemonic gendered discourses for claiming
membership in the workplace is shown in all the excerpts below which cover a range of very
different kinds of workplaces.
Example 1 focuses on identity construction by a builder’s apprentice, Max. In the extract
we see Max enacting masculinity within a team in which normatively masculine ways of
behaving form part of their shared repertoire.

Excerpt 1
Context: Max, the apprentice, talking to another tradesman on site1

1. Max: [clears throat] did you get did you watch any of the league
2. or the highlights or anything
3. John: oh I’ve seen saw a few games eh
4. Max: fuck man there are some pretty decent hits man
5. John: yeah

This team of builders spend long hours on site working alongside each other. As John
and Max cooperate on their manual task, their talk drifts to non-work-related talk.
Interestingly the topic of this talk is something that we might associate with masculinity,
sport, and more specifically rugby league which has intersectional indexicalities of both
masculinity and class or region. As well as content (in the form of this topic), elements of
masculine identity can also be seen in the discursive practices which are used. Expletive
use, swearing, is a strategy overtly associated with masculine styles (Daly et al. 2004). In
New Zealand where this extract was recorded, the pragmatic particle ‘eh’ (line 3) is also
a feature which is arguably (although not exclusively) linked to masculinity (Vine and
Marsden 2016).

215
Jo Angouri et al.

This example also highlights the role of industry on gendered context. The men are draw-
ing on highly stereotypical indexes of masculinity which are acceptable and appropriate in
their negotiated repertoires. Arguably the embedded nature of their community within the
construction sector, also stereotypically a masculinised industry, supports these behaviours.
There is increasing research on sex-segregated and gendered professions (see McDowell
2021). While the building site might represent a masculinised industry within which Max is
able to balance his gender and professional identities easily, what does it mean to construct
a masculine identity in a stereotypically feminine context?

Gendered occupations
We now turn to eldercare as an example of an industry which is typically gendered as femi-
nine. We choose this extract because not only do you see more evidence of the balancing
of a range of identities by carer Afato, but you also see his identity being co-constructed by
resident Ida for whom he is caring. Gendered norms at both a macro-societal and micro-
interactional level are relevant to any workplace and are mobilised through interaction.

Excerpt 2
Context: Afato is checking on Ida as part of his duties as a carer in a residential aged-care
facility. While this medicalised discussion is taking place, Ida refers back to a lamp that
Afato had accidently knocked over on the previous day. Frank is another carer in the facility
interacting with Ida. Like Afato, Frank identifies as Samoan, but he is 20 years his senior.
Before this extract, Afato and Ida have been laughing about his involvement in the recording
process for our research.

1. Afato: so are we going back into the wheelchair or the lazy boy love
2. Ida: (over there)
3. Afato: alright […]
4. they want an incident form for that thing +
5. the the lights //going off\
6. Ida: /okay\\ of course
7. Afato: yes
8. Ida: I didn’t even see what you do did
9. but you were doing it rather vigorously //obvious\ly
10. Afato: /[laughs]\\
11. Ida: did you see what he’s done Frank
12. Frank: yeah [laughs]
13. Ida: Frank
14. Afato: not me?
15. Ida: crashed the light off one of them
16. Afato: (and this)
17. Ida: you just can’t let him loose
18. Afato: eh?
19. Ida: can’t let some people loose can you
20. Afato: yeah
21. Ida: they run amok
22. Afato: it is

216
 Identity construction in the workplace

23. Ida: they run amok I love those expressions + you know what amok means
24. [Ida explains amok to Afato who doesn’t know the expression]
25. Ida: amok //Afato\ ran amok
26. Frank: /(no)\\
27. Afato: what’s that
28. Ida: he’s got out of control and started smashing lights
29. Afato: it’s not like I did it deliberately
30. Ida: [laughs]
31. Afato: it is
32. Ida: but it happened
33. Afato: yeah ++ okay anything else Ida you need anything else
34. Ida: no thank you Afato

Excerpt 2 provides a useful example of how gendered norms are oriented to within a ‘femin-
ised’ workplace. Throughout, there is evidence of Afato’s skilful negotiation of overlapping
identities, as well as co-construction in action through Ida’s positioning of Afato.
Immediately of note is Afaoto’s use of ‘love’ (line 1) as a term of endearment towards
Ida. Analysis has shown that this term is commonly used by eldercare workers (Marsden
and Holmes 2014), and is indexical to a professional identity in feminised care contexts.
Further evidence of this attention to professional identity is seen in the collaborative, or
even conspiratorial, discussion between Afato and Ida about the paperwork he has to do
(lines 3–7) with a demanding and commonly understood ‘they’ (line 4) constructed as in
opposition to Afato and Ida. Ida’s teasing of Afato (lines 11–32) is revealing in that as well
as being playful and relational, it draws on entrenched notions of masculinity, in this case,
aggression. Positioning Afato as ‘out of control’ and ‘smashing lights’ (line 28), we see
how gender identity construction is a two-way process depending on moment-by-moment
linguistic negotiation (micro-level) within the sense-making frameworks of gender ideolo-
gies (macro-level). The interaction is, of course, a humorous one, with the impetus deriving
from the fact that Afato has not in any way acted aggressively (in line with stereotypes of
masculinity), something that all three participants are well aware of. The fluidity of these
relationships is also evident in the teaching moment between Ida and Afato (lines 23–25). In
negotiating gender within this context, professional and relational concerns are paramount.
This unsettles the traditional assumption that maleness equates to masculine norms irrespec-
tive of context.

Normative femininities and masculinities

The examples above demonstrate that there is more than one way to do femininity and mas-
culinity and that the narrow association with biological gender is problematic. We therefore
prefer the terms ‘masculinities’ and ‘femininities’ which align with our focus on the dyna-
mism and multiplicity of identities. Forms of femininity and masculinity also change over
time and space, and certain forms of gender are traditionally accorded more societal status
(i.e. the ‘gender order’, Connell 1987). These forms have been normalised to the extent that
deviation from expected gender behaviours and identities is still seen as marked in many
contexts today. The example involving Afato above shows that non alignment with gen-
dered stereotypes is noticeable enough to create the incongruity required for humour. Yet,
attention to normativity does not capture the full story of gender in the workplace. Attention

217
Jo Angouri et al.

to emancipatory practices which push back against restrictive understandings of femininity

and masculinity afford access to the real-life sensitivities of gender identity. Understandings
of gender as dynamic and fluid are increasing in many parts of the world challenging the
status quo. People draw on resources to agentively, and sometimes fleetingly, perform femi-
ninities or masculinities, which may align with or contest stereotypical gender expectations,
and they may do both within the context of one interaction. Workplace research offers many
useful examples of this interactional intricacy.
To capture some of this nuance, Excerpt 3 includes a number of complexities. Like
Excerpt 1 it comes from data collected on the building site. In this case, however, the norms
of the community of practice actively challenge the stereotype of builders as hypermascu-
line, while in some ways also arguably reinforcing the stereotype at the same time.

Excerpt 3
Context: Rodney (the foreman) teases Darren (the client and his friend) about the pink shirt
he is wearing.

1. Rodney: might take my t shirt off er at least my shirt off + Darren

2. Darren: mhm
3. Rodney: got that pink shirt on again //bloody me\
4. Darren: /I think\\ this is the same one actually
5. Rodney: no (it isn’t) nah
6. Darren: I think the pink I think the sh- original pink shirt
7. is out there
8. Rodney: really? still going? +++ I’ve got a very pink one at home
[Darren goes to get the shirt] […]
9. Darren: that’s the original shirt
10. Rodney: oh très pink très très pink mon ami [laughs]

Our title for the chapter comes from this example because of the attention to dynamism
and detail which is increasingly central to the analysis undertaken by workplace discourse
analysts. In this example the participants are arguably ‘playing’ with the stereotype when
indexing their identities.
The colour pink has some societal connotations with femininity and with homosexual-
ity, which the team seem to be referencing in their humour. The humour relies on a level
of ambiguity: for these men wearing a pink t-shirt is perfectly acceptable, but they know
that the stereotypical builder might think the colour choice is unacceptable. Using poorly
accented French (a language associated with otherness, pretentiousness, and perhaps even
femininities), the humour is made more explicit. By referencing the stereotypes of an ‘unac-
ceptable’ masculinity, however, they are also validating it as having currency in their inter-
action. It is important to note that in our interviews with this group they actively talked about
themselves as standing in contrast to the stereotype of builders – they were all university-
educated and came to the industry after other professional careers. Their data includes them
listening to classic rock and classical music on site, and they prioritised their children and
children’s events over hypermasculine stereotypes of sports and drinking which featured in
the site represented by Excerpt 1.
Gendered points of reference are widely available in any societal context. Although gen-
der researchers have been showing the complexity and coexistence of masculinities and

218
 Identity construction in the workplace

femininities, stereotypical binaries persist in society – consider the ‘pink aisles’ in any toy
shop. Excerpt 3 can be read as simultaneously perpetuating and questioning the gender
order. Disestablishing gender dominance, however, is bigger and stronger than any one
individual or work group, as the following sections will show.

Impact of structure/ideologies on agency and identity construction

Adding a critical ethnographic lens to IS allows for further subtleties. Capturing gender in
interaction requires constant attention to complexity and to the multitude of ways gender can
be instantiated or challenged. We take as a starting point that IS is inherently critical and
political (cf. Heller 2013), reminding us to keep a constant eye on power and inequality in our
data. ‘Acceptable’ versions of masculinity and femininity are often felt acutely by those who
experience marginalisation (Dawson 2019; Holmes 2018), creating ideological constraints.
To enact this stance, we return to the methodology for data collection and analysis
described at the beginning of the chapter. To move more firmly into the critical space which
may have been at the heart of IS, but has often been somewhat implicit or even neglected
in practice, researchers increasingly engage with questions of power and inequality, and
how the sociolinguistic practices we observe have consequences for the people involved.
Simultaneously, there is a recognition of the importance of reflexivity; regular and ongo-
ing reflexivity involves analysing ourselves as researchers, our various identities, and the
assumptions we may bring to our data. Explorations of workplaces are in-depth and situated
and relationships are prioritised in the ongoing commitment to researching ‘with’ partici-
pants (Roberts 2003). We draw on our knowledge of context to strengthen interpretations
and make regular field notes which describe and question the practices we witness and are
part of. This critical ethnographic layer is a natural extension of IS, and allows us to move
between emic and etic perspectives (see Marra and Lazzaro-Salazar 2018).
In workplace contexts, analyses of gender can be ethnographically enhanced at different
levels. At a meso level, developing understandings of the particular workplace is essential.
Attention to the organisational structure, goals, expectations, and norms provides valuable con-
textualisation, and offers insights into the wider sociocultural discourses which inform them. In
the abstract, it remains imperative to engage with context at the ideological level. The gender
order and heteronormativity (as ingrained and normalised ideologies) do not limit themselves
to a particular time or place. The shape they take may be locally variable, but these ideologies
have colonised much of the world imperialistically (cf. Kress 1989: 7). In other words, in nego-
tiations of gender, the dominant ideological framing stays similar irrespective of place. What
differs, however, is the shape and strength of the particular norms and discourses that give voice
to these ideologies. As we have begun to demonstrate in the previous examples, IS provides the
tools to unveil the connections from micro-linguistic features to the level of abstract ideology.
Ethnographic approaches therefore operationalise aims of calling upon wider context
while also enhancing understandings of the subtleties of actual ‘on the ground’ interactions
in which gender is salient. As with dichotomous views of gender, we see that the so-called
analytical macro/micro divide is actually a false dichotomy. This leads us back to Gumperz’
insight that you cannot have one without the other when investigating social life (e.g. Heller
2013), and this point gains even more traction when gender is the focus of inquiry.
IS has not traditionally been used for exploring issues of power and politics. Recent
work, however, revisits its affordances and makes a case for the relevance of the framework
for a critical study of professional interaction (e.g. Angouri 2018). As some workplaces
are highly hierarchical (see, for instance, healthcare contexts), the IS approach becomes

219
Jo Angouri et al.

particularly relevant, allowing us to take into account both the institutional roles and power
asymmetries at work, as well as the here-and-now interactions. In order to meet this goal, a
combination of datasets such as meetings and interview interactions, and a systematic anal-
ysis of the latter, allows us to unpack different layers of context and draw robust inferences.
As noted earlier, as a field workplace discourse analysis has prioritised analysis of record-
ings of naturally occurring talk. It is now time that interviews are repositioned as a method
and dataset. Interviews constitute co-constructions between the researcher and participants.
As such they should be considered to be real-life interactions just as any other and can use-
fully be analysed to reflect on the negotiation of ideals between key stakeholders in any
project, the researcher and the participants. They also provide a window for the researcher
to problematise inferencing about the local context which is typically subject to less scrutiny
when compared to our detailed analysis of encounters in other professional events, such as
meetings or other kinds of interactions.
We show this by analysing an interview excerpt from a higher education (HE) context
and a project on ‘Gender and INequality in HE’ (GaIN). As we turn to a more critical use of
IS, it is important to recognise and acknowledge gender inequality in analysis. This societal
inequality, as for example reflected in the gender pay gap, applies both horizontally as well
as vertically across professional settings. Horizontal and vertical gender asymmetries make
local negotiation of imbalance always subject to negotiating dominant hegemonies and par-
allel structures that affect individuals across domains of human activity (consider the rela-
tionship between ‘home’ and ‘work’). GaIN brought these considerations to the forefront,
aiming to provide an understanding of the dominant discourses and ideologies that circulate
in one HE setting and aimed to the way for establishing the landscape of gender equality and
diversity in HE. The project involved review of institutional documents and interviews with
academic staff members from various levels and disciplines.

Excerpt 4
Context: Mary (the interviewer) and Lucy, the interviewee, discuss how Lucy navigates
male-dominated spaces. Lucy is a junior staff member working in a quantitative field, which
entails, according to her, a lack of women in academic events.

1. Mary you are in a very quantitative field aren’t you?

2. Lucy yes I am + and I guess that it stands out because sometimes
3. because sometimes I go to conferences where + I can be + like +
4. the only woman in the room
5. Mary right
6. Lucy so it’s (1.0) it does stand out sometimes
7. Mary was that + I suppose (1.0) how do you navigate or
8. how do you + kind of + work around that I suppose if you are
9. does it have an effect on you [drawls]: or: ?
10. Lucy I don’t think it does er + no (1.0) no obviously no +
11. yeah I think I’m +I’m quite lucky being like + a childless
12. well not lucky + [chuckles] but I am a kind of childless woman
13. [drawls]: so: in these circumstances I can + behave like a man
14. [chuckles] if you know what I mean
15. as mothers have to go home and look after their children so +
16. I guess that gets me accepted more in these situations yeah

220
 Identity construction in the workplace

17. Mary right + what do you mean behave like a man

18. Lucy well I guess + these men like +
19. if you go to events and stuff they can just stay and go for a drink
20. or something + and there are a few women who quite often
21. they would say + oh I have to get back to have my kids
22. or if we are in a conference away sometimes they will say +
23. oh I am just gonna go and skype my kids
24. and tell them a bedtime story and things +
25. but I don’t have to do these things
26. so you kind of (1.0) remain in the group
27. Mary mhm + mhm + do you find your character has changed at all?
28. kind of like + because of that + social interaction with uh?
29. Lucy um + I did + one guy I do research with + he often says +
30. oh + I’m one of the boys and that’s why it works to do research
31. with them + because they can all talk to me
32. because I’m one of the boys
33. Mary right
34. Lucy so I don’t know + I don’t know what that kind of means
35. I don’t know if that’s a compliment or an insult [laughs]
36. Mary but how does he mean it + you’re one of the boys?
37. Lucy I think he’s + he’s like+ you know-
38. they can all open the talk in the group
39. and if I go in the group it’s not like er oh + there’s a woman there +
40. it’s like I’m part of the group

Excerpt 4 is a succinct illustration not only of how gendered norms are (re)produced at
work, but also of the value of interview data under an IS approach. Throughout the excerpt,
hegemonic gender norms are recycled and gender is foregrounded both in the interviewer’s
prompts and the interviewee’s uptake; Mary makes an association between ‘quantitative’
and ‘male dominated’ which is consistent in the dataset. Lucy ratifies this and in the unfold-
ing narrative, femininity is associated with childcare responsibilities (lines 11, 15, 20–24),
while drinking behaviour (line 19) and lack of childcare (throughout the excerpt) are index-
ing biological maleness and masculinity.
The prefacing of ‘quite lucky to be childless’ (line 11) in the navigating of male-dom-
inated spaces frames the narrative in hegemonic terms where femininities are associated
with caring responsibilities, the same ideals we saw in Excerpt 2 earlier. Lucy’s immediate
well-prefaced self-correction followed by a short pause and a chuckle are indicators that
she perceives a need for self-correction. The repair however results, again, in the reproduc-
tion of gender norms and a performance of masculinity through a Goffmanian ‘passing’
(Goffman 1959): ‘I can behave like a man’ (line 13). This presupposes success indexed by
being one of ‘them’. The conceptualisation of femininity as closely linked to motherhood
becomes a thread throughout and is evident in the sequential order of interaction; Lucy turns
from ‘women’ participating at conferences (lines 3–4) to ‘mothers’ who ‘have to’ go home
and look after their children (line 15). The use of a modal verb of obligation (‘have to’)
twice, in lines 15 and 21 further highlights women’s obligation to take care of their children.
Conforming to the dominant gendered norms, Lucy overemphasises the gendered stereotype
of women as carers later in the excerpt, too, by choosing to draw on a prototypical example
of maternal devotion and affection, that of reading bedtime stories to the children (line 24).

221
Jo Angouri et al.

The negotiated performance between Mary and Lucy provides a rich insight into hegem-
onic discourses of femininity in the academic context and is very relevant to the discussion
in this chapter; Lucy mobilises gendered stereotypes, associates them to a female identity
in the academic sphere, and also reinforces them throughout the episode. Note that the
objective of the project and this particular encounter is to shed light on the asymmetries
and by extension to empower women; dominant ideologies however are apparent through-
out. Whether and to what extent this challenges the status quo or reinforces it is a complex
matter (consider the use of stereotypes for humour in Excerpt 3) and one our field needs to
critically address. Despite the attempt to empower those marginalised or oppressed, our own
agendas often reproduce dominant binaries.
Turning back to the excerpt, the equation of masculinity with gender male is quite promi-
nent here. See, for instance, line 13, ‘I can behave like a man’, and line 32, ‘I’m one of the
boys’; the implication here is that Lucy herself identifies going out for a drink or conducting
research with males as prototypical masculine behaviours; to be part of those requires sup-
pressing her feminine identity. Although academia is a very different sector to those we have
seen so far, and one in which we may expect different processes of in/equality, acceptance
in work groups is subject to gendered negotiation of norms; in/exclusion in the academic
context is also gendered. Lucy’s ‘masculine’ construction (being childless; being able to go
out for drinks; not standing out as a woman in a male group) enables her to claim access to
the dominant professional group for and with the sympathetic interviewer who also aligns
with the macro-narratives performed here. Claiming inclusion is constructed as important
to Lucy, as the thread of belonging to the group is recurrent throughout the excerpt: e.g.,
‘that gets me accepted’ (line 16); ‘remain in the group’ (line 26); ‘I’m one of the boys’ (line
32); ‘I’m part of the group’ (line 40). And arguably all professionals negotiate ‘belonging’
in their daily routines at work. Although CofPs have their own ways of negotiated gendered
‘normativities’ this does not suggest that societal macro-narratives are not visible. In all
our examples, societal understandings are made relevant, from the swearing, league-loving
builders, to the aggressive male caregiver, the linking of the colour pink to (‘unaccepta-
ble’) homosexuality and pretentiousness, and to childcare being a disadvantage to being a
‘proper’ member of the team. Putting together the discussion of the different professional
encounters, we see the ways in which excerpts can feed into one another and enable the
researcher to follow their own positioning as well as the ideological threads that are mobi-
lised by the participants in different contexts.
In sum, gender, commonly seen as the very fabric of social life, operates across several
levels; we cannot move forward in understandings of gender as a social process without
exploring that very social process in action. An ethnographically and critically informed
IS allows for a unified examination of gender, bringing together micro-linguistic practices,
norms, identities, sociocultural discourses, and ideologies. It acknowledges gender as a
social structure and the appearance of reality this entails, while also understanding gender
as being in a state of perpetual evolution. This blurring of ontological understandings is
the first step in moving towards social change. What remains to be fully developed are the
concrete ways in which these insights can inform workplaces committed to principles of
inclusion and social equality.

Future directions
Workplace scholars with a focus on gender have laid a robust platform from which to build.
Embracing a social justice perspective and harnessing political salience are important and

222
 Identity construction in the workplace

exciting future directions. In order to do this, we need to include more diverse workplaces
in our investigations. In times where the very shape of ‘work’ is rapidly changing (think
the ‘gig economy’) as part of increasing mobility and new forms of technology, we need
to widen the scope of workplaces we investigate. Workplaces are increasingly intricate
and complicated, not least because of the various modes of interacting which make work
simultaneously virtual and face-to-face, synchronous and asynchronous, dependent on
new AI technologies and on the (constrained) choices and actions of the human work-
force. Even more importantly, we need more gender-diverse voices in our research (both
in terms of researchers and participants). Emphasising the plurality of experience will
trouble the predominant cis hetero framing of gender research in the workplace and move
us further into a richer and more revealing space. Relatedly, the area of ideologies offers
further possibilities for analytical expansion. A dual focus on hegemonic ideologies (and
the discourses that give them voice) and budding, counter-ideologies will allow work-
place scholars to engage with both existing structures and fledgling new forms which con-
test the status quo. Only through this shared focus will we be able to access the subtleties
needed to move beyond boxes and dichotomies into more expansive and inclusive terrain.
And this is exactly where we need to be.

Transcription conventions
[laughs]: Paralinguistic features and editorial information in square brackets;
colons indicate beginning and end
+ Pause of up to one second
... //......\ Simultaneous speech
/.......\\ ...
() Unclear utterance
? Questioning intonation

Note
1 Names of workplace participants and workplaces are pseudonyms.

Further reading
Angouri, J. and Marra, M. (eds) (2011) Constructing identities at work. Houndmills: Palgrave.
This edited volume considers how interactants do identity work and how identity is indexed
in workplace discourse. It brings together a range of professional and institutional contexts, from
corporate workplaces, to courtrooms, classrooms, and academia, and introduces intersectionality.
Baxter, J. (2014) Double-voicing at work: power, gender and linguistic expertise. Basingstoke:
Palgrave Macmillan.
This book explores how women leaders use double-voicing more than men as a means of gaining
acceptance and approval in the workplace. The readers will benefit from Baxter’s attempt to combine
a nuanced and rich analysis of empirical data with a critical agenda.
Holmes, J. (2006) Gendered talk at work. Oxford: Blackwell.
Drawing on a rich body of workplace interactional data, Holmes provides a solid illustration of
the ways in which both women and men draw on gendered discourse resources to enact a range of
workplace roles.

223
Jo Angouri et al.

Tannen, D. (1995) Talking from 9 to 5: women and men in the workplace. New York: Avon.
Tannen probes how gender roles shape the ways men and women communicate in the workplace,
and the ways in which these differences can lead to misunderstandings. It is one of the first systematic
attempts to investigate the workplace dynamics using an IS approach.
Vine, Β. (ed) (2017) The Routledge handbook of language in the workplace. New York: Routledge.
Covering a variety of theoretical and methodological approaches and exploring different types of
workplace settings, this handbook provides a comprehensive overview of research on language in the
workplace.

Related topics
Gender and sexuality normativities; interactional sociolinguistics in language and sexuality research:
benefits and challenges; the accomplishment of gender in interaction: ethnomethodological and
conversation analytic approaches to gender; interactional sociolinguistics: foundations, developments
and applications to language, gender and sexuality; leadership and humour at work: using interactional
sociolinguistics to explore the role of gender.

References
Angouri, J. (2015) ‘Studying identity’, in Hua, Z. (ed) Research methods in intercultural
communication. Oxford: Blackwell, pp. 37–52.
Angouri, J. (2016) ‘Online communities and communities of practice’, in Georgakopoulou, A. and
Spilioti, T. (eds) The Routledge handbook of language and digital communication. Routledge
Handbooks in Applied Linguistics. Routledge, pp. 323–338.
Angouri, J. (2018) Culture, discourse, and the workplace. London: Routledge.
Butler, J. (1990) Gender trouble. New York: Routledge.
Cameron, D. (2009) ‘Theoretical issues for the study of gender and spoken interaction’, in Pichler, P.
and Eppler, E. (eds) Gender and spoken interaction. Basingstoke: Palgrave Macmillan, pp. 1–17.
Connell, R. (1987) Gender and power: society, the person and sexual politics. Sydney: Allen & Unwin.
Daly, N., Holmes, J., Newton, J., and Stubbe, M. (2004) ‘Expletives as solidarity signals in FTAs on
the factory floor’, Journal of Pragmatics, 36(5), pp. 945–964.
Dawson, S. (2019) ‘΄Bitch I’m back, by popular demand΄: agency and structure in a study abroad
setting’ , Gender and Language, 13(4), pp. 499–468.
Eckert, P. and McConnell-Ginet, S. (2013) Language and gender, 2nd edn. Cambridge: Cambridge
University Press.
Goffman, E. (1959) The presentation of self in everyday life. New York: Anchor Doubleday.
Hall, S. (2000) ‘Who needs identity?’, in Evans, J., Du Gay, P., and Redman, P. (eds) Identity: a
reader. London: SAGE, pp. 15–30.
Heller, M. (2013) ‘Gumperz and social justice’, Journal of Linguistic Anthropology, 23(3), pp.
192–198.
Holmes, J. (1995) Women, men and politeness. London: Longman.
Holmes, J. (2006) Gendered talk at work. Oxford: Blackwell.
Holmes, J. (2018) ‘Negotiating the culture order in New Zealand workplaces’, Language in Society,
47(1), pp. 33–56.
Holmes, J. and Stubbe, M. (2003a) Power and politeness in the workplace: a sociolinguistic analysis
of talk at work. Harlow: Longman.
Holmes, J. and Stubbe, M. (2003b) ‘“Feminine” workplaces: stereotypes and reality’, in Holmes, J.
and Meyerhoff, M. (eds) The handbook of language and gender. Oxford: Blackwell, pp. 573–599.
King, B. W. (2019) Communities of practice in language research: a critical introduction. London:
Routledge.
Kress, G. (1989) Linguistic processes in sociocultural practice. Oxford: Oxford University Press.

224
 Identity construction in the workplace

Lave, J. and Wenger, E. (1991) Situated learning: legitimate peripheral participation. Cambridge:
Cambridge University Press.
Marra, M. and Lazzaro Salazar, M. (2018) ‘Ethnographic methods in pragmatics’, in Jucker, A. H.,
Schneider, K. P., and Bublitz, W. (eds) Methods in pragmatics. Berlin: De Gruyter Mouton, pp.
343–366.
Marsden, S. and Holmes, J. (2014) ‘Talking to the elderly in New Zealand residential care settings’,
Journal of Pragmatics, 64, pp. 17–34.
McDowell, J. (2021) De-gendering gendered occupations: analysing communicative practices in the
workplace. Abingdon: Routledge.
Mesinioti, P., Angouri, J., O’Brien, S., Bristowe, K., and Siassakos, D. (2020) ‘“Get me the airway
there”: negotiating leadership in obstetric emergencies’, Discourse and Communication, 14(2), pp.
150–174.
Paechter, C. (2003) ‘Masculinities and femininities as communities of practice’, Women’s Studies
International Forum, 26(1), pp. 69–77.
Roberts, C. (2003) ‘Applied linguistics applied’, in Sarangi, S. and Van Leeuwen, T. (eds) Applied
linguistics and communities of practice. London: Continuum, pp. 132–149.
Tannen, D. (1990) You just don’t understand. New York: Ballentine.
Tannen, D. (1994) Talking from 9 to 5: women and men at work. New York: William Morrow.
Vine, B. and Marra, M. (2017) ‘The Wellington language in the workplace project: creating stability
through flexibility’, in Marra, M. and Warren, P. (eds) Linguist at work: festschrift for Janet
Holmes. Wellington: Victoria University Press, pp. 181–201.
Vine, B. and Marsden, S. (2016) ‘Eh at work: the indexicality of a New Zealand English pragmatic
marker’, Intercultural Pragmatics, 13(3), pp. 383–405.
Wenger, E. (1998) Communities of practice: learning, meaning and identity. Cambridge: Cambridge
University Press.

225
 15
Interactional sociolinguistics in
language and sexuality research
Benefits and challenges

Corinne A. Seals

Introduction
One of the core tenets of interactional sociolinguistics (IS) as a type of discourse analysis
research is that ‘[L]anguage contextualises and is contextualised, such that language does
not just function “in” context, language also forms and provides context’ (Schiffrin 1994:
134). Taking an IS perspective, then, it is through language that sexuality becomes meaning-
ful, and our experiences with sexuality further contribute to the development of language
and how we talk about it. Therefore,

meaning is located in the interaction between the act (text) and context. So gender and
sexuality are not simply brought into being by speaking or doing, but via a process of
interaction between the speech act and the grounded sociocultural and situational con-
text in which it is produced and received.
(Sauntson and Kyratzis 2007: 3)

This chapter focuses on IS methodology as applied to language and sexuality research.

First, the foundations of IS research, along with key concepts within this method of dis-
course analysis, are presented. This is then followed by a discussion of the development of
language and sexuality research, which necessarily begins with its historical background
in language and sexuality studies. The ongoing applicability of IS to language and sexual-
ity research is also discussed, followed by an empirical case study illustration of a lesbian
comedian performing stand-up comedy in New York City. After the case study, there is a
further critical examination of the benefits of using IS, as well as insights that can be gained
from also adding the poststructuralist concept of intertextuality, providing readers with a
better knowledge of IS and how it works in practice.

Foundations of interactional sociolinguistics

The overall goal of an IS approach to discourse analysis is to examine how people interpret
discourse in interaction, including how people do this similarly and differently from each

226
 Benefits and challenges of IS

other. Furthermore, ‘IS assumes that interpretive assessments always build on locally or
context specific background knowledge that takes the form of presuppositions that shift
in the course of an encounter’ (Gumperz 1999: 458). In interpreting discursive interaction
through an IS approach, sequential turns at talk are examined for literal meaning, as well as
inferred meaning (e.g. through the use of presuppositions). Additionally, contextual infor-
mation is drawn upon by the researcher to uncover how a particular interaction is situated
within micro- and macro- community, cultural, and social norms. These social norms are
used as the contextual framework through which interaction is analysed. After repeated pat-
terns of interaction are found within the discursive data, instances of these are transcribed as
evidence of the types of key interactions observed. When analysing the interactional data,
research using IS encompasses a number of key concepts to enact this type of discourse
analysis, including ‘contextualisation cues’, ‘frame’, ‘key’, ‘face’, ‘footing’, and ‘polite-
ness’. While these terms are covered across the chapters in this section, the current chapter
focuses on contextualisation cues, framing, and key.
‘Contextualisation cues’ were famously defined by Gumperz (1982: 131) as ‘any fea-
ture of linguistic form that contributes to the signalling of contextual presuppositions’. This
can include intonation, gestures, facial expressions, prosody, formality of language (mor-
phologically, lexically, and/or syntactically), and language choice, among other features.
Contextualisation cues are therefore key in helping the interlocutor(s) interpret the inter-
action at hand, including the discourse. Furthermore, contextualisation cues in discourse
function indexically (Ochs 1992). That is, through discourse and/or paralinguistic cues, con-
textualisation cues point to or call to mind certain concepts with which the hearers are famil-
iar. Often, in order to interpret the indexical meaning, hearers rely upon schemata – ‘frames
and cultural scripts in which social expectations are viewed as the classifying device of
experience’ (Kuzio 2014: 2), including stereotypes and wider social discourses1 (Gee 1996).
In order to interpret the contextualisation cues put forth by a speaker, the interlocutor pre-
supposes certain background information about that speaker. For example, if an (assumed)
woman lowers her voice while talking, and if this is interpreted by the interlocutor to mean
that the person is ashamed, then the interlocutor has made several presuppositions. First,
the interlocutor has presupposed that a conventional woman in this society would have a
reason to be ashamed regarding the content of what was being shared. Second, there is a
presupposition that a conventional woman would display shame in this society through the
lowering of her voice when talking. Third, the interlocutor is presupposing that the speaker
identifies as a conventional woman in this society. Fourth, the interlocutor is presupposing
that the speaker shares the same indexical meaning for lowering her voice in this society.
However, any one of these presuppositions could be wrong, which would then lead to a mis-
interpretation of the contextualisation cues, and therefore misinterpretation of the discursive
interaction. In fact, investigating miscommunication is one of the foundational strengths of
IS research (cf. Gumperz 1982).
In addition to Gumperz’ contribution to IS of the key concept of contextualisation cues,
Goffman (1974, 1983) also played a central role in the development of ideas that became
adopted into IS. From Goffman, we get the concepts of ‘interaction order’, ‘frame’, and
‘key’, among others. In the concept of interaction order, Goffman was interested in explor-
ing how people conduct themselves interactionally in face-to-face situations (while IS does
not specify this mode of interaction). In particular, he noted that within an interaction, a
person’s main goal is to be heard and understood, such that a person is speaking in order
to achieve this goal and to provide evidence for their position (Goffman 1974). In order
to understand the interaction order, Goffman explains that we must pay attention to the

227
Corinne A. Seals

context, particularly settings. We must do so in order to understand the expectations of the

interaction (frame), the tone of the interaction (key), the ways that people try to maintain
their presented self and to avoid being challenged (face), and how shifts in interactional
events and/or participants can alter who is included or excluded (footing).
Indeed, Goffman’s concepts, as listed above, work very well with Gumperz’ concept
of contextualisation cues to explain what happens in a number of modern-day situations,
including that presented in the case study below: a stand-up performance by a lesbian come-
dian in New York City. For the empirical example presented later in the chapter, ‘frame’
and ‘key’ are focused on in particular, alongside ‘contextualisation cues’ (similar analytical
approaches can also be found in Bing and Heller’s (2003) study of lesbian humour, Pichler’s
(2017) study of couples’ talk, and Warner-Garcia’s (2015) study of young women’s talk of
spirituality and sexuality). As previously mentioned, ‘frame’ has to do with the context of
an interaction, and is meant to be used for interpreting the expected events to take place
within it. More specifically, frames are associated with the interactional ‘rules’ that enable
or constrain specific social encounters (e.g. attending class, visiting a relative, presenting at
a conference) and accompanying actions, including discursive moves. Additionally, ‘key’
adds to ‘frame’ by highlighting the tone that is given to an interaction, whether for example
it is a serious, sarcastic, or humorous encounter. Furthermore, contextualisation cues (usu-
ally including intonation) help the interlocutor know how cognitively and emotionally to
interpret the discourse and interaction, including how to interpret the frame. When a ‘key
change’ is successful, the interlocutor is able to follow a shift in tone and respond accord-
ingly (e.g. from humorous to serious).

The value of interactional sociolinguistics for language and sexuality

research
Scholars who utilise the IS approach to discourse analysis view the relationship between
language and society as reciprocal: ‘Language, culture, and society are grounded in interac-
tion: they stand in a reflexive relationship with the self, the other, and the self-other rela-
tionship, and it is out of these mutually constitutive relationships that discourse is created’
(Schiffrin 1994: 134). This perspective of language and society contributing to each other
is also a key component of research into language, gender, and sexuality. With regard to
gender identity and performance, Wagner (2010: 36) explains that gender is considered by
many sociolinguists to be ‘a matter of identity formation which continually recurs through
actions such as speaking … As a social construct rather than a biological fact, gender is
renegotiated each time men and women interact in different contexts’. Wagner (2010) also
connects language and gender research with language and sexuality research by exploring
studies that have approached studying lesbian discourse through woman-to-woman talk in
an effort to broaden our understanding of the vast range of gender identities (cf. Coates and
Cameron 1989; Remlinger 1999; Spender 1985).
In order to better understand language and sexuality research, it is important to first
establish how it developed in the field. In order to do so, we need to look to queer theory,
which has been, and often still is, interwoven with our understandings of language and
sexuality (see this volume for the history of language and gender research). Additionally,
the discourse analysis studies presented below make use of IS, though, importantly, it has
not always been explicitly identified as such. I have noticed that explicit reference to the IS
method is more common recently, and even now it is not always found. Instead, references to
‘discourse analysis’ without further specification (e.g. CDA, pragmatics, etc.) have usually

228
 Benefits and challenges of IS

stood instead of an explicit mention of IS. Thus, while IS has been a very popular method
of discourse analysis since its inception, it has more often than not simply been referred to
in studies as ‘discourse analysis’ (e.g. Lopez 2014; Oute and Huniche 2017; Wagner 2010)
(notable exceptions include Angouri 2018; Angouri, Marra, and Holmes 2017; and Holmes
and Marra 2010, among others).

The rise of queer theory and language and sexuality research

With the rise of third-wave feminism (focused on individualism and expanding who is seen
as a ‘woman’) (see Part VI on poststructuralism, this volume) with its interest in diver-
sity and socially constructed identities, came the emergence of queer theory. This was con-
nected with discourse research and linguistics through Livia and Hall’s (1997) seminal work
Queerly Phrased: Language, Gender, and Sexuality. Similar to feminist research in calling
for a disruption of hegemonic discourses (i.e. a dominant societal ideology about the way
things work), queer theory ‘represents various approaches that are driven by a critical focus
on heteronormativity’ (Kendall and Tannen 2015: 649). Similarly, relating to third-wave
feminism in particular, queer theory established the idea of sexuality and sexual identity as a
continuum. This was made popular in particular by two key texts – Butler’s Gender Trouble
(1990) and Sedgwick’s Epistemology of the Closet (1990).
Thus, together, third-wave feminism and queer theory led to the rise of queer linguistics
(cf. Leap 1995; Livia and Hall 1997; for discussions on the future of queer linguistics, see
Sauntson this volume), as well as language and sexuality research (see Cameron and Kulick
2003 for a thorough overview of language and sexuality research’s origins). Furthermore,
it is important to acknowledge that research into language and gender is different from
research into language and sexuality, and both are different from research into language and
sexual identity.2 At the same time, however, these are all intertwined with each other and
therefore ‘push and pull’ each other, such that none can be completely separated from the
others in research and still be representative of real-life experience. As explained by Milani
(2014: 261–262), ‘despite acknowledging that sex, gender, and sexuality are separate cat-
egories, queer theory highlights that these are constructs that have been socially entwined
in such a way that they have developed “a unique relationship” (Sauntson 2008: 274) with
each other’.
As may be expected, these areas have been heavily influenced by poststructuralism
(Bucholtz 2014), which takes as a key principle the belief that we talk reality into being, and
it is through discourse that the world exists and is understood. As explained by the famous
feminist philosopher de Beauvoir, a person is not born a woman; rather, a person becomes a
woman as a result of social structures (1949: 18). It is thus through language that gender and
sexuality binaries are created, and through language that these binaries are contested. This
social construction of the body occurs through discourse, which contributes to and is reflective
of structures of power (Foucault 1990). Therefore, we use discourse and language literally to
talk the body into being (Bucholtz and Hall 2016). This process of socially constructing and
conceptualising the body carries with it all of the social constraints and embedded power rela-
tions, including marginalisation, that exist within a given sociocultural context (e.g. talking
about trying on dresses and how they fit, which carries assumptions that dresses are worn by
certain types of people and should fit a certain way, all of which gives power to discourses of
femininity, gender binaries, women’s bodies as desirable commodities, etc.).
Crucially, ‘social categories entail both gender and sexuality, and are socially con-
structed and culturally relative, changing over time’ (Kendall and Tannen 2015: 649).

229
Corinne A. Seals

This has been demonstrated by means of various cross-cultural analyses of the construc-
tion of gender and sexuality, which shows that the binary treatment of gender and sexual-
ity that exists in Anglo-dominated societies does not carry over to many other cultures,
including those that have been oppressed by the former. For example, Native American
Two-Spirit people (Driskill 2010; Wilson 1996), Thailand’s kathoey (Saisuwan 2016;
Totman 2004), Brazil’s travesti (Kulick 1998), and India’s hijras (Hall and O’Donovan
1996) are all well-recognised identities within their respective cultures, despite not being
acknowledged by Anglo-dominated hegemonic ideas of binary gender and sexuality.
Additionally, New Zealand Māori have embraced the revival of the term takatāpui to
represent a fluid, non-binary identity that encompasses gender, sexuality, and culture,
which was considered to be a normal way of identifying before English colonisation of
New Zealand (Kerekere 2017; Murray 2003). Similar in meaning are the terms fakaleiti
in Tonga and fa’afafine in Samoa (Farran 2010).

Socialisation into cultural expectations regarding gender, and sexuality

As evidenced through the above examples, language plays a key role in identifying the self
and other because of the wealth of cultural and social information intertwined. Furthermore,
any and all perceived differences in gender and sexuality from the hegemonic norm have
‘essentially cultural, rather than biological bases … the sum total of the parents’, the peers’,
and the culture’s notions of what is appropriate to each temperament…’ (Millett 1971:
28–31). These expectations of doing gender and sexuality are further governed within cul-
tures and societies from the very beginning of a person’s life through socialisation, in which
language again plays a key role. As argued by Goodwin (1990) who focused on children’s
playtime (with girls playing together and boys playing together), children are socialised
even at a young age into a gendered interactional order. Goodwin (1990) found that while
the girls tended to structure play jointly and indirectly, the boys tended to structure play
hierarchically and directly. By a young age, the children had already internalised the gender
interactional order into which they had been socialised, and they continued to reinforce it
through peer-to-peer socialisation during play.
Further expanding upon Gumperz’ (1982) contextualisation cues and Goodwin’s (1990)
socialisation into interactional order, Tannen’s book You Just Don’t Understand (1990) and
Holmes’s book Gendered Talk at Work: Constructing Gender Identity through Workplace
Discourse (2006) found that women and men appeared to take part in conversational rituals
reflecting hegemonic role socialisation, such that women tended repeatedly to support and
align with each other, while men tended repeatedly to ‘one-up’ each other in conversation.
Holmes (2006) furthermore specifically looked at how men and women draw upon societal
norms and expectations to perform gendered identity in the workplace, drawing upon con-
textualisation cues to make these connections. These gendered conversational rituals were
also discussed in Ochs and Taylor’s (1992) analysis of gendered dinner-time conversational
rituals in the United States, specifically examining how and what the children said to the
mothers as compared with the fathers. While enlightening from the ‘gender as difference’
perspective to IS language and gender research, this still focused on hegemonic conceptuali-
sations of sexuality. However, Wagner (2010) replicated Ochs and Taylor’s (1992) research,
examining conversational rituals in lesbian family dinner discourse. She found that while
some of the interactional norms from the original study held in the new study (such as one
parent more often than not taking a dominant role), these findings also pointed to the fact

230
 Benefits and challenges of IS

that such interactional strategies and contextualisation cues are not necessarily gendered
behaviour; rather, they are part of the way in which ritualised language functions within
families of all types.
Furthermore, the expectations of ‘doing gender’ via societal gender ideologies are
policed through language, such that people are expected to adhere to ‘precise hegemonic
notions of gender-appropriate language use’ (Bucholtz and Hall 1995: 6). Here is a further
case where contextualisation cues and the reciprocal relationship between language and
society are still highly relevant in language and sexuality research, as there still remain
stereotypes upon which people draw every day when interpreting gender performance and
sexuality (e.g. voice pitch and cadence). While third-wave feminist approaches to research
have done a fantastic job of ‘question[ing] the assumption that gender is a simple binary
opposition’ (Cameron 2006: 2), there is still a strong reliance upon ‘sex-class linked’ associ-
ations (Goffman 1977) when it comes to how people on average interpret interactions every
day. As explained by Tannen (1996: 713), Goffman’s phrase ‘sex-class linked’ describes the
regularised phenomenon by which

certain behaviours in certain cultures are more likely to be associated with members of
the “class” of females or males, but people come to regard such behaviours as associ-
ated not with the “class” but rather with each individual who is a member of it.

Benefts and limitations of using an interactional sociolinguistic

approach
As previously mentioned, IS is very useful for its micro- and macro-contextually situated
insights into interaction. However, there is a greater understanding now of how we can even
further strengthen IS by incorporating more critical perspectives, such as those central to
other types of discourse analysis (e.g. critical discourse analysis (CDA)). Acknowledging
the importance of the wider roles played by macro-societal power and pressures is an
important addition to IS as we grow our global awareness as researchers. Scholars such as
Cetina (2009) have spoken about the need for Goffman’s theory to be updated to account
for changes in modern society, such as asynchronous communication (e.g. leaving a text
message for someone to be read later) and human communication with computers (e.g.
buying movie tickets from an automated kiosk). Importantly, at the same time, scholars
such as Bullingham and Vasconcelos (2013) have argued that while not applicable in every
case, Goffman’s interaction order still explains much of what happens in modern-day com-
munication, including interaction in message boards and blogs, and identity projection via
avatars on internet-based platforms.
It is also important to address the misconceptions that lead to an often-found criticism
of all types of discourse analysis – that it is perceived to involve ‘cherry picking’ exam-
ples that best fit the researcher’s aims in that extracts are selected from a dataset then
analysed (most notably criticised by Widdowson 2004). It is important to note that this
is a possible danger with any research that relies upon illustrative examples, and this is
exactly why transparency of the research process and researcher reflexivity (i.e. account-
ing for the researcher’s own position in and influence on the data) are so crucial to validity
in qualitative reliability in research. As further explained by Gumperz (2015: 317), it is
not the role of IS to resolve all issues with ambiguity in interpretation. However, it is the
role of IS to hold the researcher accountable for the interpretations made by maintaining

231
Corinne A. Seals

transparency around these interpretations: ‘Instead of taking interpretive processes for

granted, IS analysis suggests (1) what the most likely interpretations are, (2) what the
assumptions and inferential processes are by which they are achieved, and (3) how they
relate to what is literally said’. Yet, this remains an area of caution because it is fully up
to the researcher to ensure transparency and reflexivity in the research process in order to
produce valid and trustworthy results.
Finally, IS researchers are aware of the need to curtail the heretofore reliance upon
Western conceptualisations of research. This awareness is one that is continuing to rise in
all areas of linguistics, as well as social science and humanities research broadly. This con-
sideration is particularly important when working with Indigenous and minority communi-
ties because sole application of a Western methodology (i.e. with its tradition in research by
traditional colonial settler countries, cf. Kurtz 2013; Tuhiwai Smith 2013) when working
with Indigenous and minority communities is now understood to continue oppressive prac-
tices (Tuhiwai Smith 2013). Rather, when working with such communities, the research
must be done alongside the communities and must be done for their interests. This usually
involves also adopting Indigenous methodologies alongside methodologies such as IS, with
the former having at its core ‘an understanding of the significance of Indigenous knowledge
and the ways in which Indigenous people make sense of life in today’s world’ (Kurtz 2013:
219; see also Ray 2012). Furthermore, the researcher cannot rely upon their own contextual
interpretations alone, but must consult continually with the communities with which they
are working.
Now that some of the benefits and limitations of using an IS approach have been
addressed, a case study is presented below that illustrates how IS can be useful for research
into language and sexuality. Following the case study, there is a further discussion of the
merits of using multiple methods including IS.

Discussion through a case study illustration

In order to better illustrate the usefulness of IS in studying language and sexuality, this
section looks to an example of research I conducted on a lesbian comedian’s interaction
with her audience (Seals 2016). This is an excerpt from comedian Julie Goldman’s stand-
up comedy routine in 2007 for Comedy Central in New York City. Julie Goldman is a
Jewish, openly lesbian comedian from the East Coast of the United States who identifies
as ‘butch’ in style and was in her mid-30s at the time of the performance. Comedy Central
is a major network on cable television in the United States, and their stand-up specials
are a feature of their programming. A key feature of their audience is that the comedian
does not know in advance who will be in their audience, nor does the audience know who
will be performing. The audience is recruited through hired representatives going onto
the streets of Times Square and asking people if they would like to see a comedy show
(Seals 2016). Consequently, the audience does not lean socially or politically one way
or the other, and the comedian cannot assume that the audience would have otherwise
chosen to be there had they known who the comedian would be. This unknown factor is
very important to keep in mind, as it makes the comedian’s interactional work even more
challenging.
In the example below, Goldman tells a narrative-style joke, drawing upon her own expe-
rience as intersectionally Jewish, gay, butch, a woman, a New Yorker, and partnered. In the
story below, Goldman tells of the hegemonic resistance she faced while shopping for her
upcoming wedding to her partner.

232
 Benefits and challenges of IS

Example: Shopping in New York City3

1 G: But you know. It’s hard for the butch lez. We’re very shy. Very sensitive.
2 And the butch lez shops in three places. Old Navy, The Gap, and when we
3 feel fancy Banana Republic. So.
4 (audience laughter)
5 I wanted to extend. Right? I live in New York City. I should be able to
6 find something great, right?
7 So I go to all these places, and it was very difficult for me. I went to
8 Barney’s. Went to Bergdorf’s. Sacks Fifth Avenue. All- Bloomingdales.
9 All the places. And it’s very hard.
10 You know why? Because the same bitch works in every store.
11 (quiet audience laughter)
12 ((grabs mic stand and drops it center-stage)) She looks like this!
13 (audience laughter and applause)
14 ((Gestures to show off mic stand four times)).
15 With blonde hair.
16 (quiet audience laughter)
17 And an accent from somewhere fabulous.
18 (quiet audience laughter)
19 We don’t know what country, just fabulous.
20 (quiet audience laughter)
21 And she clanked up to me with her clanky bones. Clank clank clank clank
22 clank clank clank clank clank clank clank clank clank clank clank clank
23 clank clank.
24 (quiet audience laughter)
25 ((Impersonates store clerk)) Can I help you:?
26 (quiet audience laughter)
27 The butch is shy. I was like ‘I’m looking for an outfit for my wedding!
28 Hmph!’ ((kicks and acts fussy))
29 (quiet audience laughter)
30 And it’s amazing! Even in New York City! Even in the two-thousands!
31 When we’re supposed to be enlightened and all enlightened and our heads
32 are all how-a-how.
33 But still people put you in a- in a box. People put you in ‘you’re supposed
34 to be a label. You’re supposed to be this.’
35 So when I said I’m looking for an outfit for my wedding, this woman saw
36 just a woman, which means I’m a bri:de. Which to her means I’m just a
37 woman, which means I’m just competi:tion.
38 So her whole inner monologue was fascinating.
39 (quiet audience laughter)
40 Because her whole illusion of reality just exploded!
41 (quiet audience laughter)

Insights gained from interactional sociolinguistics

In choosing a method of discourse analysis for the above excerpt, I chose to use IS due to its
focus on interpretation through micro- and macro-contextual information. This multi-scaled

233
Corinne A. Seals

contextual information proved absolutely vital to understanding the interactional choices

being made by Goldman.
First, an analysis of the contextualisation cues in Goldman’s stand-up performance shows
how she strategically uses exaggerated intonation and gesture to signal to her audience where
laughter is expected. As previously mentioned, Goldman does not know who is in her audi-
ence, nor do they know they would be listening to Goldman as the comedian of the night.
Therefore, contextualisation cues marking the expected points of audience response play an
important role in the success of her performance. For example, in line 1, Goldman emphasises
the words ‘butch lez’, ‘very shy’, and ‘very sensitive’ through elongation and changes in inten-
sity. Through doing so, she marks these concepts as being of particular salience for setting up
the story in the joke she is telling. In lines 2 and 3, she then draws upon place names that she
assumes are familiar to her audience so that she can also access the schemata associated with
them – rising in affiliated socioeconomic status as she lists them. She then utters the word ‘so’
in an abrupt stopping manner as a marker of the end of a turn, signalling the audience’s turn to
laugh, as well as a discourse marker that she will continue her joke after their laughter.
In the lines that follow, Goldman uses the same types of contextualisation cues for her
audience, in addition to rhetorical questions in an effort to keep the audience engaged and
to invite them to enter the joke setting with her from her perspective (e.g. lines 5, 6, 10).
Furthermore, when Goldman begins to hit crucial punch line points during her joke-telling
narrative, she draws upon contextualisation cues such as gesture (e.g. lines 12 and 14),
repetition (lines 21 and 22), and exaggerated intonation when voicing the other (e.g. line
25). As shown in the example above, these contextualisation cues are successful for her
comedic performance, as 9 out of 11 instances of audience laughter occur immediately after
Goldman uses exaggerated intonation, exaggerated repetition, or exaggerated gesture.
Furthermore, in analysing Goldman’s discourse, it is immediately noticeable that she
draws upon larger societal discourses and macro-contexts to present herself intersectionally
(Crenshaw 1989, 1991). Not only does she provide many references to various aspects of
her identity (e.g. lines 1, 2, 5, 27), but she even directly criticises the store clerk’s lack of
acknowledgement of intersectionality in lines 35–37. By simultaneously directly criticising
heteronormative perspectives, Goldman uses her performance to resist hegemonic frames
and social stereotypes, which has been found to be a feature of LGBTQ comedy in particu-
lar (Kulick 2014; Seals 2016).
However, Goldman does not simply present discourses of resistance in the example
above. She also draws upon stereotypes and re-keys them, making fun of them in the form
of the store clerk and therein further empowering LGBTQ communities (Bing and Heller
2003). In making fun of the store clerk’s ignorance, Goldman is also able to reframe the
interaction as a way to empower the self by resisting other-defining and, instead, self-defin-
ing (Willard 2010). In addition to the abundance of information gained from IS, the next
section shows how we can gain even more insight by also utilising intertextuality.

Supplementary use of intertextuality

This section shows the further information that can be added to the IS analysis by also utilis-
ing a poststructuralist concept such as intertextuality (Bakhtin 1981) – ‘the joining together
of ideas through texts in communication, while simultaneously reacting to and reflecting
primed prior texts and anticipating unrealised future texts’ (Seals 2012: 233).
Intertextuality provides another look at the contextualisation cues put forth by Goldman.
More specifically, intertextuality allows us to examine the underlying discursive references

234
 Benefits and challenges of IS

that Goldman draws upon for her diverse audiences and how she makes these connections
more explicit for an audience of an unknown quantity (i.e. how much of the audience are
lesbian, gay, bisexual, transgender, and/or queer (LGBTQ) community members). For
example, she refers to herself as a ‘butch lez’ multiple times and describes (albeit ironi-
cally) traits assigned by LGBTQ communities to someone who identifies as such. For an
in-group audience, this would be too much time spent on already known information, and in
fact, Goldman does not do this in her comedy for LGBTQ-specific audiences (Seals 2016).
Her assumption about her audience not having intertextual references for this information
appears to be accurate because her audience does not laugh at this description in line 1.
However, the description is humorous from an LGBTQ perspective because she has drawn
upon the irony of a discourse of a ‘butch lez’ as sensitive and shy, which is the opposite of
the dominant societal discourses of butch-identifying lesbians within LGBTQ communities
(cf. Kulick 2014; Wagner 2010). It is at this point, having her audience assumption con-
firmed, that she begins drawing more heavily upon contextualisation cues in her comedy,
such as exaggerated intonation and gesture.
This intertextual gamble again takes places when Goldman directly criticises the hegem-
onic notions of what it means to be a woman and what it means to be gay in lines 31–37.
While this type of social critique is very common in lesbian comedy and would likely elicit
laughter from an LGBTQ-specific audience due to shared experiences and therefore shared
intertextual referents (Becker 1994; Bing and Heller 2003), the degree of presence of these
referents for her current audience is unknown. In fact, her audience does not laugh, indicat-
ing the high likelihood that they do not in fact share this intertextual discourse. Consequently,
Goldman returns to employing contextualisation cues in the form of exaggerated intonation
and gesture in line 38, again rewarding her with audience laughter.
By drawing upon intertextuality as well as IS, we are able to uncover additional informa-
tion about the very fast audience measurement and negotiation that Goldman undergoes in
order to interactionally adjust her comedy to achieve a successful effect.

Concluding comment
This case study example speaks to the usefulness of employing post-structuralist intertex-
tuality theory supplementarily to IS when analysing discourses of sexuality in order to get
a more holistic understanding of the data. While IS prioritises interaction in discourse and
takes account of needed contextual information, there is still more that can be added to
this. Had only IS been used to analyse the example above, the degree of Goldman’s skill in
adjusting her discourse for her unknown audience would have gone unnoticed. The post-
structuralist concept of intertextuality added to IS allowed for a discovery of potentially
unsuccessful comedy in that Goldman’s audience may have missed LGBTQ intertextual
references, while also uncovering the quick recovery made by her employ of specific con-
textualisation cues for audience laughter. In short, while neither IS nor intertextuality would
provide as complete an analysis, utilised together they provide much more insight into the
inner workings of the interaction.

Future directions
All future IS work done with real people, and in particular with marginalised individuals and
communities, must put intersectionality at the forefront of any interpretive analysis. This is
already done in much IS research but needs to be a priority for all IS research. Furthermore,

235
Corinne A. Seals

we need more future IS research focusing on all three areas of language and gender, lan-
guage and sexuality, and language and sexual identity (instead of considering only one or
two aspects at a time) in order to get a more holistic picture of lived human experience.
Future research in IS should also focus more on supporting the experiences of marginal-
ised groups. However, when working with marginalised individuals and communities, it is
of utmost important that IS researchers remember to research with, not research on, them.
Hopefully, with increasing focus on the ethics of researching with Indigenous and minority
peoples will also come increasing interest in creating a platform for often-silenced voices
to be heard. This will require a deep, contextual understanding of how genders, sexualities,
and sexual identities are dynamically conceptualised within local communities.
Finally, this raising of social and cultural consciousness also points to the importance of
employing supplementary concepts and methods in future IS research, including Indigenous
ways of understanding sexuality in and through language. As has been reiterated several
times throughout this chapter, IS has much to offer the study of language and sexuality, and
the benefits can be further enhanced when in conjunction with other methods.

Transcription conventions
A: first initial of speaker’s surname
? rising final intonation
! strong emphasis, with falling intonation
. falling, final intonation
, low-rising intonation suggesting continuation
te:xt lengthening of the preceding sound
tex- an abrupt cut-off, with level pitch
text marked stress
((text)) non-verbal actions or researcher notes
(text) verbal feedback from the audience

Notes
1 Dominant ideas about the way the world works that exist at a larger societal level (e.g. a ‘family’
consists of two heteronormative parents and their children).
2 ‘Sex’ refers to a biological sex distinction (often societally binary) that is assigned at birth (i.e. terms
like ‘female’, ‘male’, and ‘intersex’). ‘Gender’ is the social expression of identity along a feminin
e<>‘nonconforming’<>masculine scale (i.e. people are often assigned (and assign) terms such as
‘woman’, ‘man’, or ‘gender fluid’). ‘Sexual identity’ refers to who a person is romantically or sexu-
ally attracted to, and the ways in which a person defines themselves based on this identity. ‘Sexuality’
is often conflated with sexual identity, but refers to how and what a person desires and how they
express that desire (such as through language).
3 Refer to transcription conventions at the end of the chapter.

Further reading
Cameron, D. and Kulick, D. (2003) Language and sexuality. Cambridge: Cambridge University Press.
This book provides a thorough overview of the history of research into language and sexuality,
including the many complexities involved.
Motschenbacher, H. (2011) ‘Taking queer linguistics further: sociolinguistics and critical
heteronormativity research’, International Journal of the Sociology of Language, 212(212), pp.
149–179.

236
 Benefits and challenges of IS

This article focuses specifically on queer linguistics and discusses the importance of a critical
perspective in sociolinguistics research.
Wagner, S. (2010) ‘Bringing sexuality to the table: language, gender and power in seven lesbian
families’, Gender and Language, 4(1), pp. 33–72.
This article revisits a classic gender and language study, replicating it with lesbian families to test
the original claims made and to incorporate research into language and sexuality.

Related topics
Performance in action; applying queer theory to language, gender and sexuality research in schools;
doing gender and sexuality intersectionally in multimodal social media practices; interactional
sociolinguistics: foundations, developments and applications to language, gender, and sexuality;
gender and sexuality in discourse: semiotic and multimodal approaches.

References
Angouri, J. (2018) Culture, discourse, and the workplace. Abingdon: Routledge.
Angouri, J., Marra, M., and Holmes, J. (eds) (2017) Negotiating boundaries at work. Edinburgh:
Edinburgh University Press.
Bakhtin, M. (1981) ‘Discourse in the novel’, in Holquist, M. (ed) The dialogic imagination: four
essays. Austin, TX: University of Texas Press, pp. 259–422.
Becker, A. L. (1994) ‘Repetition and otherness: an essay’, in Johnstone, B. (ed) Repetition in discourse,
vol. 2. Norwood: Ablex, pp. 162–175.
Bing, J. and Heller, D. (2003) ‘How many lesbians does it take to screw in a light bulb?’, Humor:
International Journal of Humor Research, 16(2), pp. 157–182.
Bucholtz, M. (2014) ‘The feminist foundations of language, gender, and sexuality research’, in
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) The handbook of language, gender, and sexuality,
2nd edn. Oxford: Wiley-Blackwell, pp. 23–47.
Bucholtz, M. and Hall, K. (1995) ‘Introduction: 20 years after language and woman’s place’, in Hall,
K. and Bucholtz, M. (eds) Gender articulated: language and the socially constructed self. New
York: Routledge, pp. 1–22.
Bucholtz, M. and Hall, K. (2016) ‘Embodied sociolinguistics’, in Coupland, N. (ed) Sociolinguistics:
theoretical debates. Cambridge: Cambridge University Press, pp. 173–200.
Bullingham, L. and Vasconcelos, A. C. (2013) ‘The presentation of self in the online world: goffman
and the study of online identities’, Journal of Information Science, 39(1), pp. 101–112.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Cameron, D. (2006) On language and sexual politics. London: Routledge.
Cameron, D. and Kulick, D. (2003) Language and sexuality. Cambridge: Cambridge University Press.
Cetina, K. (2009) ‘The synthetic situation: interactionism for a global world’, Symbolic Interaction,
32(1), pp. 61–87.
Coates, J. and Cameron, D. (eds) (1989) Women in their speech communities: new perspectives on
language and sex. New York: Longman.
Crenshaw, K. (1989) ‘Demarginalizing the intersection of race and sex: a black feminist critique of
antidiscrimination doctrine, feminist theory and antiracist politics’, University of Chicago Legal
Forum, 140, pp. 139–167.
Crenshaw, K. (1991) ‘Mapping the margins: intersectionality, identity politics and violence against
women of color’, Stanford Law Review, 43(6), pp. 1241–1299.
de Beauvoir, S. (1972[1949]) The second sex. New York: Routledge.
Driskill, Q.-L. (2010) ‘Doubleweaving two-spirit critiques: building alliances between native and
queer studies’, GLQ: A Journal of Lesbian and Gay Studies, 16(1), pp. 69–92.
Farran, S. (2010) ‘Pacific perspectives: fa’afafine and fakaleiti in Samoa and Tonga: people between
worlds’, Liverpool Law Review, 31(1), pp. 13–28.

237
Corinne A. Seals

Foucault, M. (1990) The history of sexuality an introduction, vol. 1. New York: Random House Inc.
Gee, J. P. (1996) Social linguistics and literacies: ideology in discourses, 2nd edn. London: Taylor &
Francis.
Goffman, E. (1974) Frame analyses: an essay on the organization of experience. Cambridge: Harvard
University Press.
Goffman, E. (1977) ‘The arrangement between the sexes’, Theory and Society, 4(3), pp. 301–331.
Goffman, E. (1983) ‘The interaction order: American Sociological Association, 1982 presidential
address’, American Sociological Review, 48(1), pp. 1–17.
Goodwin, M. H. (1990) He-said-she-said: talk as social organization among black children.
Bloomington: Indiana University Press.
Gumperz, J. J. (1982) Discourse strategies. Cambridge: Cambridge University Press.
Gumperz, J. J. (1999) ‘On interactional sociolinguistic method’, in Sarangi, S. and Roberts, C. (eds)
Talk, work and institutional order: discourse in medical, mediation and management settings. New
York: Mouton de Gruyter, pp. 453–472.
Gumperz, J. J. (2015) ‘Interactional sociolinguistics: a personal perspective’, in Tannen, D., Hamilton,
H. E., and Schiffrin, D. (eds) The handbook of discourse analysis, 2nd edn. Oxford: Wiley
Blackwell, pp. 309–323.
Hall, K. and O’Donovan, V. (1996) ‘Shifting gender positions among Hindi-speaking hijras’, in
Bergvall, V. L., Bing, J. M., and Freed, A. F. (eds) Rethinking language and gender research:
theory and practice. New York: Longman, pp. 228–266.
Holmes, J. (2006) Gendered talk at work: constructing gender identity through workplace discourse.
Malden, MA: Blackwell.
Holmes, J. and Marra, M. (eds) (2010) Femininity, feminism and gendered discourse. Newcastle upon
Tyne: Cambridge Scholars Press.
Kendall, S. and Tannen, D. (2015) ‘Discourse and gender’, in Tannen, D., Hamilton, H. E., and
Schiffrin, D. (eds) The handbook of discourse analysis, 2nd edn. Oxford: Wiley-Blackwell, pp.
639–660.
Kerekere, E. (2017) Part of the Whānau: the emergence of Takatāpui identity: He Whāriki Takatāpui.
Unpublished thesis. Victoria University of Wellington.
Kulick, D. (1998) Travesti: sex, gender, and culture among Brazilian transgendered prostitutes.
Chicago: University of Chicago Press.
Kulick, D. (2014) ‘Humorless lesbians’, in Chiaro, D. and Baccolini, R. (eds), Gender and humor:
interdisciplinary and international perspectives. New York: Routledge, pp. 85–99.
Kurtz, D. L. M. (2013) ‘Indigenous methodologies: traversing indigenous and western worldviews in
research’, AlterNative: An International Journal of Indigenous Peoples, 9(3), pp. 217–229.
Kuzio, A. (2014) Exploitation of schemata in persuasive and manipulative discourse in English,
Polish and Russian. Newcastle upon Tyne: Cambridge Scholars Publishing.
Leap, W. L. (ed) (1995) Beyond the lavender lexicon: authenticity, imagination, and appropriation in
lesbian and gay languages. New York: Routledge.
Livia, A. and Hall, K. (eds) (1997) Queerly phrased: language, gender, and sexuality. New York:
Oxford University Press.
Lopez, Q. (2014) ‘Aggressively feminine: the linguistic appropriation of sexualized blackness by
white female characters in film’, Gender and Language, 8(3), pp. 289–310.
Milani, T. M. (2014) ‘Queering masculinities’, in Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) The
handbook of language, gender, and sexuality, 2nd edn. Oxford: Wiley-Blackwell, pp. 260–278.
Millett, K. (1971) Sexual politics. London: Granada Publishing.
Murray, D. A. B. (2003) ‘Who is takatāpui? Māori language, sexuality and identity in Aotearoa/New
Zealand’, Anthropologica, 45(2), pp. 233–244.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds), Rethinking context: language
as an interactive phenomenon. Cambridge: Cambridge University Press, pp. 335–358.
Ochs, E. and Taylor, C. (1992) ‘Mothers’ role in the everyday reconstruction of “father knows best”.’
In: Hall, K., Bucholtz, M., and Moonwomon, B. (eds) Locating power proceedings of the second

238
 Benefits and challenges of IS

Berkeley women and language conference. Berkeley. Berkeley Women and Language Group, pp.
447–463.
Oute, J. and Huniche, L. (2017) ‘Who gets involved with what? a discourse analysis of gender and
caregiving in everyday family life with depression’, Outline: Critical Practice Studies, 18(1), pp.
5–27.
Pichler, P. (2017) ‘“You are stupid, you are cupid”: playful polyphony as a resource for affectionate
expression in the talk of a young London couple’, Gender and Language, 11(2), pp. 153–175.
Ray, L. (2012) ‘Deciphering the “Indigenous” in indigenous methodologies’, AlterNative: An
International Journal of Indigenous Peoples, 8(1), pp. 85–98.
Remlinger, K. (1999) ‘Widening the lens of language and gender research: integrating critical
discourse analysis and cultural practice theory’, Linguistik Online, 2. Available at: www.lingui
stik-online.de/heft1_99/remlinger.htm.
Saisuwan, P. (2016) ‘Kathoey and the linguistic construction of gender identity in Thailand’, in
Levon, E. and Beline Mendes, R. (eds) Language, sexuality, and power: studies in intersectional
sociolinguistics. Oxford: Oxford University Press, pp. 189–214.
Sauntson, H. (2008) ‘The contributions of queer theory to gender and language research’, in
Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds) Gender and language
research methodologies. New York: Palgrave Macmillan, pp. 271–282.
Sauntson, H. and Kyratzis, S. (2007) ‘Introduction: language, sexualities and desires’, in Sauntson, H.
and Kyratzis, S. (eds) Language, sexualities and desires: cross-cultural perspectives. New York:
Palgrave Macmillan, pp. 1–16.
Schiffrin, D. (1994) Approaches to discourse. Oxford: Blackwell Publishers Ltd.
Seals, C. A. (2012) ‘When a “non-issue” becomes an issue in discourse surrounding LGBT
communities’, Journal of Language and Sexuality, 1(2), pp. 230–255.
Seals, C. A. (2016) ‘The constrastive use of humor by a lesbian comedian for LGBT and general
audiences’, de genere, Journal of Literary, Postcolonial and Gender Studies, 2(1), pp. 95–107.
Spender, D. (1985) Man made language. London: Pandora Press.
Tannen, D. (1990) You just don’t understand: women and men in conversation. New York: Ballantine.
Tannen, D. (1996) ‘The sex-class-linked framing of talk at work’, in Gender and discourse. New
York: Oxford University Press, pp. 195–221.
Totman, R. (2004) The third sex: kathoey – Thailand’s ladyboys. London: Souvenir Press.
Tuhiwai Smith, L. (2013) Decolonizing methodologies: research and indigenous peoples, 2nd edn.
London: Zed Books Ltd.
Wagner, S. (2010) ‘Bringing sexuality to the table: language, gender and power in seven lesbian
families’, Gender and Language, 4(1), pp. 33–72.
Warner-Garcia, S. (2015) ‘“His belly dancer”: young women’s interactional negotiation of sexual
bodies and desire at a Baptist University’, Gender and Language, 9(2), pp. 255–277.
Widdowson, H. G. (2004) Text, context, pretext: critical issues in discourse analysis. Oxford:
Blackwell.
Willard, M. (2010) Homo-humor: laughter as a protective factor in the face of homophobia and
heterosexism. Unpublished thesis. Alliant International University.
Wilson, A. (1996) ‘How we find ourselves: identity development and two spirit people’, Harvard
Educational Review, 66(2), pp. 303–318.

239
 Part IV
Ethnomethodological and
conversation analytic approaches
 16
The accomplishment of gender in
interaction
Ethnomethodological and conversation
analytic approaches to gender

Lorenza Mondada (Part IV lead)

Introduction
Ethnomethodology (EM) and conversation analysis (CA), together known as EMCA, con-
stitute an interdisciplinary conceptual and methodological programme that does not have
gender as its main focus, but has contributed to the issues of gender and sexuality in origi-
nal ways in the last two decades (see McIlvenny 2002b; Speer and Stokoe 2011; Stokoe
and Weatherall 2002). This chapter aims at characterising the specificity of the analyses
of gender and sexuality that can be offered from this viewpoint. These analyses can be of
interest for feminist studies, masculinities studies, queer studies, LGBTQ+ studies. They
also provided arguments in favour of activist and empowering approaches to the diversity of
gendered and sexual practices. This chapter does not, however, aim to discuss these connec-
tions that have largely been debated elsewhere, sometimes in controversial ways (Kitzinger
2000; see Tennent and Weatherall, this volume). Rather, this chapter reviews the principles
of EMCA in order to sketch its original contributions to the study of gender as it is made
relevant by and oriented to the participants within social interactions.
EM and CA are two related paradigms that emerged in the 1970s. Both appeared as
distinctive approaches in the landscape of sociology and social theory for their primary
focus on social action as locally situated and jointly accomplished (rather than on individ-
ual action as being governed by interiorised rules, beliefs, values, and norms). Garfinkel
founded EM with a special interest in how practical activities are locally organised by
members as well as in how they are produced and recognised as having a public intel-
ligibility, that is as accountable (Garfinkel 1967, 2002; see Heritage 1984 for an introduc-
tion). Issues of accountability are crucial for the way in which gender and sexuality are
not pre-existing essential features but practical achievements constantly scrutinised and
interpreted by others.
This intellectual background is also shared by CA, originating from the thinking of Sacks
(1992), who had worked closely with Garfinkel, as well as in collaboration with Schegloff

243
Lorenza Mondada

and Jefferson, with whom he launched CA (Sacks et al. 1974). A basic principle of conver-
sation analysis is that social interaction is methodically organised through the situated, and
yet systematic, ordering of participants’ actions in a temporal, sequential, and interactional
way. This order is based on procedures for allocating turns, on the recognisable design of
actions, on sequence organisation, and on practices for repairing trouble (Schegloff 2007:
xiv), which are at the same time context-sensitive (that is, contingently situated) and con-
text-free (that is, operating across contexts).
In both EM and CA, a central issue is thus how action is meaningfully formatted within
the ongoing interaction; how its order, intelligibility, and normative expectations are pub-
licly recognised and responded to by the co-participants. Accountability concerns both intel-
ligible and normative features. It relies on what the participants make relevant and towards
which they demonstrably orient in a way that matters for them and that is consequential for
the further progression of the activity. In both EM and CA, these questions are fundamental
for understanding the establishment, reproduction, and change in what constitutes the social
order as it is achieved in and through the detailed organisation of practices within social
interaction.
These principles have generated a substantial body of analytical work questioning the
issue of gender in social interaction, which has contributed to gender, feminist, queer, mas-
culinities, and LGBTQ+ studies. An important characteristic of these contributions is the
focus on situated actions and their sequential environment in social interaction, documented
as they temporally unfold in naturally occurring settings thanks to carefully transcribed and
analysed audio/video recordings. As put by Schegloff (1997), issues of gender – like other
phenomena – are discovered as endogenous phenomena that matter for members and are
unpacked by rigorous analysis (rather than imposed exogenously from the perspective of the
analyst). The use of naturalistic data – data that is not elicited, orchestrated, or arranged by
the researcher – rather than interviews, is of great importance in EMCA because they are the
materials that exhibit how members orient and treat gender categories within the local con-
tingencies and contextures of the situated activity. For example, EMCA studies of gender
have looked at very different social settings and activities, ordinary and institutional, involv-
ing adults and children, including: how presumptions of heterosexuality are manifested in
calls to various institutions (Kitzinger 2005a, 2005b; Land and Kitzinger 2005); how gender
issues permeate divorce mediation sessions (Garcia and Fisher 2011) and marriage guid-
ance counselling (Edwards 1998); how gays and lesbians engage in subtle forms of coming
out (Kitzinger 2000); how heterosexuals mimic the practice of coming out (Mondada and
Oloff 2015); how transsexual women organise passing in the identity clinic (Speer 2009);
how children make relevant their distinct gender culture in the playground (Goodwin 1990,
2006); how children playing engage in passing (Bulter and Weatherall 2011); how victims
of violence establish themselves as such in calls for help (Tennent and Weatherall, this
volume); how the heteronormative image of the family is reproduced in everyday activities
(Ericcson, this volume); how gendered identities are performed in girls’ gossip disputes
(Evaldsson, this volume; see also Goodwin 1990); and how drag kings perform walking in
a masculine way (Greco, this volume).
This chapter reviews some EMCA work on gender. It begins with Garfinkel’s pioneering
study of Agnes (1967), a case that has been consequential for uncovering gender as a social
accomplishment. Next, it analyses gender as a category in reference to Sacks’ membership
categorisation analysis (1972). Finally, the importance of the sequential organisation of talk
for how gender is made locally relevant by the participants is discussed on the basis of
empirical analyses.

244
 Accomplishing gender in interaction

Gender as a situated practical accomplishment

The case of Agnes, an intersexed person who had requested sex-reassignment surgery, is
exemplary for Garfinkel’s conceptualisation of how accountability is constantly achieved
within the situated and methodical organisation of practical actions. In this sense, the case
of Agnes is both a key contribution to the overall conceptual framework of EM, and a foun-
dational contribution to the understanding of how members treat gender as both naturally
evident and constantly practically managed.
Garfinkel met Agnes (a pseudonym) at the Department of Psychiatry at the University of
California at Los Angeles in 1958. Since she had a female body shape with male genitals,
she was diagnosed as having a testicular feminisation syndrome and was asserting the right
to surgery. Garfinkel recorded 35 hours of conversation with her. He was fascinated by the
fact that Agnes was reproducing the ‘natural attitude’ towards gender and sex – an unprob-
lematic binary vision of sexes and a heterosexual vision of gender – although she had a body
that did not correspond to these ‘facts’. Agnes thus became a perspicuous case for under-
standing at the same time the natural attitude adhering to sex as a taken-for-granted biologi-
cal ‘fact’, and the social achievement of sex status (the term used by Garfinkel for gender)
through repeated, ordinary, and situated practices. In her everyday life, Agnes would con-
stantly learn how to be a female by observing other women in their ordinary affairs and
listening to their female autobiographies, immediately integrating these features in her own
displays. She would engage in passing as a woman, behaving like a woman, managing
an appropriate and accountable feminine behaviour, talking, sitting, walking, going to the
beach as a woman, carefully avoiding contingencies that could disclose her genitalia. In
front of Garfinkel and the doctors of the Department of Psychiatry, Agnes was claiming
her moral right to have a surgical operation, defining herself as a normal natural female,
despite the incongruous anatomical factuality of her genitalia. The operation was for her a
‘correction’ of an ‘original error’, a necessary remedy of her condition in the direction of
that ‘intended by nature’.
In sum:

We learned from Agnes, who treated sexed persons as cultural events that members
make happen, that members' practises alone produce the observable-tellable normal
sexuality of persons, and do so only, entirely, exclusively in actual, singular, particular
occasions through actual witnessed displays of common talk and conduct.
(Garfinkel 1967: 181)

Agnes produced and reproduced her sex status as an institutional fact, showed how the
‘natural attitude’ was a practical and continuous accomplishment, and demonstrated the
reflexive relationship between social practices and accounts of those practices.
Garfinkel’s contribution has been influential for further ethnomethodological explora-
tions of gender (Kessler and McKenna 1978) and studies of passing (Butler and Weatherall
2011; Speer 2009), as well as for understanding gender as a ‘practical accomplishment’ in
relation to wider discussions about gender as ‘social construction’ and gender ‘performa-
tivity’ and their conceptual differences (Bucholtz and Hall 2004; Butler 1990; McIlvenny
2002a; Speer 2005; West and Zimmerman 1987).
Garfinkel’s exemplary analysis opened up important avenues for the study not only of
the achievement/performance of gender, but also of the way in which sexuality is actually
achieved in institutional and other settings. Further studies of how sexuality is discovered

245
Lorenza Mondada

remain few. A few examples are studies of parents inspecting the sex of newborns (Lindström,
Näslund, and Rubertsson 2015), deciding the sex of infants born intersexed (Kessler 1998),
and transformed, for example, in surgical operations (Hirschauer 1991) or studies of drag-
king make-up sessions (Greco, this volume).

Gender and membership categorisation devices

Sacks’ membership categorisation analysis (1972, 1992) provides another way of reflecting
upon the accountability of social conduct. Inspired by his fieldwork in a suicide prevention
call centre, Sacks described the search of suicidal persons for help as a methodical inspec-
tion of their close social environment, identifying and categorising who could become an
incumbent of the category ‘helper’. A ‘helper’ is not only somebody who is available to give
help, but also somebody who accepts to enter into a ‘helper’/’helped’ standardised pair rela-
tionship, characterised by rights and obligations to ask and receive help (Sacks 1966, 1972).
Sacks showed that the trajectory of the suicide ends when the conclusion of this search is, ‘I
have no one to turn to’ (1966). On the basis of this categorical description of the actions cru-
cially defining the life of the suicidal, Sacks defined membership categorisation devices as
constituted by a collection of categories (such as age, nationality, profession, gender), each
of which contained a set of categories that could be applied to a population. He also defined
a set of application rules, including the economy rule (one category is generally enough)
and the consistency rule (in order to categorise a population, the categories of the same col-
lection might be used), as well as a series of maxims that defined how the categories can be
heard and seen by the co-participants (1972). Sacks also highlighted that explicit categories
are not always used. Instead, categorisation practices are tacitly recognisable in relation
to category-bound activities; that is, actions that are typically attributed to a category. In
this case, the action enables participants to infer the category (e.g. Sacks 1992, vol. I: 46).
Another way actions can be associated to categories is through the accountable conduct that
can be identified as ‘doing being’ a given category, such as ‘doing being ordinary’ (Sacks
1992, vol. I: 215) or ‘doing gender’ (Kitzinger 2009; West and Fenstermaker 1995; West
and Zimmerman 1987). These latter points show that categorisation is not only built upon
the naming of the category in lexical items but is also (mostly) inferred.
Sacks insisted on the endogenous attribution of categories and categorisation work: cate-
gories are made relevant by and for members in the accountable organisation of their affairs;
not all categories that could be referentially valid are relevantly oriented (Sacks 1963). The
relevance of categories is thus the analytic central focus. This relevance is established in
and through actual instances of social interactions, and possibly evolves dynamically as the
activity unfolds.
In this sense, membership categorisation analysis and sequential analysis are deeply
interconnected, even if these two approaches within EMCA have sometimes been separated.
They are closely related in Sacks’ own work, This is shown, for example, by his analysis of
a recurrent sequential pattern noticed in the suicidal calls in which callers reported that when
they announced their suicide tendency (as in, ‘I’m going to kill myself’), they were often
met with laughter (Sacks 1992, vol. I: 14–16). Laughter is an action that treats the second
position within this sequence as a locus for aligning and affiliating to the search for help – or
not. Laughter is a response that fills this slot. It treats the first turn as a joke, thereby rejecting
the action initiated in the first position, also rejecting the rights and obligations that go with
it. This shows how a category, and the moral duties associated with it, can be negotiated
within the turn-by-turn sequential organisation of the interaction.

246
 Accomplishing gender in interaction

In this sense, categorisation practices and sequential organisation of action are deeply
intertwined. This is demonstrated by a simple example commented on by Sacks:

Extract 1 (from Sacks 1992, vol. I: 774)

Passenger: Do you have a cigarette?
Stewardess: No we don’t. They don’t provide that service anymore.

The example is not from a recorded corpus, but from a personal experience of Sacks himself
on a flight (at a time in which people were still smoking on flights). Sacks explained that the
passenger action – a request – is very different if addressed to the stewardess or to another
passenger. The response indeed displays that the request is heard as an institutional action,
made by a client to a service provider, and not, for example, by a man to a woman:

She is talking “as a stewardess”, the question is not whether she has a cigarette, but
whether, as a stewardess she has a cigarette, which stands altogether independent of
whether she has cigarettes in her purse. The issue is whether, as a stewardess, she pro-
vides cigarettes.
(Sacks 1992: 774)

The use of pronouns in the response is significantly oriented to that: the stewardess responds
first with the first-person plural (‘we don’t’) and then adds an account in which she uses
the third-person plural (‘they don’t provide that service anymore’), further pointing at a
responsibility that falls neither on her nor the crew. Both pronouns index that she responds
as a member of an organisation and not as a private person. This is very different than if
the passenger had asked for a cigarette from a woman sitting next to him, using that action
as a legitimate procedure for engaging with her, in particular. It is striking that the relevant
categories are explicitly inscribed in the identification of the participants in the transcripts:
the result of the analysis is incorporated in the presentation of the data itself (for a discus-
sion, see Mondada 2002).
This example shows the importance of the sequential organisation of talk, in which the
relevance of a category is projected and negotiated turn-by-turn. This dimension has been
repeatedly emphasised by Schegloff (1987, 1991, 1997), who relates the issue of relevance
to the notion of procedural consequentiality:

the problem of showing from the details of the talk or other conduct in the materials
that we are analysing that those aspects of the scene are what the parties are oriented to.
For that is to show how the parties are embodying for one another the relevancies of the
interaction and are thereby producing the social structure.
(Schegloff 1991: 51)

This issue has generated discussions concerning gender, its relevance in social interaction,
and the way members’ orientations can be demonstrated on the basis of the details of their
talk and embodied actions. This also touches upon the notion of the ‘omni-relevance’ of the
category of gender. Sacks spoke of categories belonging to an ‘omni-relevant device’ (1992:
313) in referring to collections of categories characterising certain settings in which they
are always potentially relevant for organising and interpreting the activities taking place in
these settings. Although Garfinkel (1967: 118) refers to the omni-relevance of gender in

247
Lorenza Mondada

Agnes’ case, in relation to the constant passing work she does in order to accomplish the
naturality of her feminine category, the omni-relevance of the category of gender cannot a
priori be decided in general but has to be analytically demonstrated in each empirical case
(Schegloff 1997). This has generated heated debates, namely about the relation between
CA and feminism (Kitzinger 2000; Kitzinger and Frith 1999; Stokoe and Smithson 2001;
Tennent and Weatherall, this volume; Weatherall 2002a, 2002b; Wilkinson and Kitzinger
2003). Nevertheless, linking the relevance of a category to its sequential consequentiality
does not mean a priori that gender is not omni-relevant. This just has to be demonstrated,
and the demonstration is far from limited to the explicit lexical mention of the category (the
lexical item itself is not always a warrant for the relevance of the category; see Kitzinger
2007), but includes a large variety of ways of ‘doing being’ the category (Kitzinger 2009;
West and Zimmerman 1987).
The use of and orientation to gender categories in social interaction has produced fruitful
analyses of the normative evidence of the heterosexual matrix (Kitzinger 2005b) as well as
of the gendered vision of professional activities (Ekberg and Ekberg 2017). For instance,
Kitzinger (2005a, 2009) showed how the mention of the category ‘my wife’ or ‘my husband’
in an emergency call to a doctor not only provided for the intelligibility of the situation and
engaged in ‘doing heteronormativity’ (Kitzinger 2009: 96), but also impacted the smooth
progressivity of the call and, in some cases, its outcome. For example, requests for the
patient’s address routinely treated the addresses of the caller and the patient as the same,
supposing a common residence for them as a couple. Significantly, when the caller referred
to ‘my baby/wife/husband’, the question about the address was formulated as, ‘Where do
you live?’ When the caller referred to the patient as ‘my friend’, the question was rather,
‘Where does she (or he) live?’ (Kitzinger 2009: 97). Likewise, it was possible for the doctor
to infer from the mention of ‘my baby’ that there might be other incumbents of the family.
This kind of heteronormative inference works ex negativo also when there is no evidence of
a heterosexual family. For instance, when the person calling is not a member of the family,
the reference to the family as a normative framework still holds and the caller has to account
for deviating from it:

Extract 2 (from Kitzinger 2005a: 489)

01 Clr: I've got u:m .h my next door neighbor's baby's
02 not very well.<She keeps losin' 'er breath an' .hh
03 um (.) bringin' up sick and everything an' she
04 keeps cryin'. <She's been cryin' for about four
05 hours,.hhh They don't know what's wrong with 'er.
((15 lines of diagnostic questioning omitted))
21 Doc: Are you the mother.
22 Clr: No, I'm the next door neighbor.h .h
23 Doc: Right.
24 Clr: The mother's lookin' after the baby at the
25 moment. hh .hh
26 Doc: An' she's- I see. Okay doke.

The caller is phoning on behalf of a neighbour, and although she discloses this early on in
the call (1), the doctor does not consider it when he checks a few seconds later if the caller
is the mother (21), thus orienting to the normative standardised pair of ‘baby’ and ‘mother’.

248
 Accomplishing gender in interaction

This generates not only a reminder of the category by the caller (22) but also, after a mini-
mal response of the doctor, not progressing in the call (23), an explicit account about the
whereabouts of the baby’s mother (24–25), orienting to the possible moral implications of
the fact that it is not the mother who calls. The doctor’s final response (26) treats that expla-
nation as providing for the necessary accountability of the circumstances of the call. On the
basis of such cases, Kitzinger (2005a: 495) highlighted the fact that the normative orien-
tation to heterosexuality and family generates ‘interactional trouble’ when it is breached,
causing hitches, repairs, and inserted sequences that make the call more complex and delay
its progressivity. This is particularly striking in cases in which a gay person or a lesbian calls
on behalf of their partner (Land and Kitzinger 2005).
Heterosexual normativity can surface in multiple ways, inclusive when the category of
‘heterosexual’ is explicitly addressed. In a study of survey questions made at the end of calls
to a health information public service, Ostermann (2017) pointed at the difficulty raised by
explicit questions about the heterosexuality of the caller, originally designed by the institu-
tion as a Wh-question, ‘What is your sexual orientation?’, but most frequently transformed
by the person doing the survey into a polar question, ‘Are you heterosexual?’. The yes/
no interrogative is clearly biased towards heterosexuality as being the norm and projects
a preferred positive answer. Nonetheless, the callers’ responses showed a variety of trou-
bles. Whereas a few answers are straightforwardly positive and are not problematised, lead-
ing to closing, the responses most frequently showed delays, hesitations, and in-breaths. In
these cases, even when the caller responded, it occasioned the initiation of a repair by the
call-taker:

Extract 3 (from Ostermann 2017: 357)

1 CT: ce::rto (.) .h e: po- en↓t~ao senhora- (.) a senhora
oka::y (.) .h ah: you so- ↓okay mam- (.) do you
2 se considera heteressexu↑al
consider yourself heterosexu↑al
3 (1.4)
4 CR: n~a:o
no:
5 (0.6)
6 CT: heterossexual senhora e a pessoa que tem afinidade pelo
heterosexual mam is a person who has affinity with the
7 sexo o↑posto >ou seja a mulher ter afinidade
opposite ↑sex. >in other words< the woman who has affinity
8 por um homem<
with a ↑man.
9 CR: a: ent~ao sim (hh) eu n~ao entendi muito bem
oh: then yes (hh) I hadn’t understood it very well
10 (.)
11 CT: oque:i senhora ent~ao o ministerio da saude agra↑dece
oka:y mam then the ministry of health thanks you

The question (1–2) is answered after a delay (3) by the caller with a stretched negative
particle (4). After a gap (5), instead of accepting the answer, the call-taker provides an
explanation of the meaning of ‘heterosexual’, which is first defined using a generic term

249
Lorenza Mondada

(‘pessoa’/’person’) and then exemplified from the perspective of a woman (‘mulher’) as

being attracted by a man, which is recipient oriented towards the gender attributed to the
caller. The explanation is responded to by a change-of-state token (‘a:’/’oh:’ 9) and an
account. This is treated by the call-taker as operating a repair of the initial response and is
followed by a close of the sequence and of the call.
In these calls, the category ‘heterosexual’ is often misunderstood, and most often
rejected; moreover, negative answers are not taken seriously by the call-taker, who often
initiates repair, finally leading to a revision of the initial answer (Ostermann 2017). These
cases reveal that the use of an explicit category for referring to the taken-for-granted norm
generates interactional trouble, showing that naming the norm is not a straightforward mat-
ter (a number of callers actually understood the category as referring to homosexuality).
Descriptions of conduct in institutional settings also show normative orientations to gen-
der categories, which are alluded to and inferred rather than explicitly used. This is the case
with trials and police interrogations, where the choice of words for describing victims and
offenders matters for their intelligibility and morality. For instance, Wowk (1984) showed
how a murderer describing the victim’s behaviour as ‘sexual propositioning’ produced
moral inferences about ‘what kind of girl’ she was, thus shifting blame onto her. Likewise,
Drew (1992) showed how an attorney (A) interrogating a victim of an alleged rape (W) used
descriptions of what happened that were biased towards blameable conduct:

Extract 4 (From Drew 1992: 489/ex. 12)

1 A: it’s where uh (.) uh gi:rls and fella:s meet isn’t it?
2 (0.9)
3 W: people go: there.

Extract 5 (From Drew 1992: 489/ex. 13)

1 A: an’ during that eve:ning: (0.6 s) uh: didn’t Mistuh ((name))
2 come over tuh sit with you
3 (0.8)
4 W: sat at our table.

Drew commented that, although all of the questions were designed to elicit a yes/no answer,
the victim neither confirmed nor disconfirmed, but instead provided for an alternative
description. These descriptions resisted the inferences that characterised the first formula-
tions, which were hinting at portraying the bar where the fact happened as a place for sexual
encounters (extract 12, l.1) and the relation with the offender as established and consented
(extract 13, l.1–2), Instead, the revised version by W provided for a more neutral and less
morally loaded version.
A similar case was evoked by Edwards (1998: 29) showing how gendered inferences
are mobilised in marriage guidance counselling, where the wife explains in front of her
jealous husband what a ‘girls night out’ is, describing it as ‘it’s all married women talking
about our kids’; that is, a non-sexualised event, involving category-bound activities related
to ‘married women’ and ‘family’ (see also Stokoe and Edward 2011 about similar gendered
inferences in complaints about problematic neighbour behaviour).
In sum, categorisation practices are made relevant in specific sequential environments
in doing certain types of actions, and the sequential organisation further confirms their

250
 Accomplishing gender in interaction

establishment or their rejection, as well as their accountability relative to the issues at stake.
Participant orientations to the relevance of a category are visible in the way they produce
a next turn or action that might align or disalign with the previous and display acceptance
or resistance towards the inferences and consequences of the categories used. In this sense,
categorisation analysis fundamentally relies on sequential analysis.

Sequential organisation and the relevance of gender

Categories are sequentially relevant (Schegloff 1991): the details of sequential organisation
manifest convergent or divergent categorical orientations of the participants. CA’s vision of
situated actions in interaction is based on the principle of sequentiality, which refers to the
moment-by-moment stepwise emergent organisation of turns-at-talk and actions through
which a speaker or a doer not only formats their actions, but also locally and reflexively
adjusts to possible incipient responses of the co-participants (Goodwin 1979). In this sense,
‘why that now?’ and ’what next?’ are questions through which participants monitor the pro-
gression of an ongoing action as well as the provision of a response (Schegloff and Sacks,
1973; Schegloff 2007). Reponses display alignment or disalignment, affiliation or disaffili-
ation, and exhibit retrospectively their interpretation of the previous turn. The progressivity
of the encounter can be delayed and stopped at any moment to address troubles of hear-
ing, of understanding, or of acceptability, thanks to the practices of repair and correction
(Schegloff et al. 1977; Schegloff 1992).
Interactional troubles are often related to gender categorisations and descriptions. These
troubles are handled by practices of repair and correction, which are pervasive phenomena
that significantly reveal participant orientations to gender relevance and norms, displayed
in their use and repair of pronouns, lexical items, and categories. Given that a rich array of
CA studies deals with repair as revealing normative orientations to gender (e.g. Land and
Kitzinger 2005; Stokoe 2011; Weatherell 2002b about children’s disputes concerning the
gender of a doll; Weatherall 2015), this section exemplifies a few issues on the basis of this
literature.
Jefferson’s distinction (1983) between ‘embedded’ and ‘exposed’ correction has been
fruitful to show how (hetero)sexist presumptions are handled by co-participants. In the for-
mer, the problematic item simply gets subsequently replaced and therefore minimised, while
in the latter, the correction becomes the main business of the ongoing turn. For instance,
Ekberg and Ekberg (2017) showed how patients sometimes challenged sexist presumptions
of healthcare professionals, such as by correcting a reference to an unknown doctor referred
to as ‘he’ or an unknown nurse referred to as ‘she’. An example of embedded correction is
given here:

Extract 6 (from Ekberg and Ekberg 2017, from ex. 7)

17 CLI: I've been seeing my doctor about them
18 for years and was reluctant to
19 believe that there was no physical
20 cause, but eventually relented.
21 The: So he has suggested you speak to
22 someone to try and get to the bottom
23 of whether there is an anxiety link?
24 Cli: Yes, she thinks I'm suffering from

251
Lorenza Mondada

25 anxiety and depression. I think

26 she's probably right, although I
27 feel the anxiety probably started first...

The client used the category ‘the doctor’ (17) without disclosing her gender, but the thera-
pist supposed that the doctor was a man (21) in the next question. The patient first responded
to the question (‘yes’ 24) and then elaborated on it while providing an embedded correction
of the therapist’s gender presumption, referring to ‘the doctor’ with the pronoun ’she’ (24).
The use of the embedded correction did not disrupt the ongoing action; indeed, the correc-
tion was not responded to by the therapist.
Likewise, Stringer and Hopper (1998) identified the use of ‘pseudo-generics’, routinely
assuming that the referent is male, although they could be female. Some of these default
assumptions are simply ignored; that is, not treated as a source of trouble and not repaired.
Others are corrected, as in the following case:

Extract 7 (abbreviated from Stringer and Hopper, 1998: 214)

1 AVA: Well- what’d HE say
2 BEV: He is a she- and everything’s fine

Ava uses the pronoun ‘he’ to refer to the doctor. Bev’s response orients and first treats this
assumption with an exposed correction (‘he is a she’ 2) and only then answers the question.
Bev still minimises the disruption of the activity, by immediately continuing her turn with
an answer to the prior question. From this, we get a sense of the work participants do in
order to both correct erroneous assumptions and maintain the progressivity of the activity.
Land and Kitzinger (2005), studying phone calls made by lesbians to institutions, showed
that in calls to unknown persons they referred to their partners by routinely selecting gender-
neutral terms (‘partner’, ’spouse’) rather than the gender-specific ones selected by hetero-
sexuals. This often occasions later repairs when the call-taker interprets the ‘spouse’ within
a heteronormative conception of the couple, as in this case:

Extract 8 (Land and Kitzinger, 2005: 396–397)

13 (10.5)
14 Jan: .hhh I’m wanting insurance fo:r uhm: (.) ,
15 two named drivers self and spous:e.=
16 Clt: Yeah< ‘v cou:rse.
17 (13.0)
18 Clt: (Right) I’ve got you down as a doctor. Do
19 you have the use of any other vehicle
20 within the househo:ld.
21 Jan: Yes I do.
22 (0.8)
23 Clt: An: (.) you said you’d like to insure
24 your husband to drive the car.
25 Jan: mcht Uh:::m It’s not my husband it’s my
26 wi:fe and yes I would l[ike t- ]
27 Clt: [Oh I do] beg your pardon.
28 Jan: I would like to insure her.

252
 Accomplishing gender in interaction

29 Clt: Yep >thank you<

30 (11.5)
31 Clt: (’Kay) Could I take your wife’s name please.

Janice had requested an insurance for ‘self and spouse’ (15), which is a formulation that
does not give any explicit information about the gender of the spouse. Later on, the sales-
person referred to ’your husband’ (24) in a quotative turn referring back to what Janice
just said, interpreting and transforming it, assuming Janice’s spouse was male and presum-
ing her heterosexuality. Janice proceeded to an exposed correction (25–26), with the effect
that the negotiation of the insurance – for which the salesperson turn was a preliminary
– was momentarily suspended: the sequence was now devoted to the correction. Janice’s
turn beginning (‘mcht Uh:::m’ 25) displayed that a problem was surfacing. She initiated
repair by naming the repairable (‘it’s not my husband’ 25) and then correcting the category
(‘it’s my wife’ 25–26). Then she immediately moved on with the response to the previous
question (‘and yes I would like’ 26), with an and-prefaced continuation (see the same prac-
tice in extract 7, l. 2). So even if Janice operated an exposed correction, she minimised the
sequential consequences by resuming and progressing in the activity. But the salesperson in
overlap responded to the correction – possibly treating it as complainable – and apologised
(27). We notice that again Janice did not respond to the apology and came back to business.
The salesperson displayed realignment by using the category ‘your wife’ in the subsequent
management of the request (31).
Not only other-corrections, but also self-repairs and corrections, are significant dis-
plays of normative orientations to gender, revealing ongoing possible inferential processes
to which the speakers orient. For example, Stokoe (2011) discussed ‘marked self-repairs’
that showed how the speaker realised, as the production of their turn was in progress, that
improper inferences were possibly produced by the initial choice of a category (‘girl’), occa-
sioning a self-repair into another category (‘woman’), accompanied by an apology (‘sorry’).
This exhibits the speaker becoming locally ‘gender aware’ in the course of his action, orient-
ing to the possible offensive character of ‘girl’ vs. ‘woman’ (Stokoe 2011: 99). Interestingly,
if this is the case of marked self-repairs (as in ‘you didn’t– (0.2) y’didn’t admit. to this girl
then. (.) °that-° this woman >sorry< that uh:: you punched ’er door.’), alternative formula-
tions (‘girl or woman’) display lower commitment and minimal orientation to the difference
between the two categories (Stokoe 2011: 105).
Describing the emergent process through which ‘gender creeps into talk’, Hopper and
LeBaron (1998) also showed how speakers might suddenly notice the relevance of gender
and focus on it, as in the following example:

Extract 9 (from Hopper and LeBaron, 1998: 69)

1 Jill: I’ve signed up for one of those informal classes about
2 car maintenance and repair.
3 Pip: That’s a good idea. A lot of women can really learn a
4 lot from these classes
5 ((short pause))
6 Pip: Well, I guess there’s a lot of guys who can learn from
7 ’em too.

As Hopper and LeBaron noticed, Jill’s announcement contained a possible ‘implicit index-
ing of gender’ (1998: 69). Pip’s response established a categorical link between the kind of
253
Lorenza Mondada

class mentioned by Jill and a gender category (‘women’ 3). But the pause in which Jill did
not respond and did not align with it occasioned Pip’s reselection, in which she revised her
previous stereotypical gender association (6–7).
A final instance of self-repair is observable in the next extract:

Extract 10 (from Ekberg and Ekberg 2017, ex.8)

01 Aud: I said the ↑only other thing I
02 wanted to do:: was <write a letter>
03 to the gee pee:,
04 Cli: Yeah
05 Aud: just for him to sign >him or her< to sign,
06 Cli: Yeah

As commented on by Ekberg and Ekberg (2017), the audiologist first referred to the client’s
doctor with the gender-neutral ‘GP’ (3) and then continued with a pronoun. He used the
masculine form first, continuing to progress into his turn, but then quickly suspended the
turn and inserted a retrospective self-repair (‘>him or her<’ 5) before resuming it. The client
here minimally acknowledged the turn (6).

With this repair, the audiologist retrospectively displays that she is unaware of the gen-
der of the client’s GP. In addition, she displays recognition of the gender presumption
she has just made, and an intention to correct it. This conduct establishes the audiologist
as someone who is sufficiently “gender aware” (Stokoe 2011) to realise that she has
made a presumption and is able to correct it.
(Ekberg and Ekberg 2017: 11)

Repairs and corrections are practices that have been largely studied in CA approaches of
gender because they index in exemplary ways how participants orient to categories and
manifest normative gender assumptions. The way in which these practices are formatted
and sequentially positioned show how speakers not only minimise or exhibit gender inap-
propriateness, but also how they embed these concerns in the ongoing activity in ways that
occasion disruptures and suspensions or, quite the opposite, minimise them. This, in turn,
maximises or limits the possibility for the co-participant to respond, in some cases minimis-
ing the slot for doing so. The detailed organisation of sequentiality shows here again the
subtle and differentiated orientations that participants can display to gender issues.

Conclusion
The chapter has reviewed the original contributions EMCA offered to gender, insisting on
the lessons gathered in the field from the perspective on action and interaction as a continu-
ous practical accomplishment, on social interaction as making locally relevant specific cat-
egorisation devices, and on interaction as sequentially organised in detailed ways.
Ethnomethodology and conversation analysis have introduced a perspective in social
sciences, anthropology, communication, and linguistics that is radically focused on the situ-
ated activities of the participants. This emphasis on the centrality of action, on its detailed
organisation, its systematicity as well as its situatedness, provides for a specific perspective
on gender. It highlights how gender creeps into the smaller details of talk and how this

254
 Accomplishing gender in interaction

is accountable and recognisable by the way participants orient and respond to it. It also
highlights the options co-participants have to respond to these relevancies, assumptions,
categories, and to do so in minimal or exhibited ways, aligning or disaligning, affiliating
or disaffiliating with the previous action and its presuppositions. This shows how social
order in general, and gender normativity in particular, are both reproduced and challenged
or reinvented in a diversity of settings, being constantly built, reproduced, subverted, and
transformed through micro-sequential accomplishments.

Conventions
The transcription conventions adopted here are those provided by Gail Jefferson for talk
(2004).

References
Bucholtz, M. and Hall, K. (2004) ‘Identity and interaction: a sociocultural linguistic approach’,
Discourse Studies, 7(4–5), pp. 585–614.
Bulter, C. and Weatherall, A. (2011) ‘”This is Charlotte”: a case of passing in children’s talk-in-
interaction’, in Speer, S. and Stokoe, E. (eds) Gender and conversation. Cambridge: Cambridge
University Press, pp. 231–249.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Drew, P. (1992) ‘Contested evidence in a courtroom cross-examination: the case of a trial for rape’,
in Drew, P. and Heritage, J. (eds) Talk at work: interaction in institutional settings. Cambridge:
Cambridge University Press, pp. 470–520.
Edwards, D. (1998) ‘The relevant thing about her: social identity categories in use’, in Antaki, C. and
Widdicombe, S. (eds) Identities in talk. London: Sage, pp. 15–33.
Ekberg, K. and Ekberg, S. (2017) ‘Gendering occupations: persistence and resistance of gender
presumptions about members of particular healthcare professions’, Gender and Language, 11(1),
pp. 100–120.
Garcia, A. and Fisher, L. M. (2011) ‘Being there for the children: the collaborative construction of
gender inequality in divorce mediation’, in Speer, S. A. and Stokoe, E. (eds.) Conversation and
gender. Cambridge: Cambridge University Press, pp. 272–293.
Garfinkel, H. (1967) Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Garfinkel, H. (2002) Ethnomethodology’s program. working out Durkheim’s aphorism. Lanham:
Rowman and Littlefield.
Goodwin, C. (1979) ‘The interactive construction of a sentence in natural conversation’, in Psathas,
G. (ed) Everyday language: studies in ethnomethodology. New York: Irvington Publishers, pp.
97–121.
Goodwin, M. H. (1990) He-said-she-said: talk as social organization among black children.
Bloomington: Indiana University Press.
Goodwin, M. H. (2006) The hidden life of girls: games of stance, status, and exclusion. Malden, MA:
Blackwell.
Heritage, J. (1984) Garfinkel and ethnomethodology. Cambridge: Polity Press.
Hirschauer, S. (1991) ‘The manufacture of bodies in surgery’, Social Studies of Science, 21(2), pp.
279–319.
Hopper, R. and LeBaron, C. (1998) ‘How gender creeps into talk’, Research on Language and Social
Interaction, 31(1), pp. 59–74.
Jefferson, G. (1983) ‘On exposed and embedded correction in conversation’, Studium Linguistik, 14,
pp. 58–68.
Jefferson, G. (2004) ‘Glossary of transcript symbols with an introduction’, in Lerner, G. H. (ed)
Conversation analysis: studies from the first generation. Amsterdam: Benjamins, pp. 13–31.

255
Lorenza Mondada

Kessler, S. (1998) Lessons from the intersexed. New Brunswick: Rutgers University Press.
Kessler, S. and McKenna, W. (1978) Gender: an ethnomethdological approach. New York: Wiley.
Kitzinger, C. (2000) ‘Doing feminist conversation analysis’, Feminism and Psychology, 10(2), pp.
163–193.
Kitzinger, C. (2005a) ‘Heteronormativity in action: reproducing the heterosexual nuclear family in
after-hours medical calls’, Social Problems, 52(4), pp. 477–498.
Kitzinger, C. (2005b) ‘"Speaking as a Heterosexual”: (How) does sexuality matter for talk-in-
interaction?’, Research on Language and Social Interaction, 38(3), pp. 221–265.
Kitzinger, C. (2007) ‘Is “woman” always relevantly gendered?’, Gender and Language, 1(1), pp.
9–49.
Kitzinger, C. (2009) ‘Doing gender: a conversation analytic perspective’, Gender and Society, 23(1),
pp. 94–98.
Kitzinger, C. and Frith, H. (1999) ‘“Just say no?” the use of conversation analysis in developing a
feminist perspective on sexual refusal’, Discourse and Society, 10(3), pp. 293–316.
Land, V. and Kitzinger, C. (2005) ‘Speaking as a lesbian: correcting the heterosexist presumption’,
Research on Language and Social Interaction, 38(4), pp. 371–416.
Lindström, A., Näslund, S., and Rubertsson, C. (2015) ‘The interactional organisation of sex
assignment after childbirth’, Gender and Language, 9(2), pp. 189–222.
McIlvenny, P. (2002a) ‘Critical reflections on performativity and the “un/doing” of gender and
sexuality in talk’, in McIlvenny, P. (ed.) Talking gender and sexuality. Amsterdam: John Benjamins,
pp. 110–149.
McIlvenny, P. (ed.) (2002b) Talking gender and sexuality. Amsterdam: Benjamins.
Mondada, L. (2002) ‘Pratiques de transcription et effets de catégorisation’ [Practices of transcription
and categorization effects], Cahiers de Praxématique, 39, pp. 45–75.
Mondada, L. and Oloff, F. (2015) ‘“Hetero oder homo?”: la performance d’un coming out hétérosexuel
à la radio’, Langage et Société, 152(2), pp. 17–40.
Ostermann, A. C. (2017) ‘“You are heterosexual”: whose language? ‘No mam. You are heterosexual’:
whose language? Whose sexuality?’, Journal of Sociolinguistics, 21(3), pp. 348–370.
Sacks, H. (1963) ‘Sociological description’, Berkeley Journal of Sociology, 8, pp. 1–16.
Sacks, H. (1966) The search for help: no one to turn to. PhD. University of California, Berkeley.
Sacks, H. (1972) ‘An initial investigation of the usability of conversational data for doing sociology’,
in Sudnow, D. (ed) Studies in social interaction. New York: Free Press, pp. 31–74.
Sacks, H. (1992) Lectures on conversation [1964–72], 2 vols. Oxford: Basil Blackwell.
Sacks, H., Schegloff, E. A., and Jefferson, G. (1974) ‘A simplest systematics for the organization of
turn-taking for conversation’, Language, 50(4), pp. 696–735.
Schegloff, E. A. (1987) ‘Between micro and macro: contexts and other connections’, in Alexander,
J., Giesen, B., Munch, R., and Smelser, N. (eds) The micro-macro link. Berkeley: University of
California Press, pp. 207–234.
Schegloff, E. A. (1991) ‘Reflections on talk and social structure’, in Boden, D. and Zimmerman, D.
(eds) Talk and social structure. Cambridge: Polity, pp. 44–70.
Schegloff, E. A. (1992) ‘Repair after next turn: the last structurally provided defense of intersubjectivity
in conversation’, American Journal of Sociology, 95(5), pp. 1295–1345.
Schegloff, E. A. (1997) ‘Whose text? Whose context?’, Discourse and Society, 8(2), pp. 165–187.
Schegloff, E. A. (2007) Sequence organization in interaction: a primer in conversation analysis.
Cambridge: Cambridge University Press.
Schegloff, E. A., Jefferson, G., and Sacks, H. (1977) ‘The preference for self-correction in the
organization of repair in conversation’, Language, 53(2), pp. 361–382.
Schegloff, E. A. and Sacks, H. (1973) ‘Opening up closings’, Semiotica, 8(4), pp. 289–327.
Speer, S. A. (2005) Gender talk: feminism, discourse and conversation analysis. London: Routledge.
Speer, S. A. (2009) ‘Passing as a transsexual woman in the gender identity clinic’, in Wetherell, M.
(ed) Theorizing identities and social action. Basingstoke: Palgrave Macmillan.

256
 Accomplishing gender in interaction

Speer, S. A. and Stokoe, E. (eds) (2011) Conversation and gender. Cambridge: Cambridge University
Press.
Stokoe, E. (2006) ‘On ethnomethodology, feminism, and the analysis of categorial reference to gender
in talk-in-interaction’, The Sociological Review, 54(3), pp. 467–494.
Stokoe, E. (2011) ‘“Girl - woman - sorry!”: on the repair and non-repair of consecutive gender
categories’, in Speer, S. A. and Stokoe, E. (eds) Conversation and gender. Cambridge: Cambridge
University Press, pp. 85–111.
Stokoe, E. and Edward, D. (2011) ‘Mundane morality and gender in familial neighbour disputes’, in
Cromdal, J. and Tholander, M. (eds) Morality in practice: exploring childhood, parenthood and
schooling in everyday life. London: Equinox, pp. 165–192.
Stokoe, E. and Smithson, J. (2001) ‘Making gender relevant: conversation analysis and gender
categories in interaction’, Discourse and Society, 12(2), pp. 217–244.
Stokoe, E. and Weatherall, A. (ed) (2002) ‘Gender, language, conversation analysis and feminism’,
Discourse and Society, 13(6), pp. 707–713.
Stringer, J. L. and Hopper, R. (1998) ‘Generic he in conversation?’, Quarterly Journal of Speech,
84(2), pp. 209–221.
Weatherall, A. (2002a) Gender, language and discourse. London: Routledge.
Weatherall, A. (2002b) ‘Towards understanding gender and talk-in-interaction’, Discourse and
Society, 13(6), pp. 767–781.
Weatherall, A. (2015) ‘Sexism in language and talk-in-interaction’, Journal of Language and Social
Psychology, 34(4), pp. 1–19.
West, C. and Fenstermaker, S. (1995) ‘Doing difference’, Gender and Society, 9(1), pp. 8–37.
West, C. and Zimmerman, D. H. (1987) ‘Doing gender’, Gender and Society, 1(2), pp. 125–151.
Wilkinson, S. and Kitzinger, C. (2003) ‘Constructing identities: a feminist conversation analytic
approach to positioning in action’, in Harré, R. and Moghaddam, F. (eds) The self and others.
Westport, CT: Praeger, pp. 157–180.
Wowk, M. T. (1984) ‘Blame allocation, sex and gender in a murder interrogation’, Women’s Studies
International Forum, 7(1), pp. 75–82.

257
 17
Feminist conversation analysis
Examining violence against women
Emma Tennent and Ann Weatherall

Introduction
Conversation analysis is a rigorous methodology for feminist analyses of gender, language,
and sexuality. Its focus on the accomplishment of actions in naturally occurring social inter-
action is different from other approaches to gender and language. Aspects of a conversation
analytic mentality have been critiqued by both feminists and language scholars for con-
straining the study of gender and communication. Nevertheless, we show that a sustained
focus on the minutiae of social interaction and close attention to what people say and how
they say it in the service of action can further advance knowledge about key feminist con-
cerns including gender, power, identity, and oppression.
Conversation analysis influenced some of the earliest research on gender and language.
For example, by analysing patterns of turn-taking in couples’ talk, Zimmerman and West
(1975) reported that men interrupted women more frequently than the inverse. However,
‘feminist conversation analysis’ was only coined at the turn of the century by Kitzinger
(2000), who promoted it for feminist research. For some critics, conversation analysis’
objective empirical approach is ill-fitted for investigations with a political, feminist stance
(Speer 1999; Whelan 2012). Yet using conversation analysis need not preclude political
views. We share Kitzinger’s position on the value of bringing feminism and conversation
analysis together. It is our view that feminist conversation analysis offers a productive meth-
odology for the field of gender and language research.
The research we present in this chapter demonstrates a conversation analytic approach
to the study of violence against women. As feminist scholars, we are concerned with the
gendered meanings of victimhood, and the difficulties women face in disclosing vio-
lence and seeking support from institutions. As conversation analysts, we examine these
questions using recorded data from real-life interactions where women seek help from
a victim support helpline. A focus on the details of interaction can yield novel insights
into how the meanings of victimhood are negotiated when the relevant actions are callers
seeking help and call-takers delivering services. Empirical findings can then be applied
to improve services, thereby making a practical difference for women seeking support
from violence.

258
 Feminist conversation analysis

Conversation analysis and participants’ orientations

Conversation analysis is an empirical, observational method that examines how social life is
organised to produce shared meaning and accomplish coordinated action. Despite its name,
conversation analysis examines more than just talk. Data are recordings and transcripts of
‘talk-in-interaction’ which includes what people do and say, movements of the body, and
interaction with the environment. A foundational assumption is that people are competent
‘members of a culture’ (that is, parties or participants) who go about their daily lives with an
understanding of how the social world works. The aim of conversation analysis is to explain
how members build and recognise actions that result in a remarkable coordination in the
accomplishment of everyday activities. To that end, analytic observations are grounded in
the detail of actual social interactions, with a focus on what participants do and how they do
it. As feminist scholars using conversation analysis, one key interest is in how sex, gender,
and identity are accomplished through the practices of everyday life.
Conversation analysis is concerned with people’s own understandings of what they and
others are doing in social interaction. In technical terms, this is referred to as ‘participants’
orientations’. What participants treat as relevant is precisely (in fact, only) what conversa-
tion analysts attend to when analysing the interaction. For example, whether talking at the
same time as someone else is an interruption is not for analysts to decide, but something par-
ticipants display – on the spot – as part of the ongoing interaction (Weatherall and Edmonds
2018).
One way to determine participants’ orientations is through the ‘next turn proof proce-
dure’. This procedure is based on the principle that the way someone responds in a next
turn of talk displays their understanding of what was happening in the previous turn. In the
extracts presented below, we show how call-takers’ turns display an understanding of call-
ers’ turns which preceded them. We also present a case where the next-turn proof procedure
is not available as an analytic tool (when the recipient does not take a next turn of talk).
Nonetheless, we show how callers’ actions are built in ways that demonstrate their orienta-
tions to how their turns will be understood and responded to by others.
Focusing on participants’ own words and what they are demonstrably doing is one way
to conduct feminist research that examines people’s lived experiences in their own terms
(Kitzinger 2000), but the analytic focus of participants’ orientations has been controver-
sial for feminist conversation analysis. One critique is that examining participants’ orien-
tations cannot speak to the sociopolitical, cultural, and historical contexts of gender and
power (Whelan 2012). However, Schegloff (1997: 167) argued that analysts who treat their
concerns as more important than participants’ are engaged in ‘theoretical imperialism’. He
was responding directly to feminist language research that presupposed the relevance of
gender or sexuality when analysing speakers’ ways of talking and behaving. Just because
participants can be categorised as men or women doesn’t mean these are legitimate analytic
categories to make sense of how they act. The very same participants could just as easily
be categorised based on their age, occupations, food preferences, hair styles, and so on.
Schegloff argued that it is not legitimate to pick out gender identities from this list, just
because the analyst happens to be interested in gender. Instead, conversation analysts exam-
ine the identities that participants treat as important, and demonstrate how these identities
are used by the participants to make sense of the interaction they’re involved in.
Examining participants’ orientations does not, as some critics might have it, limit analy-
sis to apolitical descriptions of technical phenomena. Nor does a political feminist analy-
sis necessarily compromise conversation analysis’ theoretical roots (see Wowk 2007 for a

259
Emma Tennent & Ann Weatherall

critique). Instead, feminist conversation analysis can provide a disciplined approach for
investigating how sociocultural matters such as gender, sexuality, identity, and power are
visibly relevant as participants build and interpret their social worlds and progress daily
activities. A feminist interest in gender and power can be valid if analysts do not assume
how or where this will become manifest in the data, but instead undertake to demonstrate
that and how their feminist interests are noticed and used by participants (Speer 2012). The
following section presents some illustrative examples of the emerging field of work at the
intersection of feminism and conversation analysis.

A brief review of feminist conversation analytic research

A conversation analytic approach has been used productively by feminist scholars to exam-
ine a variety of issues. Below we present three strands of research that utilise conversation
analytic methodology to examine feminist concerns. First, feminist conversation analysis
has demonstrated how gender and sexuality operate as normative, ordinary parts of social
life and how everyday oppression works by sanctioning certain identities and behaviours
as non-normative. Second, conversation analysis can be used to examine participants’ ori-
entations to common-sense cultural knowledge about the gendered social and moral world.
Finally, grounded insights about the workings of the social word have been used to consider
effective interventions for social change, and develop services to provide practical help for
people in need.

Analysing everyday oppression

A key aim of conversation analysis is to understand the methods members use in ‘being an
ordinary person’ (Sacks 1984: 415); in other words, how people as culturally competent
members of society go about their everyday lives accomplishing actions with others and
interacting with the physical and social world. Conversation analysts examine the resources
used by participants to build and interpret in/equality in diverse settings. For example, ana-
lysts have documented how turn-taking is organised in large political meetings to support
democratic participation (Mondada 2012).
Building on foundational ethnomethodological approaches to gender (e.g. Kessler and
McKenna 1978), feminist conversation analysis has demonstrated how gender and sexual-
ity are part and parcel of common-sense knowledge. Certain ways of being in the social
world are treated as ordinary and unremarkable, while others can be subject to scrutiny. For
example, from a young age, children use the notion of gender differences to assess whether
others’ behaviours are acceptable or not (Weatherall 2002).
One notable programme of research has demonstrated how heterosexuality manifests in
interaction as ordinary and unremarkable, while homosexuality is treated as non-normative.
For example, a heterosexual person may mention in passing a ‘wife’ or a ‘husband’, making
their sexuality available to listeners while engaged in another activity entirely (Kitzinger
2005). When lesbian women do the same thing, their recipients treat them as ‘displaying’
their sexuality and may apologise for assuming they were heterosexual (Land and Kitzinger
2005). It is through these everyday practices, Kitzinger argued, that oppression is built,
maintained, or potentially challenged.
Conversation analysis can document the linguistic practices that underlie structures of
inequality, and how these can be challenged or undermined. One way to observe challenges

260
 Feminist conversation analysis

to widespread understandings about gender and sexuality is through the technical practice
of repair, where participants correct their own or others’ talk. Simply shifting pronouns can
work to undermine sexist presumptions, such as doctors being male (Ekberg and Ekberg
2017). Speakers can interrupt their own turn or others’ to change an exclusive masculine ref-
erence term, ‘he’, to more gender-inclusive references (Weatherall 2015). Correcting sexist
assumptions or orienting to gender-exclusive language are members’ practices for challeng-
ing what counts as normal.

Culture and the gendered social-moral order

Feminist conversation analysis can study the gendered social-moral order by demonstrat-
ing how participants understand sex, gender, and sexuality in diverse but ordinary set-
tings. Examples include the way a cross-gender identity can be accomplished by children
when playing during school break-time (Butler and Weatherall 2011) and identifying the
interactional organisation of sex categorisation of a child soon after they have been born
(Lindström, Näslund, and Rubertsson 2015).
Analytically, gender as a social construction can be studied through participants’ actual
use of description and categorisation in talk, two practices linked to common-sense knowl-
edge and moral beliefs. Any person, object, or event can be described and categorised in
an almost infinite number of ways. This means that any actual description or categorisation
will be selective. Category terms are a store of common-sense knowledge (Schegloff 2007),
and a way participants make moral judgements, of which gendered expectations can play
a large part (Stokoe 2006, 2010). Examining how participants describe people and events,
and the categories they choose, effectively provides a window into the working of gendered
culture in action.
The link between categories, activities, and inferences is a key way participants man-
age gender and morality in practice. The same people can be categorised in different ways,
depending on which aspects the speaker highlights or downplays. For example, a husband’s
description of his wife’s ‘girls’ night out’ invoked the activities of heavy drinking and flirt-
ing with men, while she described the same people as ‘married women’ who merely talked
about their children in order to defend against his accusation (Edwards 1998). Certain activ-
ities are culturally understood as ‘proper’ for some category members, but not for others.
For example, complaining neighbours described ‘improper’ activities for a mother as stay-
ing out late and leaving children alone, invoking normative gender expectations and moral
judgements (Stokoe and Edwards 2012). Perpetrators of domestic violence used the same
category-based moral logic, such as describing partners as unfaithful or untidy to implic-
itly justify their violence (Le Couteur and Oxlad 2011). Even without naming a category
directly, participants can use the link between activities and categories to let others infer
moral judgements. For example, a man confessing to murder described how his victim had
propositioned him for sex, letting his listeners make judgements about ‘what kind of girl’
she was based on the unspoken category of ‘prostitute’ (Wowk 1984).

Interventions and social change

According to Wilkinson and Kitzinger (2014: 154), an ‘essential prerequisite’ for feminists
advocating social change ‘is an accurate understanding of how the world is now’, and con-
versation analysis provides an empirical basis of knowledge about the social world and how

261
Emma Tennent & Ann Weatherall

participants understand it. A grounded focus on interaction can avoid many of the pitfalls
of other language-based intervention programmes. For example, Kitzinger and Frith (1999)
critiqued the feminist consent slogan ‘just say no’ as based on a mistaken belief about how
talk works. Conversation analysis has shown that refusals overwhelmingly occur without
the word ‘no’ at all, and regularly include features such as accounts, delays, and mitiga-
tions. Although well-intentioned, date-rape prevention programmes that advise women not
to hesitate, make excuses, or give explanations require women to violate culturally shared
norms for doing refusals, making these campaigns arguably misguided.
Support services can also be improved through a detailed analysis of how they operate by
examining audio, or if possible, video recordings of what they do in practice. Concepts like
‘empathy’ and ‘empowerment’ that are often central to feminist support services can be better
understood by grounding what they mean in actual practices. For example, research on the UK
Birth Crisis Helpline has documented how feminist principles of women-to-women support,
advocacy, and empathy are accomplished through talk-in-interaction. Well-established find-
ings show that finishing someone else’s turn of talk can be used to show shared understand-
ing (Wilkinson and Kitzinger 2014) and assessments of a caller’s situation can demonstrate
empathy (Kitzinger and Kitzinger 2007). In our own research, we intend to adapt our research
findings into evidence-based training to help the organisation better respond to the needs of
women who have experienced violence. A turn-by-turn analysis of how women ask for help
in this context not only contributes to scholarly examination of the difficulty of disclosing
violence (see Jordan 2004) but has practical applications. Training programmes such as the
conversation analytic role-play method (CARM) have been developed to translate analytic
findings to improve service-delivery across institutions (Stokoe 2014). In our own work we
assume a sound first step to develop processes that effectively meet the needs of victims of
violence is to examine what actually happens when they attempt to secure support.

Using conversation analysis to study violence against women

Violence against women is an issue of long-standing feminist concern, and has been
approached from a number of academic and activist perspectives. Using a conversation ana-
lytic approach, we examine real-life interactions in which women turn to a victim support
agency for help. Our analysis of calls to this service reveals how women disclose violence
and negotiate the meanings of victimhood, while shedding light on the actual practices
involved in women’s engagement with support services and the justice system. With a
focus on participants’ descriptions and categorisations, we demonstrate how they invoke
common-sense knowledge about victims of violence. Our findings broadly align with other
feminist research that points to a set of cultural beliefs about violence which structure and
maintain a gendered social-moral order that functions to disadvantage women.
Our data are calls to a New Zealand community organisation which offers practical
advice and emotional support to victims of crime and trauma. A call-centre is the first point
of contact and operates 24 hours a day, 7 days a week. Call-takers connect callers to sup-
port workers or transfer them to alternative agencies. Calls are routinely recorded for train-
ing purposes, and the organisation amended their pre-recorded message and online privacy
statement for us to inform callers that recordings could be used for university research pur-
poses. To ensure confidentiality, identifying information such as names and addresses have
been edited from the sound-files and replaced with pseudonyms on the transcripts.
The organisation initially provided us with a sample of 48 calls, and we subsequently
collected a purposive sample of all calls across a seven-day period. In total, our sample

262
 Feminist conversation analysis

consists of 396 calls and over 23 hours of recordings. We listened to each call and cata-
logued them with a unique identifying number, brief summary, and keywords. Calls were
indexed based on details like length, gender of caller, reason for the call, and call outcome
to create a searchable database. Calls were transcribed following conversation analytic con-
ventions that capture the details of interaction such as silences, overlapping talk, intonation,
and speed (Hepburn 2004; Jefferson 2004). A transcription key is provided at the end of the
chapter.
Listening to recordings and examining transcripts, we familiarised ourselves with the
data in a process called unmotivated looking. Rather than beginning with a hypothesis,
searching for a specific phenomenon, or presupposing gender underpins what people are
saying, we approached the data with an open – albeit feminist – stance. Although we were
interested in women and victimhood, we did not presuppose its relevance for callers seek-
ing help.
An initial observation was that callers contacted the service for many different reasons
and presented their problems in different ways. From this observation, we set out to collect
instances where callers identified themselves and presented their problems. This can occur
in many places, but in the cases below we focus on the opening moments of the call. We
were interested in the different ways callers presented themselves as victims of gendered
violence, and so narrowed our focus to cases regarding sexual or domestic violence. The
organisation is a generic victim support service, and so only 57 calls concern sexual or
domestic violence, which formed our subset for analysis. For each case, we examined how
caller and call-taker reached a joint understanding of caller as a victim of violence. We
found that callers regularly used description and categorisation to identify themselves and
present their problem, and that common-sense cultural knowledge was a resource for call-
takers to make inferences about callers and their circumstances. Below we present three
cases that show different ways callers described themselves and their problems.
The opening of a call is an important moment where participants identify themselves and
establish the reason for the call. It is also the first opportunity for callers to describe their
experiences and ask for help. Extract 1 shows the routine opening contact service workers
(CONTCT) are trained to use,1 and the way the caller responds.

Extract 1
01 CONTCT: kia ora victim support this is Claire
02 (1.4)
03 CALLER: .hh oh hi um my name’s Leanne Alweather .hh um (0.2)
04 I: had s- (0.6) dealt with you guys befo:re um with
05 problems at ho:me .hh and um (0.8) the: lady that
06 I did deal with and I can’t remember her name she did
07 put me on to .hh a (.) um lawyer? .hh (0.8) and (0.8)
08 yeah I was just w- (0.2) wanting to find out inf- (0.2)
09 the lawyer’s name again hh
10 (0.8)
11 CONTCT: o:h okay (0.8) o:h (0.6) what- what’s your name please?

The caller first provides her name (a pseudonym) and then categorises herself as a client
of the service. She does this by describing that she has ‘dealt with you guys befo:re’ (line
4), which identifies herself as someone who has previously used victim support services.

263
Emma Tennent & Ann Weatherall

The caller also provides the reason for her prior contact with the description of ‘problems
at ho:me’ (line 5). Already, just a few moments into the interaction, the caller’s description
of herself as a client who has problems at home provides resources for the contact service
worker to understand her circumstances. In the context of this victim support service, some-
one who is a client is a victim of either crime or trauma. This means the contact service
worker can understand ‘problems at home’ as a description of victimisation. There are a
range of possible crimes and trauma that could occur at home, such as burglary, fire, or
death of a loved one, but the caller’s word selection ‘problems’ could suggest an ongoing
situation.
The contact service worker asks for the caller’s name (line 11) to retrieve her casefile
from the database. This next turn ratifies the caller’s identity as a client of the service. The
contact service worker thus displays an understanding that the caller’s ‘problems’ constitute
victimisation. Common-sense cultural knowledge about the kind of problems that occur at
home, coupled with the caller’s request for a lawyer, makes inferences available about the
nature of the caller’s problem. These inferences are confirmed later in the call, when the
caller explains that the call may be possibly cut off (Extract 1a).

Extract 1a
21 CONTCT: °okay thank [ you° ]
22 CALLER: [and if ] I if I hang up on ya it’s cos my
23 husband’s come home .HH HHUH .hhh
24 CONTCT: °oh(h) right°

The category term ‘husband’ (line 23) displays that the caller is in a married relationship,
and invokes her paired identity as a wife (Sacks 1972). Category terms act as a store of
cultural common-sense knowledge, in this case allowing the contact service worker to
make inferences about the caller and her circumstances. Both descriptions (of ‘problems’
and the consequences of her husband’s return) are linked to the location ‘home’ (line 5;
23). Note also the breathiness (marked by capital ‘h’ line at 23) that represents the call-
ers’ loud exhale of breath, a nonverbal way of displaying emotional stance (Potter and
Hepburn 2010).
Common-sense knowledge about married couples is that they live together in a shared
family home. Generally, home is understood as a place ‘where one belongs, whose presence
there is not accountable’ (Schegloff 1972: 98). Yet for this caller, home is where her prob-
lems (i.e. victimisation) occur, and she does account for her husband coming home and the
consequences that would entail. The two activities (a husband returning home and a wife
hanging up the phone) are not conventionally linked together, but if the categories are mor-
ally qualified (Cuff 1993), for example into ‘abusive husband’ and ‘victimised wife’, then
the caller’s description makes sense. Although she never says the words ‘victim’ or ‘vio-
lence’ directly, the categories she uses and the activities she describes allows the inference
that she is a victim of domestic violence perpetrated by her husband. By responding with
‘oh right’ (line 24) in the next turn, the contact service worker displays an understanding of
this as an account (rather than asking why or querying the caller’s explanation).
In the following extract, another caller describes her experiences in quite a different
way. In contrast to Extract 1 and 1a, this caller refers to herself using the category ‘victim’.
Nevertheless, her first description of how she came to contact the service also provides
inferences about her identity and the nature of her experiences.

264
 Feminist conversation analysis

Extract 2
01 CONTCT: kia ora victim support Molly speaking
02 (0.6)
03 CALLER: .hh hey Molly: um: it’s Mara here I’m just
04 ringing up um (0.6) hey: u:h look I’ve just got
05 one of these um pamphlets from the: Taura drive
06 police station?
07 CONTCT: mhm?=
08 CALLER: =Ngawhare? ((place name))
09 (1.0)
10 CALLER: and um I was just reading <through it> u:m (0.8)
11 y’know for victim information,
12 (0.2)
13 CALLER: if um: (0.4) if I feel that (.) myself an:d (1.0)
14 my son are victims of psychological abuse o:r
15 (0.6) some sort of you know (0.4) um (1.0) >how do
16 you say it< um (1.0) u-uh::h: abuse (.) (of/or) (.)
17 like THREAt,
18 (0.4)
19 CONTCT: mhm
20 CALLER: you know like um (1.4) you know like I’m being
21 threatened that (0.2) you know like (0.6) I’m not
22 allowed to: (0.8) leave with my so:n from the address?

Like the caller in Extract 1, this caller returns the contact service worker’s greeting and
introduces herself by name. She begins to describe what it is she is ‘just ringing up’ for
(lines 3–4), but abandons this turn-in-progress to instead describe how she found out about
the service. In contrast to the caller in Extract 1 who described her prior dealings with the
organisation to identify herself as a client, the caller here presents herself as a first-time
service-seeker by describing how she found out about the service.
Yet even in her description of how she came to call the organisation, the caller presents
herself as a potential client, which in this case means a victim. She refers to the place she
found a pamphlet about victim support as the ‘police station’ (line 6). Like ‘home’ in Extract
1, this description of place invokes inferences that the contact service worker can use to
make sense of the caller’s identity and the nature of her problem. Police stations are places
associated with certain activities (e.g. reporting a crime) and certain categories of people
such as police officers, suspects, witnesses, and victims (Dingemanse, Rossi, and Floyd
2017; Schegloff 1972). That the caller is looking for ‘victim information’ (line 11) implies
she is a victim, and that she is doing so in such a place suggests police involvement, lending
her the legitimacy of being recognised by a criminal justice institution.
The caller’s description of looking for victim information implicitly categorises her as
a victim. The caller goes on to to explicitly categorise herself and her son as ‘victims’ (line
14) but pauses throughout her turn (line 13) to display some difficulty in formulating her
experiences. By describing someone as ‘my son’ (line 14), the caller categorises herself as
a mother. These two categories are paired together, but can also be grouped in the wider
collection of categories, ‘family’. Following the consistency rule of categorisation (Sacks
1972), if two members of a family are victims, it is consistent to categorise the perpetrator

265
Emma Tennent & Ann Weatherall

within this same collection. In other words, the perpetrator of violence against mother and
son is also a family member. The caller invokes the common-sense knowledge that family
members live together in a shared home when she describes her problem as being unable
to leave with her son (line 21–22). As with Extract 1, the location of the caller’s problems
and the categories of people involved make inferentially available that the problem is about
domestic or family violence. The contact service worker provides only minimal responses
(lines 7, 19), rather than taking a full turn of talk which would display her understanding of
the caller’s description. Nonetheless, the caller describes her situation in ways that makes
domestic violence inferentially available, displaying her understanding of a relevant reason
to seek help from the victim support service.
In the third extract, a caller specifies that she is ringing about a ‘court appearance’ (line
4–5). Courts (like police stations) are settings that allow recipients to infer certain people
and activities (Dingemanse, Rossi, and Floyd 2017) such as lawyers, judges, offenders,
and so on. In the context of seeking court information from a victim support agency, this
description identifies the caller as a victim in the criminal justice system, a societal institu-
tion that has been the target of considerable feminist critique for re-victimising women (see
Jordan 2004).

Extract 3
01 CONTCT: kia ora victim support Molly speaking
02 (0.6)
03 CALLER: .hh hullo Molly it’s Taydi Letonen speaking, .hh
04 Molly I’m ringing with regard to um tch a court
05 appearance .hhh of Dayton Matthew Fenterwild
06 today in relation to a breach of protection order?
07 (0.2)
08 CALLER: .hhh um it’s got to: hh. now and I- .hh there was
09 a note to be made on the file by uhm (.) Stephen,
10 from the Norswith police .hhh uh rather than the
11 co:urts ring me they were going to ring me. hh
12 (0.2) uhm (.) >immediately as soon as he’d< been
13 dealt with (.) as far as l-like whether he was
14 released or in fact there were bail conditions!
15 CONTCT: mh[m,]
16 CALLER: [.h ]hh
17 (0.4)
18 CALLER: and I haven’t (0.4) ah heard anything yet and I’m
19 just starting >to get a little bit< (0.2)
20 [worried (about) where I’ll] stay
21 CONTCT: [ yeah no that’s all good]
22 (0.2)
23 CALLER: [ plan ]
24 CONTCT: [so what I-] d- what I will do: is I’ll pop you
25 ↑↑through to the police non emergency number,

By naming the person involved (line 5), the caller identifies the referent without mak-
ing explicit her relationship with him (contrast this with Extract 1a’s use of ‘husband’).

266
 Feminist conversation analysis

Nevertheless, her description of a ‘breach of protection order’ (line 6) provides inferences

about the named person and the nature of her victimisation. Protection orders are granted by
the New Zealand family court in cases of domestic or family violence. Thus, the recipient
(and analysts) can infer that the named man breached a protection order against the caller
and appeared in court on that charge. Common-sense knowledge about when protection
orders are given allows the inference that the named man is either a family member or (ex-)
partner of the caller; and the perpetrator of violence against her.
The caller’s reference to time is an important way that she presents the urgency of
her problem. The court appearance was ‘today’ (line 6), but at the time of the call, she is
yet to receive the information she was promised. Note the sound-stretch and exhalation
on line 8 before the caller describes the present time, ‘now’, and the way she cuts off
(‘I-’) to describe the prior arrangement that has not yet been met. She specifies that the
police were to contact her ‘immediately as soon as he’d been dealt with’ (lines 12–13),
making the fact she hasn’t ‘heard anything yet’ (line 18) a cause of worry. Here we see
the caller’s displayed orientation to the urgency of her circumstance – whether the per-
petrator of violence has been released, or the nature of his bail conditions are serious
matters that have practical ramifications of where she will stay (line 20) as well as her
emotional state.
The caller’s report of events that should have happened can be understood as a com-
plaint. However, the contact service worker responds by offering to connect the caller to
the police (lines 24–25). This offer ratifies the caller’s identity as a victim entitled to help.
However, the contact service worker displays her understanding that the appropriate agency
to help the caller is police, rather than victim support. Thus, in the next turn, the contact ser-
vice worker displays her understanding of the caller’s problem as a situation that somebody
else can resolve rather than as a complaint she can directly respond to. Here, the next-turn
proof demonstrates participants’ orientations to the relevant action, showing the different
ways participants can seek and provide help.

Concluding comments
Conversation analysis can provide new insights into long-standing feminist questions.
Detailed analysis can reveal the workings of issues such as women’s experiences with
the justice system, the difficulty of disclosing violence to institutions, and the practicali-
ties of help-seeking or service-provision. By focusing on participants’ orientations we can
also demonstrate how people make sense of shared cultural meanings about violence and
victimhood.
We set out to investigate how callers and contact service workers come to a joint under-
standing of the caller’s circumstances. We found that callers presented themselves and their
experiences in a range of ways. Descriptions of people, activities, and places are everyday
ways that participants invoke common-sense cultural knowledge about the gendered social-
moral world. In this way, we can examine how participants themselves orient to cultural
beliefs about violence (such as who is involved and where it occurs) and use them for
practical activities like seeking support. Calls to a victim helpline are a consequential site in
which shared beliefs about gendered violence are built, reinforced, or potentially challenged
in ways that can impact the provision of support.
Feminist scholars have long documented the struggles women face in having their expe-
riences recognised as victimisation and securing help through the institutions of criminal

267
Emma Tennent & Ann Weatherall

justice (Jordan 2004). Our research speaks to these issues by examining real-life cases
where women turn to a support service for help. In Extract 1, the caller references the dif-
ficulty of seeking support over the phone when her home is not safe. In Extract 2, the caller
identifies as both a victim and a mother, identities which can lead to contradictory expecta-
tions of protecting children and upholding the nuclear family (Powell and Murray 2008).
In Extract 3 the caller describes how miscommunication between police and courts has left
her in fear for her safety.
Empirical findings grounded in detailed analysis of recordings that show how women
understand their experiences and explain it in their own terms can be used to design practi-
cal applications to improve support services. One internationally recognised programme is
the conversation analytic role-play method (CARM) which develops research findings into
evidence-based training (Stokoe 2014). In this way, a focus on what participants actually
say and do as they seek support can inform the delivery of service, making a practical dif-
ference to the lives of women in need.

Future directions
For feminist conversation analysis to become widespread, more opportunities are needed
for scholars to learn about the methodological approach. Conversation analysis, including
the study of membership categorisation practices, is a highly technical discipline, and can
appear intimidating for a novice. Yet as we have shown, attention to the fine level of detail
can yield important insights into the operation of the gendered social moral order, and the
way participants invoke, reproduce, or challenge shared common-sense knowledge. An area
of exciting potential for feminist research is the analysis of embodied interaction, where
the tools of conversation analysis are used to examine how participants use gaze, gesture,
touch, movement, and interact with features of the physical environment as they go about
their daily lives.
The sharing of data corpora within the conversation analytic research community has
fruitful potential for collaborative feminist scholarship. Different researchers can reana-
lyse the same data to analyse different practices. For example, in our analysis above we
focused on description and categorisation, but the same data could be analysed in terms of
the sequential unfolding of the interaction, the way participants repair their talk, and so on.
We hope that the growth of feminist conversation analysis is accompanied by a spirit of col-
laborative endeavour, so that shared corpora become a valued resource for future feminist
research.
There are controversies regarding feminist conversation analysis. Yet to move beyond
these debates, we offer the parallel with feminist psychology. Once considered a contra-
diction in terms (see Fine and Gordon 1992), feminist psychology is now accepted as a
legitimate field of enquiry. Conversation analysis is a rigorous empirical methodology
that grounds claims about the social world in the observable conduct of participants.
Feminist conversation analysis provides a powerful tool to demonstrate that and how
the gendered social-moral order is understood by participants, and used to practical
ends. We encourage students and scholars of language, gender, and sexuality to take
up conversation analysis as a research approach. As we have demonstrated, grounded
observations of recordings of naturalistic data, and a deep engagement with participants’
own understandings holds remarkable potential for feminist work that remains largely
untapped.

268
 Feminist conversation analysis

Transcription conventions
Symbol Description
(0.2) Numbers in brackets represent silences in seconds
(.) Silence less than one-tenth of a second
[word] Square brackets represent overlapping talk – talk by different speakers
[word] that occurs at the same time
word= Equals signs represent ‘latched’ talk. This can be a continuation of talk
=word within a turn or across different lines on the transcript
wo:::rd Colons indicate a sound-stretch; more colons represent a longer stretch
>word< Arrows surrounding talk indicate faster speech
WORD Capitalisation indicates greater volume
°word° Speech between degree signs is quieter relative to the rest of talk
word Underlining represents emphasis
~word~ Words within tildes represent ‘wobbly voice’
. Full-stop indicates falling intonation at the end of a unit of talk
, Comma indicates continuing intonation at the end of a unit of talk
? Question mark indicates rising intonation at the end of a unit of talk
↑word Upward arrows before talk indicate a sharply increased pitch
wo:rd Underlining and colons represent intonation contours within words
.hh hhh Breathiness is represented with the letter ‘h.’ The length indicates the
relative length of breathiness. Preceded by a full-stop indicates an
in-breath; without a full stop indicates an out-breath
wo(h)rd ‘h’ within words represents breathiness (often laughter or crying) within a
spoken utterance
.snih Huhh These represent sniffing and sobbing respectively
( ) Empty brackets represent spoken words unable to be transcribed
(word) Words in brackets represent uncertain hearings
((ring)) Double brackets represent transcriber comments

Note
1 ‘kia ora’ (line 1) is a Māori greeting used in New Zealand English.

Further reading
Kitzinger, C. (2000) ‘Doing feminist conversation analysis’, Feminism & Psychology, 10(2), pp.
163–193.
Kitzinger’s article coined the term ‘feminist conversation analysis’ and argued for the compatibility
of feminism and conversation analysis. This work became the target of debates about the legitimacy of
feminist conversation analysis.
Speer, S. A. and Stokoe, E. (2011) Conversation and gender. Cambridge: Cambridge University Press.
This edited collection was the first book to bring together studies of gender using a conversation
analytic approach. Studies in a range of settings including children’s play, mediation, psychiatric
assessments, and helpline calls provide clear demonstrations of conversation analysis in use.
Stokoe, E. (2006) ‘On ethnomethodology, feminism and the analysis of categorial reference to gender
in talk-in-interaction’, The Sociological Review, 54(3), pp. 467–494.

269
Emma Tennent & Ann Weatherall

Conversation analysis is committed to an ethnomethodoligcal mentality, and Stokoe’s article

demonstrates how gender can be studied using those perspectives. She draws particular attention to
the way gender can be studied sequentially and categorially.
Weatherall, A., Stubbe, M., Sunderland, J., and Baxter, J. (2010) ‘Conversation analysis and critical
discourse analysis in language and gender research: approaches in dialogue’, in Holmes, J. and
Marra, M. (eds) Femininity, feminism and gendered discourse. Newcastle upon Tyne: Cambridge
Scholars Publishing, pp. 213–243.
In this compartive chapter, four renowned feminist language researchers analyse the same piece of
data with different methodological approaches. This provides a valuable consideration of the strengths
and limitations of various discursive approaches, including conversation analysis.

Related topics
The accomplishment of gender in interaction: ethnomethodological and conversation analytic
approaches to gender; gender and sexuality normativities; gender, stance, and category work in girls’
peer language practices; feminist poststructuralism – discourse, subjectivity, the body, and power;
semiotic representations of women criminals.

References
Butler, C. and Weatherall, A. (2011) ‘This is Charlotte’: a case of passing in children’s talk-in-
interaction’, in Speer, S. and Stokoe, E. (eds) Gender and conversation. Cambridge: Cambridge
University Press, pp. 231–249.
Cuff, E. C. (1993) Problems of versions in everyday situations. Washington, DC.: International
Institute for Ethnomethodology and Conversation Analysis & University Press of America.
Dingemanse, M., Rossi, G., and Floyd, S. (2017) ‘Place reference in story beginnings: a cross-linguistic
study of narrative and interactional affordances’, Language in Society, 46(2), pp.129–158.
Edwards, D. (1998) ‘The relevant thing about her: social identity categories in use’, in Antaki, C. and
Widdicombe, S. (eds) Identities in talk. London,: Sage Publications, pp. 15–33.
Ekberg, K. and Ekberg, S. (2017) ‘Gendering occupations: persistence and resistance of gender
presumptions about members of particular healthcare professions’, Gender & Language, 11(1),
pp. 100–120.
Fine, M., Gordon, S., and M. (1992) ‘Feminist transformations of/despite psychology’, in Fine, M.
(ed) Disruptive voices: the possibilities of feminist research. Ann Abor: University of Michigan
Press, pp. 1–25.
Hepburn, A. (2004) ‘Crying: notes on description, transcription, and interaction’, Research on
Language & Social Interaction, 37(3), pp. 251–290.
Jefferson, G. (2004) ‘Glossary of transcript symbols with an introduction’, in Lerner, G. (ed)
Conversation analysis. Studies from the first generation. Philadelphia: John Benjamins, pp. 13–31.
Jordan, J. (2004) ‘Beyond belief? Police, rape and women’s credibility’, Criminology & Criminal
Justice, 4(1), pp. 29–59.
Kessler, S. J. and McKenna, W. (1978) Gender: an ethnomethodological approach. New York: Wiley.
Kitzinger, C. (2000) ‘Doing feminist conversation analysis’, Feminism & Psychology, 10(2), pp.
163–193.
Kitzinger, C. (2005) ‘Speaking as a heterosexual: (how) does sexuality matter for talk-in-interaction?’,
Research on Language and Social Interaction, 38(3), pp. 221–265.
Kitzinger, C. and Frith, H. (1999) ‘Just say no? The use of conversation analysis in developing a
feminist perspective on sexual refusal’, Discourse & Society, 10(3), pp. 293–316.
Kitzinger, C. and Kitzinger, S. (2007) ‘Birth trauma: talking with women and the value of conversation
analysis’, British Journal of Midwifery, 15(5), pp. 256–260.
Land, V. and Kitzinger, C. (2005) ‘Speaking as a lesbian: correcting the heterosexist presumption’,
Research on Language & Social Interaction, 38(4), pp. 371–416.

270
 Feminist conversation analysis

LeCouteur, A. and Oxlad, M. (2011) ‘Managing accountability for domestic violence: identities,
membership categories and morality in perpetrators’ talk’, Feminism & Psychology, 21(1), pp.
5–28.
Lindström, A., Näslund, S., and Rubertsson, C. (2015) ‘The interactional organization of sex
assignment after childbirth’, Gender & Language, 9(2), pp. 189–222.
Mondada, L. (2012) ‘Embodied and spatial resources for turn-taking in institutional multi-party
interactions: participatory democracy debates’, Journal of Pragmatics, 46(1), pp. 39–68.
Potter, J. and Hepburn, A. (2010) ‘Putting aspiration into words: “laugh particles”, managing
descriptive trouble and modulating action’, Journal of Pragmatics, 42(6), pp. 1543–1555.
Powell, A. and Murray, S. (2008) ‘Children and domestic violence: constructing a policy problem in
Australia and New Zealand’, Social & Legal Studies, 17(4), pp. 453–473.
Sacks, H. (1972) ‘On the analyzability of stories by children’, in Gumperz, J. and Hymes, D. (eds)
Directions in sociolinguistics: the ethnography of communication. New York: Rinehard & Winston,
pp. 325–345.
Sacks, H. (1984) ‘On doing “being ordinary”’, in Atkinson, J. M. and Heritage, J. (eds) Structures
of social action: studies in conversation analysis. Cambridge: Cambridge University Press, pp.
413–429.
Schegloff, E. A. (1972) ‘Notes on a conversational practice: formulating place’, in Sudnow, D. N. (ed)
Studies in social interaction. New York: Macmillan, pp. 75–119.
Schegloff, E. A. (1997) ‘Whose text? Whose context?’, Discourse & Society, 8(2), pp. 165–187.
Schegloff, E. A. (2007) ‘A tutorial on membership categorisation’, Journal of Pragmatics, 39(3), pp.
462–482.
Speer, S. A. (1999) ‘Feminism and conversation analysis: an oxymoron?’, Feminism & Psychology,
9(4), pp. 471–478.
Speer, S. A. (2012) ‘Feminist conversation analysis: who needs it?’, Qualitative Research in
Psychology, 9(4), pp. 292–297.
Stokoe, E. (2006) ‘On ethnomethodology, feminism and the analysis of categorial reference to gender
in talk-in-interaction’, The Sociological Review, 54(3), pp. 467–494.
Stokoe, E. (2010) ‘‘I’m not gonna hit a lady’: conversation analysis, membership categorization and
men’s denials of violence towards women’, Discourse & Society, 21(1), pp. 59–82.
Stokoe, E. (2014) ‘The Conversation Analytic Role-play Method (CARM): a method for training
communication skills as an alternative to simulated role-play’, Research on Language & Social
Interaction, 47(3), pp. 255–265.
Stokoe, E. and Edwards, D. (2012) ‘Mundane morality: gender, categories and complaints in familial
neighbour disputes’, Journal of Applied Linguistics & Professional Practice, 9(2), pp. 165–192.
Weatherall, A. (2002) ‘Towards understanding gender and talk-in-interaction’, Discourse & Society,
13(6), pp. 767–781.
Weatherall, A. (2015) ‘Sexism in language and talk-in-interaction’, Journal of Language & Social
Psychology, 34(4), pp. 1–19.
Weatherall, A. and Edmonds, D. (2018) ‘Speakers formulating their talk as interruptive’, Journal of
Pragmatics, 123, pp. 11–23.
Whelan, P. (2012) ‘Oxymoronic and sociologically monstrous? Feminist conversation analysis’,
Qualitative Research in Psychology, 9(4), pp. 279–291.
Wilkinson, S. and Kitzinger, C. (2014) ‘Conversation analysis in language and gender studies’, in
Ehrlich, S. and Meyerhoff, M. and Holmes, J. (eds) Handbook of language, gender, & sexuality.
Somerset: Wiley, pp. 141–160.
Wowk, M. T. (1984) ‘Blame allocation, sex and gender in a murder interrogation’, Women's Studies
International Forum, 7(1), pp. 75–82.
Wowk, M. T. (2007) ‘Kitzinger’s feminist conversation analysis: critical observations’, Human
Studies, 30(2), pp. 131–155.
Zimmerman, D. H. and West, C. (1975) ‘Sex roles, interruptions and silences in conversation’, in
Thorne, B. and Henley, N. (eds) Language and sex: difference and dominance. Rowley: Newbury
House, pp. 105–129.

271
 18
Performance in action
Walking as gendered construction practice in
drag king workshops

Luca Greco

Introduction
Since the publication in 1990 of Butler’s groundbreaking book Gender Trouble, the con-
cept of performance as a synonym of action and impersonation has become a cornerstone
concept in gender studies and social sciences, and a key word to understand how people
act and account for their gender in their everyday lives. The concept of performance con-
tributes to renewing a focus on action in social sciences as an interactive, multimodal, and
historical phenomenon (Goodwin 2017). It also contributes to refreshing what the anthro-
pologist Conquergood (1989) calls the ‘performative turn’: a heterogeneous yet powerful
paradigmatic shift in social sciences and humanities, in which performance is mobilised as
a metaphor to explain social life as theatre (Goffman 1959).
This chapter deals with gender as performance as illustrated by walking in drag king
workshops. By focusing on gender as an embodied performance, I will demonstrate why an
integrated approach between social sciences and artistic perspectives (Sormani, Carbone,
and Gisler 2018) on performance is necessary. Such an approach resituates the emergence
of gender as performance in contemporary arts, and in social sciences, it underscores per-
formance rather than performativity and it considers the domain of artistic performance as a
theoretical resource for language, gender, and sexuality studies (LGSS).
This chapter combines ethnographic methods, sequential, and multimodal approaches.
The use of ethnography in conversation analysis has sparked a very intense debate in the
EM/CA community (Clemente 2013; Hopper 1990; Moerman 1988). It focused around the
methodological relevance of recruiting larger social structures in order to account for the
functioning of conversational structures, the perimeters of context and the ‘free-context’
nature of conversation (Maynard 2003; Schegloff 1997; Wetherell 1998). Nowadays, an
important analytical shift on multimodality, on interactions unfolding in complex semiotic
settings, and a focus on non-human agency have provoked what we can call an ‘ethno-
graphical turn’ in multimodal and ethnomethodological oriented conversation analysis, as
the work of the Goodwins has demonstrated throughout their career.
After a presentation of the heuristic power of the concept of performance for LGSS and
the value of studying walking in drag king workshops, I will analyse walking practices

272
 Performance in action

in drag king workshops through the analysis of three types of walking observed in this
context: in vitro walking, liminal walking, and in vivo walking. I will conclude with some
propositions for a closer dialogue between queer and gender studies, contemporary art, and
interactional perspectives.

Performance: a historical and theoretical overview

In this section, performance is first defined in relation to two very close concepts, which are
performativity and performative. I then present two ways through which performance has
been approached in literature: as a metaphorical and theoretical device, and as a distinctive
discursive and artistic genre.

Performance, performativity, and performative: some methodological and

theoretical issues
Performance, performative, and performativity are inextricably intertwined and must
be defined clearly. ‘Performance’ refers to action and impersonation – something that is
achieved step-by-step in the temporality of action for, with, and towards a real or imagined
audience. ‘Performative’ designates a type of verb (vs. constative) and can be used to refer
to gender as performative; i.e. gender as constructed by discursive actions. ‘Performativity’
refers to a complex process through which gender is iteratively constructed through a cita-
tional process (Butler 1990, 1993) and a normative horizon in which gender is achieved and
accounted for as a performance (Butler 2005).
Butler first refers to ‘performance’ in the preface of Gender Trouble (1990) to designate
the ways in which the actor and performer Divine accounts for their feminine gender in John
Waters’ movies. She notes three important aspects of gender performance in Divine’s play:
the questioning of dichotomies used to talk about and to explain gender, such as or through
natural/artificial, depth/surface, inner/outer dichotomies; the parody of the naturalness of
gender as part of the action repertoire of lesbian and gay cultures; and the naturalness of
gender as constructed through performative discursive acts (Butler 1990: xxvii).
What is at stake in Butler’s work is less performance than performativity. The concept
of the performative act, a cognate concept of performance in Butler’s paradigm, came from
speech act theory and philosophy of language (Austin 1962). A performative (vs. consta-
tive) utterance has the power to realise what is said by the speaker by proffering phrases
such as ‘I promise’, ‘I declare you married partners’, etc. Curiously, Austin considers the
language spoken in a theatrical context as ‘not serious’, ‘parasitic upon its normal use’, and
he excluded it from the class of performatives (Austin 1962: 22).
‘Performative’ in the Butler framework is extended to all types of action, not only to
those relating to speech and not only referring to sentences with performative verbs such
as ‘declare’, ‘promise’, etc. The Butlerian view of performatives comes from an interpreta-
tion the philosopher made of the concept of iterability (Derrida 1984). In this framework,
the realisation of what is said is guaranteed by the repetition (i.e. iterability) of the speech
act over time. In this conception, the realisation of an act depends less on the status (or
the authority; cf. Bourdieu 1991) of the speakers or on the type of institution in which the
speech act is accomplished (cf. the felicity conditions; Austin 1962) than on its repetition. In
this theoretical framework, language becomes a central feature in gender construction pro-
cesses as they are realised by participants through daily and routine actions (Butler 1997).
It is through repetition of gendered social acts that gender is constructed, stabilised, and

273
Luca Greco

acknowledged by co-participants in a number of very different contexts: in family interac-

tions, conversations among pairs, interactions at work, etc. (Speer and Stokoe 2011).
The distinction between performance and performativity has political, theoretical, and
methodological implications because gender is not just action. Gender is intertwined in a
web of power and normative relations which make gender performance possible (Butler
1993). This framework is quite far from a common-sense vision of gender as entirely a
matter of free will, in which everyone can change one’s gender as one can change one’s
clothes. Even the more subversive modes of gendered self-construction and presentation
practices are subjected to power or to what Butler calls ‘the heterosexual matrix’ (Butler
1993). If action and resistance to norms are never completely disanchored from power,
a resistance to these same norms, allowing the possibility of action and contributing to
shape it, is nonetheless possible. Methodologically and theoretically speaking, this view
entails a dialectical and dynamic view of action as a contextually shaped and renewing act
(Heritage 1984) and situated in two different levels of analysis: the ‘there’ of literal per-
formances (what people do in social exchanges) and the ‘not there’ (something that eth-
nographers must elicit using rich and dense ethnographies in order to unveil the ‘unsaid
traces’; Kulick 2005: 616).

Performance as a metaphorical and theoretical device

A focus on gender as practice, action, or doing is deeply connected with at least two differ-
ent research strands: research into feminist and gender studies ‘before Butler’, and interac-
tional approaches.
A vision of gender as performance (i.e. as doing) was developed by scholars such as the
British psychoanalyst Riviere (1929) with the concept of ‘masquerade’ – actions women
intentionally perform to be identified as submissive in relation to authoritarian men. The
French philosopher Simone de Beauvoir (1972), who coined the famous saying ‘one is not
born, but rather becomes, a woman’, contributed to shedding light on the historical context
and on the socialising practices which shaped women as a ‘second sex’. Finally, thanks to
the pioneering work of anthropologist Newton on female impersonators (1972), a view of
gender as a relational and bodily practice embedded in everyday activities becomes obvi-
ous within the constructionist approach, i.e. a theoretical framework in which the forms of
knowledge to reality are constructed by participants in the course of their repeated and daily
social practices (Berger and Luckman 1966).
Research in the fields of micro-sociology (Goffman 1959, 1976) and ethnomethodology
conceives gender as a practical accomplishment (Garfinkel 1967), a social process (Kessler
and McKenna 1978), and a process of attribution through which social actors interpret
gender cues presented by participants in interaction. Gender in this framework is treated
as something that is finely orchestrated by participants in front of an audience (i.e. co-
participants in daily interactions), an activity labelled by Goffman as performance: ‘the
activity of a given participant on a given occasion which serves to influence in any way
any of the other participants’ (1959: 8). The ‘Agnes case’ in Garfinkel’s survey (1967) is an
excellent example of gender as the work (or as the performance) accomplished by Agnes,
a transsexual woman, in order to be interpreted as an ordinary woman (Mondada, infra).
Gender transformation is presented through the lens of socialisation practices accomplished
in every moment of life, learned by participants in the course of their actions, and con-
structed through a multiplicity of semiotic resources such as speech, vocal, visual, kinesic,
and material resources.

274
 Performance in action

Performance as a discursive and artistic genre

Performance is not only a descriptive resource mobilised by scholars to understand how
people interact in the world. It is also an object of inquiry in social sciences. In linguistic
anthropology, performance is a mode of spoken verbal communication consisting of the
assumption of the responsibility to an audience for a display of communicative competence
(Bauman 1975). This focuses attention on social interaction, the aesthetic and evaluative
dimension of social life, and presents three important characteristics: to be interpretable,
reportable, and repeatable (Hymes 1981).
Performance is a particular way of speaking in which social actors evaluate the discourse
of participants in specific social contexts such as the management of gossip disputes and
confrontation-report-offence among girls through conversation (Goodwin 1990). This con-
cept allows scholars to focus their attention on the multimodal dimension of social encoun-
ters and on a dynamic and culturally anchored conception of the backstage/frontstage
dichotomy in speech events (Finnegan 1998). Gender can be a relevant social category used
to explain the diversity of speech styles in performance. In her study on forms of speech
use in a Malagasy speech community, Keenan-Ochs (1974) remarks on a type of social
gendered division of performance labour: men use a kabary speech form, a form privileg-
ing more indirect style forms, whereas women mobilise more often direct speech forms
known as resaka. Finally, sexuality, gender, and race can be finely intertwined in verbal
performances. In African American drag queen speech in gay bars, Barrett (2017) identifies
a discursive genre typical of performance. This is speech designed for an audience com-
posed of gay males sharing an important number of values and features of gay subculture
with drag queens; this includes many allusions to sexual activities and the intertwining of
some discursive features typical of Black and white bourgeois women. Barrett presents
drag queen performances as irreducibly multi-voiced (Backhtin 1973) in that participants
index a multi-layered identity which crosses different social categories: Black woman,
white Christian woman, gay men, drag queens, etc. Contrary to the Butlerian approach,
which views drag queen performances as symbols of subversion of gender norms, Barrett’s
analysis sheds light on a misogynistic potential in drag queen speech practices and offers an
interesting intersectional (Crenshaw 1991) analysis of drag queen practices in which gender,
race, class, and sexuality are intertwined.
All of these works confirm that performance is not constituted solely by texts; it is bodily
constructed1 and it creates a multiplicity of audiences and participation frameworks. These
points are shared by a corpus of artistic works and reflections on performance as an artistic
genre in the framework of performance studies (Schechner 2006; Schneider 1997).
In contemporary art, performance disrupts a classical view of theatrical representation
based on text, the linearity of narrative, the presence of a character, and a dichotomic vision
of space which includes a performer on stage in front of a comfortably seated audience.
Within this framework, we can define performance as an action accomplished once, across
from, or oriented towards a physically present, absent, imagined, or technologically medi-
ated participant. The distinctive features of performance are inter alia a focus on a racial-
ised, gendered, and sexualised body; the primacy of experience lived by the social actors;
and the processual dimension of action.
Since the 1960s at the latest, many contemporary artists and feminist activists have con-
tributed to nourishing an idea of gender as a routine through their performances, which can
be considered to be embodied processes, and ideological and artificial constructs (Jones
1998). Some of the ideas emerging in their work echoes some of the points underlined in

275
Luca Greco

gender and queer studies, and in social sciences in general, as central features for a vision of
gender as performance: the relation between body and language, the body as a process (vs.
result), and the power of bodily transformation.
Sprinkle, a post-porn activist and feminist, contributes to conceiving the body as a pro-
cess through her performances. In Anatomy of a Pin-Up (1980) she shows, through an image
of her body, the procedures mobilised by the artist to construct a perfect pin-up body. To this
end, each body part is related through an arrow to an inscription in which the artist explains
how she achieves the result of the pin-up forms. In this way, Sprinkle transforms her body
into an accountable one through which the spectators can approach the pin-up forms more
as a process than as a result.
The theme of bodily transformations through makeup activities has attracted many per-
formers since the sixties. Eleanor Antin, a visual artist, performer, and feminist, is a pioneer-
ing figure in this field. In The King, a 1972 video-recorded performance, the artist shows her
transformation into a male persona – a king. We can see the artist facing the mirror while
constructing a beard and a moustache. She wears a hat and applies some facial hair, antici-
pating what a group of drag kings would do in the 1980s and 1990s in the United States
and Europe (Greco 2018; Halberstam 1998). The theme of The King persistently dominated
her performances between 1972 and 1978. In this period, she embodied the King of Solana
Beach, who walked and met his subjects in areas of southern California. In these perfor-
mances she embodied a male character, while at the same time maintaining some feminine
attributes, such as breasts. She contributed to the emergence of a new type of performance,
anticipating some of the drag king performances under scrutiny in the next sections.

Summary
In contemporary art and performance studies, performance has a specific meaning referring
to an artistic genre whose roots are clearly situated in the modernist avant-garde, especially
Futurism and Dadaism, and in the 1950s and 1960s with the performances of artists such as
John Cage, Vito Acconci, Yoko Ono, Martha Rosler, etc. (Goldberg 1979). The way in which
performance is approached in social sciences is quite different. Performance is either related
to theatre or to a common-sense vocabulary in which performance is a synonym of a theatri-
cal, artistic show, or refers to specific discursive genres in different cultures. Inspired by the
works of performers and social scientists presented in this section, I will consider walking
practices as performances and a gender (de)construction device. My analysis focuses on the
experiential, bodily transformation, and processual dimensions of walking practices.

Presentation of the setting and methods

A drag king is generally a female-assigned person at birth, who, through the mobilisation
of a multi-semiotic repertoire – verbal, visual, tactile, material (makeup, prosthesis, cloth-
ing, glue) and bodily resources (they use hair cut and glued on the face to make a beard)
– embodies several types of masculinities. The embodiment of masculinities is achieved in
the pursuit of at least three objectives: a personal aim, which is the desire to experience gen-
der and its plasticity; an artistic desire to be on stage and create from and through gender;
and a political agenda, in that the embodiment of masculinities highlights gender’s fictional
character and helps to destabilise gender categories.
Drag king workshops are social occasions in which people take time to construct
a male body, together with the help of some experts – people knowledgeable in gender

276
 Performance in action

transformation and who perform on stage as drag kings – through the choice of a character
to embody, makeup, and some exercises in which walking constitutes a distinctive locomo-
tive feature and a topic of conversation.
The workshops are held once a month in the Rainbow House building, an LGBTQIA+2
socialisation space based in Brussels; it is an architectural setting within which drag kings
circulate and interact. At the entrance of the Rainbow House there is a bar, which is on the
ground floor. There are meeting rooms on the upper floors. The workshop takes place on
the second floor and, once it is finished, drag kings usually go down to the bar to drink with
the public. From there, they can visit the most famous square of Brussels, the Grand-Place,
which is two minutes’ walk from the workshop site.
In drag king workshops, once makeup is complete, walking is performed by partici-
pants in the workshop space and outside on the street. In both cases, participants create an
experience with and on their gender-transforming bodies and in the ways they approach
space, movement, and gender. Walking can also become a topic mobilised by participants
after or before the exercises, while standing in a fixed position. It is interesting to focus
analytical attention on an activity such as walking because it grasps gender in its locomo-
tive and discursive dimensions, i.e. how gender and bodily transformations are experienced
through walking itself and discourse on walking. Through an ethnographic and interactional
approach, we can look at practices in their temporal unfolding and have access to social
norms about walking, gender, and space as they emerge in the discourses of the social actors.
Several methodological devices were used in this study. These include interviews with
the participants, video recording of several activities (dressing and makeup sessions, visits to
public places, performances, interviews), and gathering written and visual documents about
the group’s activities. This methodology has the advantage of considering walking as a prac-
tice per se, in its corporeal and locomotive dimensions, and as a topic mobilised by partici-
pants in the workshop. Moreover, it gives the possibility of grasping the duality of gender
as performance, a practice achieved by participants in the course of their routines, and as a
reminder of performativity; i.e. related to social norms concerning the relations participants
establish between gender, walking, and space. A long-term ethnography focusing on a longi-
tudinal vision of the activities of participants from the moment they arrive to the moment they
leave is necessary to grasp the complexity of the spatio-temporal webs in which the activities
are situated and intertwined. This presupposes rethinking the methods scholars use to identify
temporal and spatial boundaries of activities, to follow participants in the course of their activ-
ity, and to question the participation of the ethnographer in the activities they observe.
Three types of spaces and walking can be found in the workshops.
A first type of walk takes place in the makeup space (Figure 18.1), which is transformed
into a space of locomotion and experimentation (Figure 18.2) after the makeup activity. This
is what I call an ‘in vitro walk’: a type of walking experienced, accomplished, and lived,
but also thematised and theorised with the help of an expert, Max, the leader of drag king
workshops in Brussels.
A second type of walk, a ‘liminal walk’, takes place in the space of the staircase – a bor-
der space, which allows the transition from the experimental space of the ‘in vitro walk’ and
space (second floor), to the bar, located on the ground floor, where masculinity is lived with
the public of the Rainbow House bar.
A third type of walk, the ‘in vivo walk’, takes place outside on the path between the
Rainbow House and the Grand-Place. This is a space in which participants test their new
identities in silence, focusing on their own feelings, and testing and being attentive to the
reactions of pedestrians.

277
Luca Greco

Figure 18.1 The makeup space.

Figure 18.2 The makeup space transformed into a walking space.

The temporality which unites these three types of space and walk makes possible a tran-
sition from what I call, following the dramaturgic vocabulary of the sociologist Goffman
(1959), the ‘backstage’ of gender, situated in the in vitro space and in the liminal space, to
the ‘stage’ of gender, located in the in vivo space. This transition from one type of space
(in vitro) to another (in vivo) constitutes new embodied and gendered subjectivities in the
practices observed in the workshop.
In a theoretical framework inspired by queer linguistics (Barrett 1997; Milani 2013) and
multimodal analysis (Goodwin 2006; Mondada 2016), I conceive walking as a social device
producing gender and subjectivities, and as something inseparable from the spaces in which
the walking is deployed. Thus, walking is not just a practice taking place in a spatial setting

278
 Performance in action

or just a topic. Walking constructs and deconstructs gender; it builds the space while, at the
same time, it is shaped by the plastic materiality of the space.

Walking in drag king workshops

Walking is an irreducible interdisciplinary object of study, which traverses the social sciences,
humanities, and arts (Lorimer 2011). As an interactional resource and a multimodal resource,
walking is finely orchestrated (Goffman 1972), intertwined with talking, and produces unex-
pected participation frameworks (Mondada 2014). As a social resource, walking is a body
technique through which class, geographical origins, and gender are indexed (Mauss 1973).
In this section, walking is approached as a gendered practice through which partici-
pants experience gender and construct articulations between gender, movement, and space
in three different cases: in vitro walking, liminal walking, and in vivo walking. As an expe-
riential practice, walking will be conceived as a performance – a privileged arena for the
transformation and the discovery of self.

Articulating gender, space, and walking in in vitro walking

It is in the context of in vitro walking that walking itself becomes a topic in Max’s speeches
and it is conceived as a place of experimentation of the self in quite a didactic register. The
male gendered transformation of the participants is achieved and Max takes the floor to
deliver some remarks about gender and walking.3

Extract 1 – Socialised as masculine or feminine

M: Max (#)

1 Μ la majorité euh qu’on catégorise au

the majority (of people) that we categorize
2 niveau masculin/ #les hommes/# ont gardé/ la
as masculine men have kept the
#makes a step#
3 démarche d’enfant/ enfin pas d’enfant mais la
child’s gait well not exactly but the
4 démarche normale du pied #(—fin) on a un pied/#
usual move of the foot (well) we have a foot
#lift their tools #
5 #et alors on met le talon# d’abord et comment
and then we put the heel first and how
#bring the heel on the floor#
6 la : #comment on appelle ça ah en kiné / c’est
now how do you call this in physiotherapy/ it’s
#reproduces with the hand foot movement-→
7 d’abord le talon fin c’est vraiment la les trues de
first the heel well it’s really the stuff of
8 la marche enfin la la logique de la marche en fait#
-----→#
the walking well the the logic of the walking like

279
Luca Greco

9 tandis que: on apprend aux filles justement/ à

while we teach girls precisely/ how to
10 contrecarrer vraiment à faire tout le
oppose really to do all the
11 contraire de ce que : le pied est prévu pour/ donc
opposite of what the foot is intended for/ so
12 euh avec les hauts talons alors là j’ai absolument pas
uh with high heels then I absolutely don’t have
13 le choix parce qu’elles doivent poser d’abord
the choice because they (girls) must first put
14 l’avant du pied et pas du tout dans le sens de la
the tip of the foot (on the ground) and not at all in direction of
15 marche (2) et done /là on peut aussi directement/
walking (2) and so/ then we can also directly/
16 uniquement par par le balayage et comme les
just through watching people and how
17 personnes se se s- déplacent et comment elles posent
people are moving and how they put
18 le pied et ben on peut voir si c’est masculin/ ou si
the foot on the ground and well we can see if it’s masculine/ or if
19 c’est féminin/ socialisé socialisé masculine
it’s feminine/ male socialized

In this excerpt we can observe a thematisation and a theorisation of the articulations

between space, walk, and gender. These are accomplished by a focus on the movement
of the foot on the ground, a naturalisation of the male step, and a denaturalisation of the
female step.
The walk of those who are categorised as men is presented as a normal process (l. 2–4)
– an evidence or a general truth which mobilises a noun group (‘les hommes’ l. 2) meto-
nymically represented by the foot step (l. 2) whose walking is characterised as ‘usual’ (l.
4). After this preface in which Max presents the terms of the problem, they continue to
speak, giving some examples in an assertive mode – ‘we have a foot’ (l. 4) – and focusing
their speech on the foot’s movements in male walking. Within this logic, Max lifts their4
foot and, then, they bring the heel onto the floor, reproducing, in a mechanical way, ‘male
walking’ (l. 4–5).
If the male walk is presented as natural, the female walk is presented as a cultural and
learned practice, which is carried out by first laying the toes and then the rest of the foot
(l. 9–15). This practice – the initial placement of the front of the foot – is presented as an
injunctive one which goes against the natural functionality of the foot in walking; i.e. ‘what
the foot is intended for’ (l. 11).
This walking theorisation practice about gendered walking produces gender generalisa-
tions such as ‘one can see by walking whether it is masculine or feminine’ (l. 17–19) and it
is accompanied by a subsequent degree of abstraction.
Walking is no longer a spoken practice, exemplified and reified through steps on the
ground by placing the heel first if masculine or the toe if feminine. Max gestures with their
hand, reproducing the movement of the foot (l. 6–8); this gives a para-medical framing to
the walking practice, highlighting the scientific, naturalistic, and objectifying dimensions of
his description.

280
 Performance in action

Walking can also be an opportunity to find new trajectories, feel new bodily sensations,
and discover some characteristics which are not necessarily planned, but which emerge as
evidence. It is in this context that Alexandre expressed a desire to discover a faster walk, as
demonstrated in the next excerpt:

Extract 2 – A faster walk

M: Max
A: Alexandre (£)

1 Μ mais testez un peu les les même c’est un personnage

but test a little bit the the same it’s a male
a »fixed position---→
2 masculin d’ailleurs hein/ c’est arléquin/ il est fort/
character by the way right/ it’s Harlequin/ he’s strong/
3 aérien/£
aerial
---→£
4 A £ouais °xxx°
yeah°xxx°
£they walk--→
5 Μ et/ so:n caractère cha::nge£/£donc on voit que même
and/ his character changes/ so we see that even
a ---→£ £fixed position---→
6 dans le théâtre/ classique\ on va dire que [ce soit
in classical/ theatre\ we can say that it is
7 A [(ss ss)
8 Μ ben y a eu pense a [plein
well there was think a lot of
9 A [xx là j’ai envie d’aller plus vite
xx now I want to go faster
10 ££(1)
--→££they walk--→>
11 Μ Ah
12 (1)
13 A ouais\ quand je fais comme ça
yeah\ when I do like that
14 they accelerate their walking-→>

While all the participants walk in the space of the workshop, Max invites them to search for
new ways to embody masculinities through the figure of Harlequin, a central character of
the commedia dell’arte, mixing features of strength and softness (l. 1–8). Alexandre, a male-
to-female trans person who participates in drag king workshops, takes the floor at line 9 to
express their desire to engage in a faster walk. A careful look at their actions shows a com-
plex multimodal trajectory accounting for their turn-taking. Starting from a fixed position (l.
1–3), they react to the first part of Max’s turn by producing an affirmative particle (l. 4) and
by engaging in a walking practice (l. 4) just in time to come back to a fixed position in the
middle of Max’s turn (l. 5). This return to a fixed position allows Alexandre to show signs

281
Luca Greco

of incipient speakership, as in line 7, and to announce a new way of walking (l. 9) which
will be achieved in line 14. This announcement (l. 9) projects several actions: a motion state
change (a return to walking, l. 10), ratification from Max (l. 11), the construction of a com-
plex sentence through the temporal conjunction ‘quand’ (when) (l. 13) relating and adding
to their previous turn (l. 9), and an acceleration of their walking (l. 14).
The walking practices, as we can see in these examples, are both decontextualised, natu-
ralised, or decomposed into micro-movements, objectified, and connected to gender norms
by an expert participant, Max, standing in motionless position (Excerpt 1). Alternatively,
they are experienced as an opportunity to research, improvise, and discover unplanned spa-
tial trajectories (Excerpt 2) by novice participants in mobile configurations.
The focus on experience, creativity, and improvisation, (to find new ways to walk and to
feel their body) and the ‘rehearsal aspect’ of the walking (one should test their body through
walking in an ad hoc situation) are typical features of performance. In this context, the use
of a video-recording device is particularly relevant as it allows the scholar to grasp the bod-
ily and temporal dimension of gender construction practices.

Articulating gender, space, and walking in liminal walking

Following the temporality of the activities planned in the workshop, once the exercises have
been achieved, drag kings can decide to go downstairs into the bar space located on the
ground floor of the Rainbow House.
This is a very important moment because this transition towards the bar marks two types
of border crossings. First, there is a movement away from a more intimate space of work
and experimentation, which has taken place only with the members of the workshop and
which involves a form of non-mixed participation. Second, there is movement towards a
mixed participative space where one has to interact with unknown people or with known
people maybe seeing them in drag for the first time.
In the next excerpt, one of the participants of the workshop gives an account of the transi-
tion from the in vitro space of walking to the in vivo space of the bar:

Extract 3 – Going down

B: Béatrice

1 Β je pense que si le fait de descendre\ personnellement

I think that yes going dovm\ personally

((several lines omitted))

13 Β ça débloque vraiment les les escaliers étaient un peu

It unlocks really the the stairs were a little bit
14 euh ((rire)) qu’est ce que je vais dire qu’est ce que
euh ((laughing)) what I’m going to say what
15 je vais faire et puis du moment où tu es là ben tu
I’m going to do and then since you are there well you
16 oublies en fait c’est là que tu re- rentres un peu
forget in fact it’s there that you get back into
17 dans ton personnage vraiment tu te dis c’est le
your character really you say to yourself it’s the

282
 Performance in action

18 fait d’être face à des gens parce que si non autrement

fact to be face to people because otherwise
19 ça reste un jeu —fin
it’s like a game/play right

In this excerpt, Béatrice gives an account of descending the staircase as a constitutive

moment of the drag king experience.
In this liminal space, participants cross a border between two different types of space
(the space of makeup and locomotion, and the space of the bar) and prepare a gradual entry
into character.
The entry into the bar, the in vivo space, is a crucial moment through which participants
leave the ‘backstage’ of gender (marked by rehearsals in which people use make up, try to
walk in different ways, etc.) towards the ‘stage’ of gender (a scene in which one gives a
presentation of themselves facing an audience).
The gradual entry into character and the passage through the liminal space of the stairs
does not allow time for reflexive activity. During this space and gender transition, Béatrice
doesn’t stop to ask themselves questions such as ‘what I can do?’ or ‘what I can say?’ (l.
14–15) once they arrive in the bar.
A focus on moments other than the exercises – occasions in which participants talk in a
spontaneous way among themselves and with me, the ethnographer – gives access to unex-
pected revelations about their experiences and helps to (re)construct the temporality of the
gender transformation.

Articulating gender, space, and in vivo walking

In this last section dedicated to in vivo walking and space, we consider a moment in which
walking is practiced in a public space – in the open air or in the bar. This is an important and
quite mysterious phase. The moments spent at the bar are the only ones I did not record with
my camera for a number of reasons.
First, I thought that because the workshops had come to an end, I no longer needed to
film and to have exploitable video-recorded data. Second, it would have been very compli-
cated to obtain consent forms from everyone present in the bar. Third, these moments of
descent, both into the bar and the exit into the big square, allowed me to leave my position
as researcher and change my observational status in my ethnographic encounters with the
drag kings. I also wanted to experience with them the meaning of the walk; thus, I decided
to dress in drag and enter the space of the bar and the big square with them.
The absence of video recordings of the practices outside of the workshop could be inter-
preted as an important limitation of the methodology; however, I think that all types of
data gathered by observation, participation, and, of course, recording activities are equally
relevant to the analysis. They nourish what the anthropologist Geertz (1973) call a ‘thick
description’ and they open unexpected analytical spaces for the researcher. In the case of
walking, I realised how walking in public spaces is a very interactional activity, because we
are constantly faced with possible contact with the people we meet in the course of walking.
It gives us an opportunity to modify some features of our walking and posture, to adapt our
way of walking to the contingencies of the situation, and to discover the desire to walk in
a certain (gendered) way. A sustained practice of the drag king workshops allows (expert)
participants to develop a political consciousness of the space, of the way it is approached by
men and women, and of the gendered nature of movements, gestures, and gazes deployed

283
Luca Greco

there. In the case of walking drag king practices, we can say that the action of walking does
not exclude a theorisation – a representation of walking as a purely corporal or political
practice. Moreover, a mechanical and an anatomical vision of the body does not exclude a
political consciousness of gendered issues in walking in a public space. Walking practices,
just as all social practices in everyday life, are agentive.
On the one hand, walking is determined by a binary gendered conception of the world –
men and women do not walk in the same way – and there is indeed a gendered socialisation
in walking practices, forbidding and/or favouring certain types of walk in persons assigned
as ‘boys’ or ‘girls’ at birth. On the other hand, walking offers favourable grounds for the dis-
covery of new sensations and subjectivities, of political awareness of the space and political
occupation of the streets, such as in the example of drag marches.

Conclusions
This chapter focuses on performance as a theoretical device, an analytical object, and an
artistic genre in social sciences and contemporary arts. I show its historical and theoretical
background and its relevance for LGSS through the case of walking practices in drag king
workshops.
Inspired by an approach mobilising queer linguistics, multimodal analysis, and perfor-
mance studies, I focused on walking practices as a gender (de)construction device underly-
ing bodily transformation, and the experiential and processual dimensions.
Focusing my attention on the temporality and spatiality of walking activities, I have shown
the necessity for an ethnographic approach. Within this perspective, I have identified three
moments and three spaces through which drag kings act during the workshop, after makeup:

• An in vitro space, in which walking is experienced in two different ways. As a topic,

walking is decomposed, decontextualised, and articulated in terms of gender and
space. As a practice, walking is lived by participants as a medium for the discov-
ery of new types of bodily consciousness, spatial trajectories, and new, unexpected
identities;
• A liminal space in which the descent of the stairs allows one to cross the threshold of a
public space and to plunge into character; and
• An in vivo space in which participants experience through walking a new body in con-
tact with pedestrians and a gendered consciousness of the space.

If the walk in these three spaces is not separable from a reflexive posture, then, we must
specify that the levels of reflexivity at work are not the same. In the in vitro and liminal
spaces, which I also call ‘the backstage of gender’, we are faced with a self-centred reflexiv-
ity, which is very focused on one’s own body and on one’s own transformation. In the in vivo
space, which I call the ‘stage of gender’, we have a rather relational reflexivity, interaction-
ally anchored, and much more oriented towards a possible encounter.
This pervasive reflexive dimension in walking practices, combined with decontextualisa-
tion, decomposition, introspection, discovery, and research operations, transforms the par-
ticipants of drag king workshops into real explorers of gender and space, and theoreticians
of walking practices. The focus on experimentation, experience, and the discovery of new
and unexpected trajectories and selves make walking practices performances in the artistic
sense of the term.

284
 Performance in action

The intertwining of stage and backstage, action and representation, permanent adjust-
ments and abstraction/decomposition of practices, self-oriented reflexivity and relational
reflexivity makes walking practices close to scientific experiments, which are observable,
accountable, and objective, and to artistic processes, because they are open to improvi-
sation, experimentation, and creation. The ethnographic and analytical focus on waking
practices as a gender (de)construction device allows me to consider gender through its cor-
poreal and locomotive dimensions, as it is experienced and thematised by participants, and
to consider bodies as powerful instruments which can experience new gendered personas
and exciting lives.

Future directions
Several issues arise from the analysis and the methodologies proposed in this chapter.

• A theoretical issue: the study of performance and performativity in LGSS needs to

be nourished by the important contributions feminist contemporary artists gave to
a vision of gender as a performance. Performance is not just a theoretical device, a
discursive genre; it is also an artistic genre raised in contemporary arts in the 1960s
and 1970s in concomitance with political activism for gender equality and sexual
freedom. In this perspective, we could consider artistic works as analytical material,
theoretical devices social scientists could integrate in their methodological toolkit
in order to give a ‘thick’ description of how gender performance works in everyday
activities;
• A methodological issue: the combination of ethnographic methods – interviews, par-
ticipant observation, audio/video recording, and collection of textual materials con-
cerning the community under the scrutiny – multimodal approaches, and sequential
analysis is something scholars could benefit from in their analysis if they want to grasp
the multilayered nature of context and of social actions. In this framework, it is possible
to account for different levels of analysis description: intra-turn (turn design), inter-
turn (sequential level from adjacency pairs to complex sequences), speech event (a
combination of sequences constituting the speech event), multimodal configurations of
the action; and extra-interactional level: background knowledge relevant for the under-
standing of interaction and pertaining to the life of institutions in which interactions
occur and to the life of participants under scrutiny;
• A political issue: the analysis of gender as a performance cannot be detached from
political and social contexts in which actions occur. The necessity to integrate into our
toolkit ethnographic approaches has been demonstrated in this chapter. This approach
calls for a Critical EM/CA perspectives in which the political and existential issues at
work in gender construction process are arising by a combination of ethnographical,
multimodal and interactional approaches.

Notes
1 A focus on body as a gender (de)construction device is curiously still neglected in LGSS, and in mul-
timodality research, with some worthy exceptions (Goffman 1976; Goodwin 2006, 2015; Mondada
2011).
2 Each of the letters included in the acronym LGBTQIA+ refers to gender minority groups: Lesbian,
Gay, Trans, Queer, Intersex, Asexual … It designates the plurality and the diversity of the political

285
Luca Greco

activism focused on gender and sexuality identities. The symbol ‘+’ signifies that the coalition build-
ings between gender minorities and their allies (sometimes, we found another ‘A’: allies) are a histori-
cal construction, a working progress construct.
3 Transcription conventions in this chapter are inspired by those proposed by Mondada: https://franz.u
nibas.ch/fileadmin/franz/user_upload/redaktion/Mondada_conv_multimodality.pdf.
4 Concerning the participants in the drag king workshops, we use the neutral gendered form ‘they’.

Further reading
Birdwhistell, R. (1970) Kinesics and context: Essays on body motion communication. Philadelphia:
University of Pennsylvania Press.
In this book, readers can find some interesting and pioneering remarks on bodily resources as a
gender display and recognition device.
Goffman, E. (1976) Gender advertisements. New York: Harper & Row Publishers.
In this paper, Goffman proposes a praxelogical and a multimodal vision of gender: the social
asymmetry between men and women is constructed through corporeal resources depicting women as
fragile and subordinate subjects.
Goodwin, H. M. (2006) The hidden life of girls: games of stance, status, and exclusion. Oxford:
Blackwell.
In this book, Goodwin focuses on multimodal resources through which participants construct
gender, power, and race in interaction.
Moerman, M. (1988) Talking culture: Ethnography and conversation analysis. Philadelphia:
University of Pennsylvania Press.
In this book, Moerman proposed a culturally contexted conversation analysis: a theoretical and an
analytical framework in which sequential approach is combined with ethnographic perspective and
cultural analysis.

Related topics
Gender diversity and the voice; gender and sexuality in discourse: semiotic and multimodal
approaches; the accomplishment of gender in interaction: ethnomethodological and conversation
analytic approaches to gender; multimodal constructions of feminism; feminist poststructuralism –
discourse, subjectivity, the body and power.

References
Antin, E. (1972) The king [video-recorded performance].
Austin, J. L. (1962) How to do things with words. Oxford: Oxford University Press.
Backhtin, M. (1973) Problems of Dostoevsky’s poetics. Ann Arbor: Ardis.
Barrett, R. (1997) ‘The «homo-genius» Speech community’, in Livia, A. and Hall, K. (eds) Queerly
phrased. Language, gender and sexuality. New York: Oxford University Press, pp. 181–201.
Barrett, R. (2017) From drag queens to leathermen. Language, gender and gay male subcultures.
Oxford: Oxford University Press.
Bauman, R. (1975) ‘Verbal art as performance’, American Anthropologist, 77(2), pp. 290–311.
Berger, P. L. and Luckmann, T. (1966) The social construction of reality: A treatise in the sociology of
knowledge. London: Penguin Books.
Bourdieu, P. (1991) Language and symbolic power. New York: Polity Press.
Butler, J. (1990) Gender trouble: feminism and the subversion of identity. New York: Routledge.
Butler, J. (1993) Bodies that matter: on the discursive limits of sex. New York: Routledge.
Butler, J. (1997) ‘Performative acts and gender constitution: an essay in phenomenology and feminist
theory’, in Conboy, K., Medina, and Stanbury, S. (eds) Writing on the body: Female embodiment
and feminist theory. New York: Columbia University Press, pp. 401–418.

286
 Performance in action

Butler, J. (2005) Giving an account of oneself. New York: Fordham University Press.
Clemente, I. (2013) ‘Conversation analysis and anthropology’, in Sidnell, J. and Stivers, T. (eds)
Handbook of conversation analysis. Malden, MA: Blackwell, pp. 688–700.
Conquergood, D. (1989) ‘Poetics, plays, process, and power: the performative turn in anthropology’,
Text and Performance Quarterly, 1(1), pp. 82–95.
Crenshaw, K. (1991) ‘Mapping the margins: intersectionality, identity politics, and violence against
women of color’, Stanford Law Review, 43(6), pp. 1241–1279.
De Beauvoir, S. (1972) The second sex, trans. Parsley, H. M. Harmondsworth: Penguin.
Derrida, J. (1984) Margins of philosophy, trans. Bass, A.. Chicago: Chicago University Press.
Finnegan, R. (1998) Oral traditions and the verbal arts. A guide to research practices. London:
Routledge.
Garfinkel, H. (1967) Studies in ethnomethodology. Cambridge: Polity Press.
Geertz, C. (1973) The interpretation of cultures. New York: Basic Books.
Goffman, E. (1959) The presentation of self in everyday life. New York: Anchor Book.
Goffman, E. (1972) Relations in public. New York: Penguin.
Goffman, E. (1976) Gender advertisements. New York: Harper & Row Publishers.
Goldberg, R. L. (1979) Performance art: from futurism to the present. New York: Thames & Hudson.
Goodwin, C. (2017) Co-operative action. Cambridge: Cambridge University Press.
Goodwin, H. M. (1990) He-said-she-said: talk as a social organization among black children.
Bloomington: Indiana University Press.
Goodwin, H. M. (2006) The hidden life of girls: games of stance, status, and exclusion. Oxford:
Blackwell.
Goodwin, H. M. (2015) ‘A care-full look at language, gender and embodied intimacy’, in Allyson,
J. (ed) Shifting visions: gender and discourses. Cambridge: Cambridge Scholar Publishing, pp.
27–48.
Greco, L. (2018) Dans Les Coulisses Du Genre, La Fabrication De Soi dans les Ateliers Drag King.
Limoges: Lambert Lucas.
Halberstam, J. (1998) Female masculinity. Durham, NC: Duke University Press.
Heritage, J. (1984) Garfinkel and ethnomethodology. Cambridge: Polity Press.
Hopper, R. (ed) (1990) ‘Ethnography and conversation analysis after Talking Culture’, Research on
Language and Social Interaction, 24(1–4), pp. 161–171.
Hymes, D. (1981) ‘In Vain I Tried to Tell You’: Essays in native american ethnopoetics. Philadelphia:
University of Pennsylvania Press.
Jones, A. (1998) Body art. Performing the subject. Minnesota: Minnesota University Press.
Keenan-Ochs, E. (1974) ‘Norm-makers, norm-breakers: uses of speech by men and women in a
Malagasy community’, in Bauman, R. and Sherzer, J. (eds) Explorations in the ethnography of
speaking. Cambridge: Cambridge University Press, pp. 125–143.
Kessler, S. and McKenna, W. (1978) Gender. An ethnomethodological approach. Chicago: The
University of Chicago Press.
Kulick, D. (2005) ‘The importance of what gets left out’, Discourse Studies, 7(4–5), pp. 615–624.
Lorimer, H. (2011) ‘Walking: new forms and spaces for studies of walking’, in Cresswell, T. and
Merriman, P. (eds) Geographies of mobilities: practices, spaces, subjects. Farnham: Ashgate, pp.
19–34.
Mauss, M. (1973) ‘Techniques of the body’, Economy and Society, 2(1), pp. 70–88.
Maynard, D. W. (2003) Bad news, good news. Conversational order in everyday talk and clinical
settings. Chicago: The University of Chicago Press.
Milani, T. (2013) ‘Expanding the queer linguistic scene. Multimodality, space and sexuality at a South
African university’, Journal of Language and Sexuality, 2(2), pp. 206–234.
Moerman, M. (1988) Talking culture. Ethnography and conversation analysis. Philadelphia:
University of Pennsylvania Press.

287
Luca Greco

Mondada, L. (2011) ‘Le genre en action: la catégorisation des locuteurs comme production située des
participants dans l’interaction’, in Duchêne, A. and Moïse, C. (eds) Langage, genre et sexualités.
Québec: Editions Nota Bene, pp. 115–143.
Mondada, L. (2014) ‘Bodies in action: multimodal analysis of walking and talking’, Language and
Dialogue, 4–3(3), pp. 357–403.
Mondada, L. (2016) ‘Challenges of multimodality: language andthe body in social interaction’,
Journal of Sociolinguistics, 20(3), pp. 336–366.
Newton, E. (1972) Mother camp: female impersonators in America. Chicago: University of Chicago
Press.
Rivière, J. (1929) ‘Womanliness as Masquerade’, International Journal of Psycho-Analysis, 10, pp.
303–313.
Schechner, R. (2006) Performance studies: an introduction. New York: Routledge.
Schegloff, E. A. (1997) ‘Whose text? Whose context?’, Discourse and Society, 8(2), pp.165–187.
Schneider, R. (1997) The explicit body in performance. New York: Routledge.
Sormani, P., Carbone, G., and Gisler, P. (eds) (2018) Practicing art/science experiments in an emerging
field. New York: Routledge.
Speer, S. A. and Stokoe, E. (eds) (2011) Conversation and gender. Cambridge: Cambridge University
Press.
Wetherell, M. (1998) ‘Positioning and interpretative repertoires: conversation analysis and post-
structuralism in dialogue’, Discourse and Society, 9(3), pp. 387–412.

288
 19
Gender and sexuality normativities
Using conversation analysis to investigate
heteronormativity and cisnormativity in
interaction

Stina Ericsson

Introduction
Research informed by queer theory assumes ‘normality’ as the object of investigation, tak-
ing a critical approach to gender and sexuality normativities (e.g. Milani 2014: 261). Thus,
rather than seeing gender and sexuality as involving distinctions between ‘femininity’ and
‘masculinity’, or ‘homosexuality’ and ‘heterosexuality’, a queer theoretical stance sees the
very creation of such binaries and dichotomies, and the different values and connotations
being attached to various genders and sexualities, as that which is to be explained. Taking a
queer approach, this chapter explores the use of conversation analysis (CA) for investigat-
ing the (re)production of normative sexualities and genders in everyday spoken interactions.
Everyday interactions are interesting in this regard because they concern both individual
actions and societal discourses, and everyday norms may be conveyed in such unremarkable
ways that they are almost invisible – turning research into exciting detective work.
CA is informed by Garfinkel’s (1967) ethnomethodological approach, where the focus is
participants’ own understandings, that is, how participants themselves make sense of their
worlds, through social actions. Theoretically, the chapter takes a queer critical approach
through the notions of ‘heteronormativity’ and ‘cisnormativity’. Heteronormativity refers to
the ‘defining, normalising power of the heterosexual assumption, which marginalises other
sexualities at best, and invalidates them at worse’ (Weeks 2010: 92). Heterosexuality is here
taken as a sexuality which is specific to a particular time and place. Cisnormativity concerns
the normalising power of the cisgender assumption, that is, the idea that gender is coherent
over time, in perception, and through actions and identities. Just as for heterosexuality, I
take cisgender to be temporally and spatially specific, and as a notion that needs to be exam-
ined critically, rather than taken for granted or essentialised.
I begin this chapter by outlining the theoretical framework that underpins the approach
taken here – a framework which includes heteronormativity and cisnormativity, and which
places these two notions in relation to each other and to other normativities. Next, I intro-
duce relevant methodological tools and assumptions that underpin CA. I then illustrate the
theory and methodology by applying them to empirical examples of heteronormativity and

289
Stina Ericsson

cisnormativity in a case study of recorded interactions, where children and parents talk
about families and relationships. The chapter ends with a concluding discussion, together
with future directions and further reading.

Theoretical framework: intersecting normativities

As a way of specifying a queer critical approach to binaries and gender and sexuality nor-
mativities, this chapter adopts the genderism model of Hornscheidt (2012, 2015). This is
a framework that outlines different discriminatory structures which are related to gender,
including how they interact with and depend on each other. In the version of the model that
I adopt here, genderism has six different realisations: categorical gendering, binary gender-
ing, cisgendering, androgendering, heterogendering, and reprogendering. Of these, I will
focus on heterogendering and cisgendering in the case study below.
‘Categorical gendering’ is the idea that all human beings are gendered. All of the other
realisations of genderism are based on this idea, and it represents a powerful imperative of
seeing people, properties, behaviour, etc. as gendered. Assumptions of more genders than
two, or a gender system that incorporates genders that are not ‘female’ and ‘male’, all rely on
the idea that gender exists as a category. ‘Binary gendering’ means the assumption of two,
and only two, mutually exclusive gender categories – ‘female’ and ‘male’. This idea per-
meates dominant thinking in many contemporary societies around the globe, but balanced
against both historical and contemporary non-binary gendering, a critical stance towards
binaries is necessary in language, gender, and sexuality research. ‘Cisgendering’ corre-
sponds to the notion of cisnormativity that I introduced above, and it ‘constructs women
and men as coherent, natural genderings which are given at birth and which remain constant
throughout a person’s life’ (Hornscheidt 2015: 37). The coherence and consistency con-
straints involved in cisgendering marginalise certain trans bodies and experiences, and by
presupposing binary gendering, and thereby ‘continuously affirming it’ (Hornscheidt 2015:
37), cisgendering marginalises intersex and various non-binary identities and experiences.
‘Androgendering’ is the ‘universalisation and the norm-setting of men/a male norm’
(Hornscheidt 2015: 32), that is, it corresponds to the notion of ‘sexism’. ‘Heterogendering’
is Hornscheidt’s term for ‘heteronormativity’, which renders heterosexuality as the norm,
marginalising other sexualities. A powerful strategy of heterogendering is couple normativ-
ity, which is ‘the normalisation of a couple as the standard and desired form of living in
Western societies’ (Hornscheidt 2015: 34–35). Finally, ‘reprogendering’ means that repro-
duction is viewed as an essential part of a heterosexual couple’s experience.
These different realisations of genderism interact in different ways. For instance, repro-
gendering has different effects depending on whether someone is privileged or discriminated
against by androgendering. This may mean that people presenting as female are assumed
to be ‘(potential) mothers and/or daughters’ (Hornscheidt 2015: 35) in a way that people
presenting as male are not correspondingly assumed to be (potential) fathers or sons.
Having briefly introduced this theoretical framework of genderism, I next turn to the CA
method employed in this chapter.

Method: conversation analysis

Queer investigations of normativities can be approached using different kinds of material
and methods. The focus in this chapter is on social actions in everyday settings – how peo-
ple produce, reproduce, and transform gender and sexuality normativities through mundane

290
 Gender and sexuality normativities

actions. That is, how people through the routine and unremarkable things that they do in
their everyday lives, also create, adhere to, or change norms that govern gender and sexual-
ity. For instance, referring to one’s partner while booking a dentist’s appointment may chal-
lenge or reinforce heteronormativity.
For the analysis of such social actions in everyday settings, CA proves a highly valua-
ble tool, providing methods for both data collection and data analysis, enabling fine-grained
investigation of human actions at the micro level. While the focus of this chapter is on methods
for analysis, data collection methods are vital to any CA enterprise. In brief, CA scholars use
audio and video recordings of actual situated activities – which contrast with data collection
methods such as interviews and experiments – to gain insight into how participants them-
selves make sense of their social worlds. As part of this enterprise, CA methodology involves
the transcription of recorded audio or video data in ways that capture sufficient information
regarding the temporal and sequential unfolding of the interaction. However, it is important
to remember that the primary data are always the recordings, not the transcriptions, and the
researcher should always return to the primary data. CA involves listening to or watching
recorded data in tandem with transcriptions several times, giving an increasing understanding
of the interaction, leading to the identification of interesting phenomena and patterns.
CA methodology has been developed in tandem with theoretical assumptions about the
organisation of human interaction and social life. In fact, the approach takes social action as
fundamental, rather than language. A central concern is sequential organisation, referring to
the way in which social actions are always carried out in the context of preceding actions or
turns of talk. This means, for instance, that the meaning of a piece of talk can only be rec-
ognised through its sequential position, and not by looking at that piece of talk in isolation.
Relatedly, CA work has shown how meaning in interaction is co-constructed, incrementally,
by participants. Thus, any analysis of gender and sexuality norms in interaction will need to
consider the actions that the expression of these norms perform, in specific contexts of other
speakers and preceding talk.
One analytical consequence of CA’s ethnomethodological basis, is the requirement that
the analysis always be grounded in what occurs in the interaction. That is, in terms of gen-
der and sexuality studies, the analysis cannot freely use gender and sexuality categories to
interpret the interaction, but instead has to show how participants themselves make gender
and sexuality relevant. For instance, knowing that a participant self-identifies as ‘asexual’
is not enough to infer that the participant makes use of ‘asexual speech’. At a general level,
this is the issue of just what counts as context (Schegloff 1992), which has been seen as both
a strength and a weakness of CA. As a strength, it means that CA observations have a solid
scientific basis in the data at hand, staying clear of unwarranted speculations. As a weak-
ness, critics argue that relevant observations may be overlooked because wider contextual
knowledge is not made use of. This issue also turns on the CA notion of ‘relevance for
participants’ or ‘participants’ orientations’, that is, the analytical imperative of demonstrat-
ing what participants themselves make relevant in interaction. These issues of context and
participants’ orientations have generated productive methodological debates in language,
gender, and sexuality research (e.g. Kitzinger 2008; Stokoe and Smithson 2001. See also
Schegloff’s debates with Billig in 1999 – all contributions from both Schegloff and Billig
are freely available from the Schegloff Publications Archive online – and the chapter on
feminist conversation analysis in this volume).
Two methodological points will be made here regarding the use of ethnomethodologi-
cally grounded CA for the analysis of gender and sexuality. First, the analysis must indeed
always be grounded in participants’ own actions and interpretations. Without this, analysis

291
Stina Ericsson

runs the risk of ‘start[ing] out by “knowing” the [sexual] identities whose very constitu-
tion ought to have been precisely the issue under investigation’ (Kulick 2000: 265) or of,
for instance, reinforcing gender dichotomies. That is, the analysis runs the risk of yielding
uninteresting results or of misinterpreting or giving a skewed picture of the data.
Second, making sense of just what participants achieve in any given interaction may
involve making use of different kinds of analytical resources. One way of doing this is
by seeing wider knowledge of gender and sexuality as another layer that is added to the
analysis, such as the ‘post-analytic’ application of the term ‘heteronormativity’ to inter-
actional patterns found in data (Kitzinger 2008: 202–203). The argument here is that to
say something that is more specific and interesting from a gender and sexuality point
of view, ‘one must [sometimes] go beyond describing data in participants’ own terms’
(Stokoe and Smithson 2001: 232). It should be noted in this regard that while language,
gender, and sexuality studies, including queer theoretically informed studies, may use
CA in combination with various other methods and theories, other CA scholars may not
take favourably to this (e.g. Kitzinger’s 2008 debate with Wowk and the Schegloff–Billig
1999 debate in the Schegloff Publications Archive). In the case study below, I will make
use of Hornscheidt’s (2012, 2015) theoretical model of genderism as a way of adding
another layer to the analysis, thus going beyond describing data in participants’ own
terms, in order to further show how the data is interesting from a gender and sexuality
perspective.
A second way of making use of different kinds of analytical resources in order to make
sense of what participants do in interaction, is by considering the multiple things to which
participants may be orienting (Kitzinger 2008), and calling for more precision regarding just
what is an orientation and what is not (Wilkinson and Kitzinger 2014: 156). For instance,
participants in a medical help-line conversation orienting to requesting a home visit by the
doctor, may also be shown to be orienting to a cultural understanding of couples living
together, e.g. through a doctor asking a husband who’s calling on behalf of his wife, ‘where
do you live?’ and not ‘where does she live?’ (Kitzinger 2008: 200). When it comes to the
queer enterprise of investigating normativities, this issue of context and participants’ ori-
entations is highly pertinent, since norms in many ways are ‘invisible’ and may therefore,
at first glance, be expected to be something to which participants do not orient. However,
in the analysis of empirical examples below, we will see just how an ethnomethodological
approach can reveal participants’ orientations to normativities.
Given that gender and sexuality research brings certain theoretical perspectives to the
analysis, an important critique of CA’s tenet of approaching data without any preconcep-
tions, concerns the background knowledge that analysts nevertheless bring to the analy-
sis. For instance, Stokoe and Smithson (2001: 200) argue that ‘culture and common-sense
knowledge, of both members and analysts, are largely unacknowledged and unexplicated
resources in CA’. Thus, for language and gender scholars using CA methodology, it is piv-
otal to be explicit about background assumptions and theoretical knowledge. In this, self-
reflexivity is an important tool, just as a critical stance and self-reflexivity in relation to
theory is important for any language and gender research. In practice, the gender and sexu-
ality scholar that uses CA can employ such techniques as being open to what data shows
and collecting also cases that are troubling for one’s theoretical perspectives, perhaps even
specifically looking for such cases; working with one’s data in data sessions with different
groups of people, both within and outside of academia (the latter may require less technical
transcriptions); spelling out and critically examining one’s own assumptions and how one is
variously privileged or discriminated by different power structures.

292
 Gender and sexuality normativities

Turning now more concretely to the methodology of CA, this basically involves iden-
tifying and describing patterns in participants’ actions in interaction. In accordance with
ethnomethodology, analytical proof is found in the data itself: evidence for identified pat-
terns and what participants’ actions do, is located in what participants themselves do and
how they respond to each other’s actions. From this, two kinds of scientific proof method
can be formulated: the identification of cases that follow the pattern and cases that deviate
(Sidnell 2013). For instance, say that we wish to establish whether a question followed by
an answer constitutes an ‘adjacency pair’, that is, whether an answer is in general interac-
tionally expected subsequent to a question. Evidence indicating that a question–answer is
indeed an adjacency pair, can be found in ‘the great many cases’ (Sidnell 2013: 80) of ques-
tion–answer examples in data (cases that follow the pattern). Evidence can also be found in
examples where a question is not followed by an answer, but where either the questioner or
the non-answerer shows that an answer should have followed (deviant cases). Examples of
the latter that Sidnell mentions include the recipient of the question apologising for not giv-
ing an answer, or the questioner asking a follow-up question. Sidnell (2013: 80) argues that
deviant cases ‘often provide the strongest evidence for the analysis because it is here that we
see participants’ own orientations to the normative structures most clearly’.
Concluding from the above, the method to be illustrated in this chapter involves
CA-informed ethnomethodological analysis of interactional data, to identify what sequenti-
ality and participants’ orientations reveal about normative structures, where these normative
structures here concern genders and sexualities. The next sections are concerned with a case
study of gender and sexuality normativities in a data set of child-parent conversations.

Case study: normativities in interaction

The data that I use to illustrate the theory and method in this chapter, are drawn from the
research project ‘Daddy, daddy, child: linguistic negotiation of family, parenthood and rela-
tions in conversations between children and adults’ (Boyd and Ericsson 2017; Ericsson
2018). This Swedish project investigated children’s and parents’ conversations about fami-
lies, living arrangements, love, and marriage, specifically focusing on gender and sexuality
normativities. 24 children, mainly aged 5 to 8 years, participated in the project together with
their parents. Participating families included single mothers by choice through insemina-
tion/IVF (in vitro fertilisation), same-gender and different-gender parental couples, binary
and non-binary gender presentations and identities, parents cohabiting or not, as well as
urban–rural and geographical variation across Sweden.
Data were collected using a purpose-designed tablet application (‘app’), which was devel-
oped as part of the project. With the help of an interactive character called Moi, who speaks
using utterances recorded by a human child, the app asks questions of the children which
they are then required to think about and discuss with a parent. The app records the child–
parent conversations, both in the form of audio and through logging screen events, such as
what images the users select. The app was designed using a norm-critical perspective, which
involves taking a critical stance to norms in the sense of explicitly showing or challenging
norms which may otherwise typically remain hidden and taken for granted. In the app such
an approach was used to probe and challenge children’s views of gender and sexuality, for
instance through app characters with ambiguous gender presentations. The families used the
app in their homes whenever they wanted, without the presence of a researcher.
Traditional CA research places great emphasis on the recording of naturally occurring
activities, that is, activities that would have occurred even without the recording or the

293
Stina Ericsson

research study. This is motivated by CA’s interest in people’s ordinary lives and actions. CA
studies of naturally occurring activities have been successfully used for language, gender,
and sexuality research (e.g. Kitzinger 2005; Speer and Stokoe 2011). The use of the app for
generating the data in the present chapter meant stepping away from the requirement of nat-
urally occurring activities, in that these interactions only occurred because of the research
project. However, in several important ways, the participants in these interactions draw on
the same resources as in any other conversational setting, such as turn-taking. They also
access discourses on normative genders and sexualities. As Cameron et al. (1993) argue,
research data – in their case interaction between researcher and researched – constitute one
form of ‘normal communication’ that ‘provide[s] important insights into the way social
relations and identities are constructed through interaction’ (1993: 87, emphasis in original).
As a general remark, it is important to be aware of the specific conditions of one’s data col-
lection, and to critically examine the consequences of data collection methods for the data
at hand, and for the analysis. With regard to the data used for the case study in this chapter,
it may for instance be important to investigate how parents and children carry out their
conversations for an audience (the researcher) rather than as part of their own daily activi-
ties. It is also important to carefully consider one’s research design before the collection of
any data, in light of the kinds of research questions one wishes to address. Here, the present
research project had a focus on ideological norms rather than naturally occurring social
actions, which helped motivate the research design using the app.
All child–parent interactions with the app were transcribed in full in the project, but
using a rough transcription mainly aimed at capturing the content of participants’ contribu-
tions. Parts of the data were then transcribed in more detail, for specific studies and publica-
tions. In practice, the transcription of one’s recorded material is a matter of resources, as
transcription is highly time-consuming. All families spoke Swedish with the app, with the
exception of one family who mainly used English.
I will now turn to a ‘hands-on’ illustration of the use of CA, first analysing sexuality nor-
mativities in the data, and then cisgender normativities. These two sections are structured
in the same way: I first consider evidence from cases that follow the pattern, then evidence
from deviant cases, and finally cases that challenge the normativities. This is typically not
the way such an analysis would be presented in a research paper, but it is used here to expli-
cate the method as clearly as possible. I also want to emphasise that it is not a matter of a
complete analysis, but rather intended to illustrate the application of theory and method.
Names of all participants and the people they talk about, as well as some other personal
details, have been anonymised in the transcriptions.

Analysing sexuality normativities

This section investigates sexuality normativities in the data, addressing the overarching
questions of which sexuality norms are produced by the participants, and how these norms
are interactionally achieved. The analysis of sexuality normativities will here focus on par-
ticipants’ replies to Moi’s question ‘Do you know someone who’s in love?’, where Moi is
the fictional app character. An inductive, bottom-up analysis is used to make claims about
the particular data at hand.
Translations into English are provided on a separate line, and a transcription key can be
found at the end of the chapter.

294
 Gender and sexuality normativities

Evidence from cases following the pattern

Consider the following example, where Oskar and his parent talk to Moi:

Example 1. ‘Henry and Edith’

593 Moi: vet du nån som e kär

do you know someone who’s in love
594 Osk: jao ((smackljud)) Henry och Edith dom e kära
yieah ((lip smack)) Henry and Edith they’re in love
595 Par: m hm:: ha dom sagt de eller
m hm:: have they said so or
596 Osk: Henry ha sagt de ti mej
Henry has said so to me
597 Par: m hm::

In this example, Oskar answers the yes/no question of whether he knows someone who’s in
love – the Swedish nån (‘someone’, line 593) formally requesting just one person – by giv-
ing an example of two people he knows, ‘Henry and Edith’ (line 594). We learn elsewhere
in the conversation that they are friends of Oskar’s from school.
In his reply, Oskar describes a couple: Henry and Edith. The mention of a couple indi-
cates an understanding of being in love as a relation that takes an object (cf. if Oskar had
replied ‘yieah Henry he’s in love’), and only one object (cf. ‘yieah Henry, Edith, and Fred
they’re in love’). Further, the couple that Oskar names is a different-gender couple. What
Oskar does, then, I argue, can be seen as an instance of what Kitzinger (2005) describes as
the production of heterosexual couples, or as an example of Hornscheidt’s (2015) couple
normativity as a strategy of heterogendering.
In line 595, the parent acknowledges Oskar’s reply through ‘m hm::,’ and then poses a
question. The question concerns the type of evidence that Oskar has for his claim, but does
not question the idea of Henry and Edith as a couple, or of different-gender couples, as such.
As another example, consider Stella’s reply:

Example 2. ‘You and mummy’

052 Moi: vet du nån som e kär

do you know someone who’s in love
053 Ste: många ☺du å mamma☺ o::ch asså de=e så många
many ☺you ’n’ mummy’☺ a::nd well it’s so many
054 Par: okej
okay
055 Ste: Jessica å Davi:::d o::ch
Jessica ’n’ Davi:::d a::nd
[removed talk about how Stella sits on her chair]
058 Ste: (å sen) Emmas mamma å pappa å Susans mamma å pappa
(’n’ then) Emma’s mum ’n’ dad ’n’ Susan’s mum ’n’ dad
059 Par: e de bara mammor å pappor som e kära
is it just mummies ’n’ daddies who’re in love

295
Stina Ericsson

060 Ste: och barn kan me vara kära

and children can also be in love
061 Par: okej
okay

Stella gives several different couples as part of her response (lines 53, 55, 58). Her response
is acknowledged by the parent, explicitly through ‘okay’ (line 54), and implicitly through a
follow-up question (line 59). All couples are different-gender couples (the ‘you’ in line 53
referring to Stella’s pappa (daddy)).
The parent’s question in line 59 concerns whether anyone else can be in love, apart from
‘mummies ’n’ daddies’. Stella understands this as a generational issue, and states that chil-
dren can also be in love, which the father agrees to, possibly indicating that Stella gave an
expected reply (lines 60–61).
Regarding participants’ orientations, as Kitzinger’s (2008) analysis of orientation to cul-
tural norms shows, such an analysis relies on inferences and requires careful and critical
analysis of just what it is that participants treat as common knowledge. For the two brief
examples considered here, it can be argued that Oskar, Stella, and their parents orient to a
cultural norm of different-gender couples as part of their understanding of being in love.
Also, participants’ understanding is co-constructed by the children and parents, through
Oskar and Stella describing the couples in their replies to Moi’s questions, and the parents
acknowledging these replies and producing follow-up questions.
In the data, there are ‘great many cases’ (Sidnell 2013: 80) of different-gender couples
being described in response to the question of knowing someone in love, similar to the two
examples which we have seen above. This is one type of argument for seeing such examples
as cases that follow a pattern.
Summing up, cases that follow the pattern of sexuality normativities in the data involve
participants producing and treating different-gender couples and living arrangements as the
norm. We have seen two examples in some detail here, and the analysis as a whole would
build upon many other, and different kinds of, examples in the data.

Evidence from deviant cases

The two examples above with Oskar and Stella, respectively, can be contrasted with the
following:

Example 3. ‘Kasper and Vincent’

247 Moi: vet du nån som e kär

do you know someone who’s in love
248 Lud: eh:: två
eh:: two
249 Par: vilka då
who
250 Lud: Kasper å Vincent
Kasper ’n’ Vincent
251 Par: e dom kära i varann
are they in love with each other

296
 Gender and sexuality normativities

252 Par: nähä vilka e dom kära i då

no okay who are they in love with then
253 Lud: nåra i en annan klass
some in the other class
254 Par: jaha
okay
255 Par: e de tjejer eller
are they girls or
256 Lud: m
257 Par: ha
okay

Highly reminiscent of the examples that follow the pattern – such as Examples 1 and 2
above – Ludwig here uses a conjunctive structure, ‘Kasper ‘n’ Vincent’ (line 250), in reply
to the question of knowing someone in love. However, unlike the parents’ next contribu-
tions in those two other examples, the subsequent contribution by Ludwig’s parent ques-
tions the very coupledom of Kasper and Vincent (line 251). From the ensuing contributions
by Ludwig and his parent, lines 252–257, it is clear that Kasper and Vincent are not to be
understood as a couple in love with each other (presumably Ludwig shakes his head or gives
some other non-verbal response, between lines 251 and 252).
What makes this a deviant case, then, is partly seen from sequential evidence, in the par-
ent’s response to the child’s answer, and the participants’ mutual meaning construction of
‘Kasper ‘n’ Vincent’, despite formal appearances, as not a couple. Additionally, one thing
that differs between Ludwig’s conjunctive structure and those used by Oskar and Stella
above, is the inclusion of two people of the same gender in Ludwig’s utterance. One of the
ways in which heterogendering works is by making sexualities other than heterosexuality
invisible. It may be this mechanism that makes the parent question a couple reading here.
As possible further support of sexuality and gender being at issue here, gender is explic-
itly brought up by the parent in line 255, and Ludwig (line 256) confirms that Kasper and
Vincent are in love with girls.
Another sequential deviation in Example 3 comes in the position immediately follow-
ing Moi’s question (here line 248). As can be seen from Examples 1 and 2, this is where an
answer is given to Moi’s question, enumerating exemplifying couples. Ludwig’s response
differs from this, giving instead the number of people that he knows are in love, ‘two’. This
answer may also be what contributes to the parent’s inference that ‘Kasper ‘n’ Vincent’ are
to be understood as two people in love with others and not each other.
In this way, Examples 1–3 have given an illustration of how a collection of exam-
ples that are similar in some way – in this case through being the utterances that follow
a specific question – can be put together and examined, revealing normative structures
through cases that follow and deviate from an identified pattern, respectively. There are
also several other kinds of examples in the data that can be examined for what they reveal
about heterogendering through deviant cases. One type of example involves participants
expressing breaches to their expectations, such as Ebba exclaiming ‘WHAT can you get
married to a girl’ when her parent probes who she wants to get married to. Another type
of example is discriminatory or negative utterances, such as Hannes expressing that two
male-presenting characters in the app cannot get married or that a boy going out with a
boy is ‘odd’.

297
Stina Ericsson

Challenges
Based on evidence from cases that follow and deviate from a pattern, it can thus be argued
that heterogendering is prevalent in the data. However, there are also other kinds of exam-
ples, which complicate the picture. Consider Example 4:

Example 4. ‘Agnes and Vera’

201 Moi: vet du nån som e kär

do you know someone who’s in love
202 Mor: Agnes å Vera
Agnes ’n’ Vera
203 Par: du få prata ordentligt så dom hör
you need to speak properly so they can hear

This example follows the pattern of the child – here Morgan – describing a couple in the
second-pair part (Stivers 2013) of the question–answer adjacency pair (line 202). It can also
be said to follow the pattern of the parent not questioning the couple as such (line 203),
meaning that Morgan and the parent interactionally co-construct the couple as unremark-
able. The couple here is a same-gender couple, involving Morgan’s sister Agnes. In other
parts of the data, Agnes herself also mentions Vera as the person that she is in love with, and
this is also interactionally conveyed as something ordinary. Similarly, Siri and her father
Jesper jointly construe ‘daddy Peter’ as the person that Jesper is in love with, and Siri also
mentions ‘Vilgot’s mummies’, among other couples, as being in love.
What do such examples mean for heterogendering? Would it be inaccurate to connect
empirical evidence to the theoretical notion of heterogendering here? Well, yes and no. No,
because of the great number of normatively conveyed other-gender assumptions that permeate
the data, attesting to the pervasiveness of heterogendering. Here, actual numbers of same-gen-
der and other-gender couples in the data set as a whole could be stated to support the argument.
Yes, because it would not give the full picture of the data. Considering Examples 1–4 again,
in light of the whole material, what comes across even more strongly than heterogendering is
couple normativity, irrespective of sexuality or gender constellations. That is, in the data as
a whole (based on many more cases than Examples 1–4), the couple as a way of organising
love, sexuality, living arrangements, etc., is superordinate to, and can thus be theoretically
partly detached from, heteronormativity. Then, given the interdependence between different
realisations of genderism as shown by Hornscheidt (2012, 2015), couple normativity can be
expected to interact with heteronormativity in different ways.
As a brief conclusion of the analysis of sexuality normativities, my aim here has been to
show how the identification of cases following and deviating from a pattern can be useful
analytical tools, and also how micro-level actions by participants can be connected to macro-
level language, gender, and sexuality theories. At the same time, my aim has also been to
show how a critical approach, investigating challenges to one’s analysis, is necessary.

Analysing cisgender normativities

Having shown how an analysis of sexuality normativities can be carried out, I will here
consider more briefly an analysis of cisgender normativities, addressing the overarching
questions of which cisgender norms are produced by the participants, and how these norms
are interactionally achieved.

298
 Gender and sexuality normativities

Evidence from cases following the pattern

In the example below, Siri and her friend Felix are doing the app activity of giving Moi a
family, by choosing from a large set of characters that are shown visually in the app (but
otherwise not verbally described in any way by the app). Having chosen a set of characters,
they talk about the ages of these fictional family members:

Example 5. He and she

205 Moi: hu gamla ä vi

how old are we
206 Sir: [m::: ]
207 Fel: [eh: han] ska va kanske:
[eh: he] should be maybe:
208 Sir: ksch arton
ksch eighteen
209 Fel: nä @ tju- [tjuge::]
no @ twe- [twenty::]
210 Sir: [@ @ ]
211 Sir: [a]
[yeah]
212 Fel: [nej] trett- tretti:nie
[no] thir- thirty:nine
213 Sir: a trettinie [å hon]
yeah thirtynine [’n’ she]
214 Fel: [å hon] ska va trettiåtta
[’n’ she] should be thirtyeight

Siri’s and Felix’s use of gendered pronouns to refer to the characters (lines 207, 213, 214)
attest to their assumption of coherent genders as something that can be read off their percep-
tion of the visually presented characters. Coherence over time can be seen through partici-
pants’ repeated use of the same pronoun for the same character on different occasions when
using the app (not shown in the extract). In this way, Siri, Felix, and other participants can
be seen to be orienting to an unnamed and taken-for-granted norm (cf. Kitzinger 2008) of
coherent gender.
The prevalence of gendering, and thereby the continuous reproduction of gender as a rel-
evant category, is further seen through numerous examples in the data (not shown in this chap-
ter). For instance, participants use gendered pronouns, gendered proper names, and gendered
family roles such as ‘mummy’ and ‘daddy’ in their utterances. These also reproduce a binarily
gendered worldview. Thus, categorical and binary gendering are both firmly attested to in the
data. Cisgendering presupposes both categorical and binary gendering, and further involves the
assumption of gender coherence over time, in perception, and through actions and identities.
It might be argued that pronoun gendering is unavoidable in languages like Swedish and
English – as opposed to e.g. languages like Finnish where the single third-person pronoun
hän does not encode gender – that is, that the language system itself constructs categorical,
binary, and cisgendering. However, language use always involves choice, and analysing
the linguistic choices that language users make is precisely what underlies discourse and
interaction analysis. Alternatives for Siri and Felix here could have been to use referring
expressions like ‘this/that one’.

299
Stina Ericsson

Evidence from deviant cases

In interaction with a parent, Matilda asks ‘but why does she wear a tie’, and ‘why doesn’t
she wear a dress’, about a bride character in the app. These two examples involve cisgender
incoherence being constructed as a clash between ‘gender assignment’ (‘she’) and ‘gender
presentation’ (‘wear a tie’, ‘not wear a dress’), using terms from Zimman (2015). A similar
example is Marika asking her little sister Sara ‘why did you say daddy to a girl’, holding
Sara accountable for what Marika sees as a clash between the ‘gender role’ of ‘daddy’ and
the ‘gender identity’ of being a ‘girl’. ‘Accountability’ (Scott and Lyman 1968) is the idea
that it is the unanticipated and non-normative which needs to be explained, and requests for
accountability in relation to gender constitute deviant cases which give good evidence for
participants’ orientations to normative structures regarding cisgender normativities.
Consider next the following exchange, which occurs a few turns after Example 5:

Example 6. She you mean

222 Sir: å HA:N ska va::

’n’ HE: should be::
223 Fel: eh hon menar du
eh she you mean
224 Sir: ho[n:]
sh[e:]
225 Fel: [h-] hon ska va
[sh]- she should be
226 Sir: typ tre
like three

In this example, Felix initiates a repair sequence in line 223, where ‘repair’ is the CA term
for participants’ treatment of problems in interaction which have to do with hearing and
understanding, and the like. In Example 6, the repair concerns the gender of the character
they are talking about. Felix’s utterance conveys a perceived gender incoherence between
Siri’s assignment of ‘he’ and the presentation, role, or identity that Felix assumes for the
character. Thus, in addition to accountability, repair of assigned gender is another way in
which deviant cases regarding gender attest to cisgender normativities.

Challenges
In the following example, Gabi is talking to their parent:

Example 7. Mapa

061 Moi: who are the people in your family

062 Par: can you say
063 Gab: ˚ma:pa:˚
064 Par: say it a little bit louder
065 Gab: u:o[: ]
066 Par: [who] am I
067 Gab: ma- mapa
068 Par: and who else is in our family

300
 Gender and sexuality normativities

069 Gab: mama

070 Par: and who else
071 Gab: Cam
072 Par: okay

Gabi here uses the term ‘mapa’ for their parent (lines 63, 67). This is a term created from the
Swedish ‘mamma’ and ‘pappa’, so in one sense it builds upon binary gendering, but in another
sense, it challenges it, by naming something outside of the binary. Cisgender being definition-
ally restricted to a binary system, ‘mapa’ then also challenges cisgendering. Thus, this exam-
ple constitutes a challenge to the cisgender normativities shown by the other examples above.
In a general way, challenges to cisgendering would involve examples of participants embrac-
ing gender incoherence in various ways. Examples of categorical gendering being challenged
exist in the data, in the form of participants using non-gendered expressions such as ‘that per-
son’ or ‘that/this one’, that is, here participants do not assign gender at all. Such examples can
perhaps also be analysed as challenging cisgendering, as cisgendering relies on gender being
assigned. An interesting case in this regard is also utterances where ‘he or she’ is used with a
specific reference, such as the Moi character. Such a referring expression seems to allow for
gender fluidity (while categorical and binary gendering remain intact). However, such examples
in the data are typically offset and overruled by the same app characters being gendered on other
occasions in the interactions. Examples that more explicitly and firmly embrace gender incoher-
ence and fluidity are rare in the data, with the use of ‘mapa’ in one family as an exception.

Conclusion
This chapter has employed the notions of ‘heteronormativity’ and ‘cisnormativity’ in the
enterprise of using ethnomethodologically grounded CA for the analysis of gender and sex-
uality normativities. Evaluating this approach, there are several distinct advantages to using
CA for gender and sexuality research in general and normativities in particular: it shows
what people are doing rather than what they believe they, or others, are doing; it can reveal
how discourses at the societal macro level are formed, used, and challenged at the interac-
tional everyday micro level; it can make seemingly invisible norms visible and thereby help
denaturalise the seemingly natural orders of cisgendering and heterogendering. By using
CA, the gender and sexuality scholar can also take full advantage of several decades’ work
on interactional practices in other areas. Disadvantages of CA as a method include practi-
cal obstacles: transcription is highly time-consuming, and, as with any field work, finding
participants and obtaining access may be difficult.

Future directions
In recent years, multimodality and embodiment have received increasing attention in CA
research, with studies of gaze, gesture, bodily posture, and movement, in relation to objects
and physical space. This can be expected to develop as a highly productive line of investi-
gation also for language, gender, and sexuality research, with detailed micro-level analyses
of, for instance, embodied gender expressions and assignments in specific interactional con-
texts; embodied interactions in sexual encounters; and gender and sexuality normativities
conveyed in embodied ways.
An area which has received increasing attention in language, gender, and sexuality
research in recent years involves gender and sexuality non-binaries. Here, queer-theoretically

301
Stina Ericsson

informed CA studies ought to be able to generate significant new knowledge of human gen-
ders and sexualities in practice, such as through analyses of how binary norms are upheld
and challenged in various ways in different interactional settings.

Transcription conventions

((…)) meta-comments and non-verbal actions

☺…☺ said while laughing or smiling
: lengthening of preceding sound
(…) uncertain transcription
@ laughter, each token marks one pulse
[… speech produced in overlap
°…° quiet voice
…- interruption
word emphasis
WORD strong emphasis

Further reading
Sidnell, J. and Stivers, T. (eds) (2013) The handbook of conversation analysis. Chicester:
Wiley-Blackwell.
This edited volume is a comprehensive guide to all aspects of CA method and theory.
Wilkinson, S. and Kitzinger, C. (2014) ‘Conversation analysis in language and gender studies’, in
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) Handbook of language, gender, and sexuality, 2nd
edn. Chichester: Wiley Blackwell, pp. 141–160.
This chapter deals with CA as a method for language and gender research, focusing on interactional
practices such as turn-taking and interruption.
http://emcawiki.net
This site contains a continuously updated bibliography database of CA publications.
Zimman, L., Davis, J. L., and Raclaw, J. (2014) Queer excursions: retheorizing binaries in language,
gender, and sexuality. Oxford: Oxford University Press.
This edited volume gives critical and complex analyses of gender and sexuality binaries, showing
both how binaries are sometimes needed for analytical purposes and how, in other settings, binary
understandings would be erroneous.

Related topics
Sexuality as non-binary: a variationist perspective; gender diversity and the voice; an ethnographic
approach to compulsory heterosexuality; feminist conversation analysis: examining violence against
women; applying queer theory to language, gender, and sexuality research in schools.

References
Boyd, S. and Ericsson, S. (2017) ‘Children’s ongoing and relational negotiation of informed assent in
child-researcher, child-child, and child-parent interaction’, Childhood, 24(3), pp. 300–315.

302
 Gender and sexuality normativities

Cameron, D., Frazer, E., Harvey, P., Rampton, B., and Richardson, K. (1993) ‘Ethics, advocacy and
empowerment: issues of method in researching language’, Language & Communication, 13(2),
pp. 81–94.
Ericsson, S. (2018) ‘The language of cisnormativity: children and parents in interaction with a
multimodal app’, Gender & Language, 12(2), pp. 139–167.
Garfinkel, H. (1967) Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Hornscheidt, L. (2012) feministische w:orte: Ein Lern-, denk- und handlungsbuch zu Sprache und
Diskriminierung, gender studies und feministischer Linguistik. Frankfurt am Main: Brandes &
Apsel.
Hornscheidt, L. (2015) ‘Trans_x_ing linguistic actions and linguistics’, in Magnusson, J., Milles, K.,
and Nikolaidou, Z. (eds) Könskonstruktioner och språkförändringar: rapport från den åttonde
nordiska konferensen om språk och kön. Huddinge: Södertörn University, pp. 29–46.
Kitzinger, C. (2005) ‘“Speaking as a heterosexual”: “Speaking as a Heterosexual”: (How) does
sexuality matter for talk-in-interaction?’, Research on Language & Social Interaction, 38(3), pp.
221–265.
Kitzinger, C. (2008) ‘Developing feminist conversation analysis: a response to wowk’, Human
Studies, 31(2), pp. 179–208.
Kulick, D. (2000) ‘Gay and lesbian language’, Annual Review of Anthropology, 29(1), pp. 243–285.
Milani, T. M. (2014) ‘Queering masculinities’, in Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds)
Handbook of language, gender, and sexuality, 2nd edn. Chichester: Wiley Blackwell, pp. 260–278.
Schegloff, E. A. (1992) ‘In another context’, in Duranti, A. and Goodwin, C. (eds) Rethinking context:
language as an interactive phenomenon. Cambridge: Cambridge University Press, pp. 191–227.
Scott, M. B. and Lyman, S. M. (1968) ‘Accounts’, American Sociological Review, 33(1), pp. 46–62.
Sidnell, J. (2013) ‘Basic conversation analytic methods’, in Sidnell, J. and Stivers, T. (eds) The
handbook of conversation analysis. Chichester: Wiley-Blackwell, pp. 77–99.
Speer, S. A. and Stokoe, E. (2011) Conversation and gender. Cambridge: Cambridge University Press.
Stivers, T. (2013) ‘Sequence organization’, in Sidnell, J. and Stivers, T. (eds) The handbook of
conversation analysis. Chichester: Wiley-Blackwell, pp. 191–209.
Stokoe, E. H. and Smithson, J. (2001) ‘Making gender relevant: conversation analysis and gender
categories in interaction’, Discourse & Society, 12(2), pp. 217–244.
Weeks, J. (2010) Sexuality, 3rd edn. Abingdon: Routledge.
Wilkinson, S. and Kitzinger, C. (2014) ‘Conversation analysis in language and gender studies’, in
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds) Handbook of language, gender, and sexuality, 2nd
edn. Chichester: Wiley Blackwell, pp. 141–160.
Zimman, L. (2015) ‘Transmasculinity and the voice: gender assignment, identity, and presentation’,
in Milani, T. M. (ed) Language and masculinities: performances, intersections, dislocations. New
York: Routledge, pp. 197–219.

303
 20
Examining girls’ peer
culture-in-action
Gender, stance, and category work in girls’
peer language practices

Ann-Carita Evaldsson

Introduction
This chapter outlines how an ethnomethodological approach to gender as a reoccurring
achievement of situated conduct (West and Zimmerman, 1987, 2009) and an emergent
property of interaction (Benwell and Stokoe 2016; Speer and Stokoe 2011) make possible
a dynamic view on gender among girls. The methodological starting point is how gen-
der and gendered norms of interaction are co-constructed and constituted among girls in
everyday peer-group interaction through their participation in language practices such as
gossip, disputes, and story-telling (Goodwin and Kyratzis 2011, 2014 for overviews). The
analytical focus is on how repeated performances of gossip disputes provide young girls
with resources for “doing gender” and strengthen feminine relations, while making others
accountable for gender in-appropriate conduct.
Following an ethnomethodological approach from Garfinkel (1967), the analytical point of
departure is the ‘doing of gender’ as a locally managed practical accomplishment (West and
Zimmerman 1987, 2009). As West and Zimmerman (1987: 125) note in their classical paper
‘doing gender involves a complex of socially guided perceptual, interactional, and micropo-
litical activities’, that casts pursuits of expressions as either gender-appropriate or, as in the
case, as gender-inappropriate, that is accountable. In this perspective the doing of gender is
a members’ phenomenon in the sense that accountability to normative conceptions of gender
involve ‘observable practices in interaction’ (West and Zimmerman 2009: 115). That is ‘to
do gender is not always to live up to normative conceptions of femininity or masculinity; it
is to engage in behavior at the risk of gender assessment’ (West and Zimmerman 1987: 136).
In the analytical section an ethnomethodological conversation analytical approach to
membership categorisation sequence work (Benwell and Stokoe 2016) combined with a
multimodal interactional approach to stance is used to explore how gender is accomplished
in girls’ peer group interaction (Evaldsson 2007; Goodwin 2007, 2011; Goodwin and Alim
2010). Sacks’ work on membership categorisation analysis (MCA; Sacks 1972, 1995; Stokoe
2006) is now a well-established approach in studies of gender and discourse both in research
on children’s peer language practices (Goodwin and Kyratzis 2011, 2014) and in feminist

304
 Examining girls’ peer culture-in-action

conversation analysis (Speer and Stokoe 2011: see chapter by Tennent and Weatherall in this
volume). A crucial feature of membership categorisations is that they both are constitutive
of and reflect conventional expectations of normative behaviours within a specific group
and culture (Evaldsson 2007). As will be further shown in the analytical section gendered
categories such as ’girls’ and ’boys’ are not used explicitly in girls’ gossip dispute activities;
instead, the co-participants make use of a variety of ways to describe, evaluate, and catego-
rise co-participants such that gender along with other social categories are made relevant in
the course of interaction. Thus, for MCA, the analyst necessarily draws on extra-contextual
resources in the form of cultural knowledge of the participants and the setting to explicate
the sense-making orientations of the participants (Evaldsson 2007; Speer and Stokoe 2011).
In this particular study, ethnographic knowledge of the girls, their interactional history, and
the social arrangements of groupings in school is used for examining girls’ peer culture
in action through an examination of members’ methods of doing gender along with other
social categories are accomplished in interaction.
In order to capture how repeated performances of particular kinds of stances and cat-
egories are habitually and conventionally associated with particular subject positions
(associated with gender, ethnicity and/or class including relations of power) an ethnometh-
odological approach to doing gender is combined here with sociolinguistic work on stance
(Bucholtz 2009; Jaffe 2009). Thus instead of treating girls’ participation in gossip disputes
as intuitively linked to a fixed form of feminine behaviour, the focus is on gender as medi-
ated by stance to explore the indexical mediation of language practices and social identities
(Ochs 1992). As argued by Jaffe (2009: 8), in taking up stances speakers ‘project, assign,
propose, constrain, define, or otherwise shape the subject positions of their interlocutors’.
Drawing in particular on Goodwin’s (2006, 2011; Goodwin and Alim 2010) ethnographic
and multimodal interactional approach to girls’ peer language practices, the analytical focus
is on how embodied acts of stances including both talk and the body are performed in inter-
action for displaying alignment/and or disalignment as well as for taking up oppositional
and/or multiple positions vis-à-vis co-present as well as absent others. The sociolinguistics
of stance situates also linguistic acts of stance within the sociocultural framework that give
stances their social meanings (Jaffe 2009). By combining ethnomethodological perspectives
on gender with a sociolinguistic approach to stance, I will demonstrate how at an analytical
level, repeated performances of stances and category-work, indexing inappropriate femi-
nine behaviours, co-exist with, and become woven into the local moral and social business
of the interaction at hand.

Research on gender and peer language practices

Ethnographically based research on gender and language in children’s peer language prac-
tices have demonstrated how children’s participation in storytelling, gossip, disputes, teas-
ing, and insults is tied to the accomplishment of social selves. Children co-construct their
social world, negotiating gendered relations and social boundaries, evaluating self and oth-
ers, creatively transforming category memberships and gendered norms in their social life
with peers (Goodwin and Kyratzis 2011 for an overview). In line with the shift that has
occurred in the field of language and gender, this implies a shift from binary gendered
norms and differences in boys’ versus girls’ behaviours and language to a focus on diversi-
ties of gender identities and variability of language practices (Goodwin and Kyratzis 2014
for an overview). The approach taken is underpinned by performative and constructionist

305
Ann-Carita Evaldsson

approaches to gender and social interaction (Bucholtz 2009; Garfinkel 1967; West and
Zimmerman 2009). What emerges is a multifaceted and dynamic view of how girls (and
boys) perform and do gender in moments of peer group interactions that problematises easy
gender dichotomies (see for example Evaldsson 2002, 2007).
In this section I discuss some findings on what makes gossip disputes pervasive features
of girls’ language practices and social life with peers (see Eder 1995; Evaldsson and Svahn
2012, 2017; Goodwin 1990, 2006; Mendoza-Denton 2008; Shuman 1986). Particularly
Goodwin (1990, 2006) has demonstrated through ethnographic and multimodal interac-
tional analyses how girls across groupings engage in extended and complex types of opposi-
tional stances in gossip dispute sessions in ways that refute simplistic notions of all females
as non-competitive or non-aggressive. Building on long-term ethnographic work combining
conversation analysis and Goffman’s (1981) notion of footing, Goodwin (1990) shows how
a group of urban African American neighbourhood girls engage in he-said–she-said gossip
events that reorganise and realign the social organisation. The girls’ instigated complex
forms of gossip-dispute stories directed towards an absent target, leading to future confron-
tations of the offending party. The findings can be compared with Shuman’s (1992) early
linguistic ethnographic work on everyday storytelling among working-class girls (African
American, European American, and Puerto Rican) in a junior high school. Shuman found
for example that groups of girls engaged in extended gossip-telling activities leading to
upgraded disputes and possible fights directed towards an instigator that were linked to enti-
tlement to talk about someone else’s offences. In her linguistic ethnographic study of Latina
gang girls in Northern California, Mendoza-Denton (2008) discusses how a form of gos-
sip, here named ‘talking shit’, is used among girls both to portray others unfavourably and
to brag about own fights. In addition Eder (1995) demonstrates in her micro-ethnographic
work in middle schools how groups of working-class girls engage in aggravated forms of
gossip disputes, indexing toughness and assertiveness as central for their gender.
In more recent ethnographic studies of girls’ peer group interaction in a multi-ethnic
school setting Evaldsson and Svahn (2012) demonstrate how a friendship group of girls’
instigate gossip disputes to sanction offensive behaviours of a targeted girl for calling on
adults for help (Evaldsson and Svahn 2012). Through a range of collectively and repeatedly
performed derogatory category descriptions and embodied actions, the target was cast as ‘a
snitch’ and then confronted for being a familiar ‘offender’. In a previous article (Evaldsson
and Svahn 2017), based on the same corpus of data as this study, we have explored the
organizing force and normative character of anger and aggression in how girls constitute
their participation in gossip disputes culminating in fights. Goodwin (2006, 2007, 2011) also
shows in her ethnographic work how girls engage in forms of social aggression in gossip
dispute and insult sessions to exclude and sanction members who want to socialise with
them. A group of white middle-class American girls constantly engaged in negative evalu-
ative acts of degradation to build power asymmetries towards a girl of African American
and low-income background, indexing the girls’ own upper-middle-class status as valued
(Goodwin and Alim 2010).
These and other findings demonstrate the importance of exploring the dynamic and
multifaceted nature of how feminine gender identities are co-produced as girls differenti-
ate themselves from other girls in peer interaction when the moral and social order is at
stake. Moreover it shows the need for long-term ethnographic analyses to explore how
broader social categories such as gender along with ethnicity, age, language, social class,
friendship, and notions of power are enacted in girls’ peer interactions both in schools and
neighbourhoods.

306
 Examining girls’ peer culture-in-action

Ethnographic analyses of girls’ gossip disputes

The data in this chapter draw on a long-term ethnographic study based on video record-
ings (108h) of children’s everyday peer language practices in a low-income multi-ethnic
school setting in Sweden. The data was collected by a former doctoral student (see Svahn
2012), as part of a research project on school bullying (Evaldsson and Svahn 2012, 2017).1
Through a long-term ethnographic fieldwork during one school year, it became possible for
the fieldworker to identify, access, and document with video, reoccurring language practices
in children’s friendship-groups. The linguistic fieldwork that focused on girls from two fifth
grade classrooms provided rich ethnographic knowledge of the girls’ language practices
in instances where there were no other adults present. Most often girls and boys were in
each other’s presence in the schoolyard, while they mainly socialised in same-sex groups
with other children that they know from school. The gossip disputes that are in focus in the
analytic section were initiated by one of the most popular groups of girls in fifth grade. The
girls often created micro-dramas in the schoolyard to deal with real and imagined offences
of particular girls (compare with Goodwin and Alim 2010; Shuman 1986). The girl group
in focus consists of seven 11-year-old girls (Yaasmiin, Salina, Tara, Emine, Beyan, Azra,
Samina) with various ethnic backgrounds (Somali; Eritrean; Kurdish; Palestinian) and with
a long relational history of socialising with one another. While it is possible to identify
the girls’ different ethnic and social backgrounds, these identity categories were seldom
articulated in the girls’ interaction. However long-term data collection made it is possible to
show how gender along with ethnicity and social class are co-constructed in girls’ everyday
interaction.

Analysing activity trajectories of gossip disputes

The interactional analysis traces an activity trajectory of video-recorded gossip dispute
activities in which members of the popular peer group stage a micro-drama where they
report about and take action against offences of a particular girl (see also Evaldsson and
Svahn 2012, 2017). Drawing on ethnomethodological membership categorisation analysis
in sequences of actions (Sacks 1992; Stokoe 2006), the focus is on how the co-partici-
pants do gender as they make normative assessments that link ‘category-bound actions’
and ‘identity-categories’ with the person being assessed (Evaldsson 2007; Goodwin 2011).
The methodology of extended fieldwork combined with an ethnomethodological and mul-
timodal conversation analytical approach to stance was critical for the analysis (Goodwin
2006). Building on Goodwin’s (2000) multimodal interactional approach to participation
the analysis focuses on how co-participants make use of different modalities (talk, prosody,
and the body) to perform actions and identities in interaction. The fact that video-recordings
provide repeated possibilities for listening to and viewing the participants’ embodied actions
and language practices was important for transcribing and analysing the data.
The video excerpts are transcribed following conventions developed within CA (see tran-
scription conventions) combined with images from video-recordings. The English transla-
tions are as close as possible to the Swedish verbatim records. To ensure the participants’
anonymity, all names have been replaced with pseudonyms. The convention of single quota-
tion marks indicates the contrastive membership categories (‘fighting girl’ versus ‘coward’)
and category-bound actions (‘fighting back’ versus ‘not standing up for oneself’).

307
Ann-Carita Evaldsson

Instigating reports about fights to negatively assess a target of offence

The unfolding gossip-telling follows the general definition of gossip as a form of storytell-
ing whereby pejorative information is passed about a (non-present) third party (Goodwin
1990). The analysis of the first excerpt focuses attention on how the girls who constitute the
core group – Yaasmiin, Emine, Tara, and Azra (Figure 20.1) – organise their participation
through instigating stories (Goodwin 1990). The instigated stories project a chain of actions
that eventually culminate in a confrontation of the target (Evaldsson and Svahn 2017). The
instigating story, as Shuman (1986: 51) notes, is aimed not so much at describing the offen-
sive actions performed by the target (Rana) but at morally evaluating the behaviour and
character of the person. In this process the co-participants take up oppositional stances and
uses ‘members’ categories’ (with negative valence), indexing gender inappropriate behav-
iours to discriminate the reported target from the members of the peer group.

Tara Yaasmiin Emine Azra

Figure 20.1 The four girls constituting the core group are standing in the hallway.

Excerpt (1)
1. BOY 1: ↑Hallå:+ Beyan bråka me Rana innan-
↑Hey: Beyan had a fight with Rana before-
+--Comes down the stairs
2. EMINE: [↑Ahh:men hon ehh. ↑C-Phh
[↑YEAhh: but she’shh. ↑REtar-ded
3. YAASMIIN: [↑JAAhh. ↑ASS↑Å::hh i klassrummet
[↑YEAhh ↑LIK↑E::hh in the classroom
4. [((Steps forward))
5. (.hh.) hon sa så här till mej (.) hon bahh
(.hh.) she said like this to me (.) she likehh
6. >”asså va vill du jävla ↑FULING”<

308
 Examining girls’ peer culture-in-action

>”like what do you want you fucking ↑UGLY person”<

7. >jahh. ba (.) >”↑Fuck you”<
>I’m like (.) “↑Fuck you”<
9. ALL: ((Laughter and mumbling comments))

The transcription demonstrates how one of the girls, Rana, through the instigated story
becomes subject to the other girls’ negative evaluation in ways that discredit both her actions
and her membership in the girls’ peer group. To start with, one of the girls, Emine, who usu-
ally takes the lead, aligns with an instigated story about a fight between one marginal group
member (Beyan) and the target (Rana), ‘↑Hey: Beyan had a fight with Rana’ (Excerpt 1, line
1). Emine’s use of the feminine pronoun ‘she’ together with an extreme case formulation
(Pomerantz 1986) in the form of a third-person insult ‘Yea:h but she’s retarded’(Excerpt 1, line
2) positions Rana as expected to act in abnormal ways (Sacks 1992; Stokoe 2006). The use of
the pejorative category ‘retarded’ (Sw. CP, i.e. Cerebral palsy) is shaped with a high-pitched
voice to intensify the deviant gendered character of the target. Already in the following turn
one girl, Yaasmiin, who stands close to Emine (Figure 20.1), affiliates with the negative evalu-
ations. Her response cry ‘↑YEAhh ↑LIK↑E::hh’ (line 3) that is also shaped in a high-pitched
voice (Goodwin and Alim 2010: 184) serves to announce a spectacular story.
By animating the actions performed by target, Rana, ‘hh.she said like this to me (.)
she likehh’, the next teller (Yaasmiin) provides further evidence of the target’s offences.
Goffman’s (1981) notion of footing (animator, principal, figure) is relevant here for show-
ing how the teller locates the negative categorisation of the target in the ‘imaginary’ of the
telling through an animated third-party report (lines 6–8). By re-enacting the reported insult
>‘like what do you want you fucking ↑ugly person<‘ (line 6) the teller, Yaasmiin, casts the
target as a principle figure (Rana) responsible for speaking badly and therefore not to be
trusted (Evaldsson 2002; Goodwin 1990). In that way Yaasmiin manages to distance herself
from the reported insult, indexing a contrastive category membership for herself as loyal to
her gender.
By inserting herself as a figure in the imaginary of the telling, the teller (Yaasmiin) pub-
licly justifies her acts of retaliation ‘(.) >I’m like “↑Fuck you”<‘. The first-person pronoun
‘I’ in the described comeback indexes a subjective stance associated with a feminine gender
identity of ‘toughness’ and ‘assertiveness’ (Jackson 2011). Yaasmiin’s comeback is publicly
staged to strengthen her own gendered identity of ‘standing up for oneself’. At the same
time the telling offers affective cues to the other girls in the group to affiliate with her social
position (Goodwin and Alim 2010). We can see here how the oppositional stance taken
towards the targeted girl cast her as acting in abnormal ways while it indexes ‘toughness’
and ‘fighting back’ as gender-appropriate peer group conduct for the rest of the girls.

Upgrading reported insults in parallel third-party reports

As an interactional event, the girls gossip disputes are organised within shifting forms of
‘participation frameworks’ (Goodwin and. Goodwin 2004) of multiple actors who contrib-
ute in various ways. In what follows, a parallel third-party report in the form of a second
story is produced by two other members (Samina and Beyan) of the peer group. The two
girls, who at the time are in the hallway, produce a second story that topically ties to the first
(Sacks 1992). The analysis will show how the gossip-telling creates a participation frame-
work of two-against-one that allows the offended party to gain in gender reputation while
the principal figure (Rana) loses hers (Evaldsson 2002; Goodwin 1990).

309
Ann-Carita Evaldsson

Excerpt (2a)
10. SAMINA: ((Comes down the stairs with Beyan))
11. ↑Ja bahh >”titta på dej [själv du”< =
↑I’m likehh (.) >”you look at [yourself”< =
12. BEYAN: [↑MEhhJ?
[↑MEhh?
13. SAMINA: = hon bahh (.) [“↑Fuling”(.)
= she likhhe (.) [“Ugly person’”(.)
14. BEYAN: [↑MEhhJ?
[↑MEhh?
15. SAMINA: = ja bahh (.) “titta på dig själv du”-
= I’m likehh (.) “you look at yourself”’
16. BEYAN: ↑MEhhJ? (.) kalla hon ↑ME:J FU;:L↑ING
↑MEhh? (.) did she call ↑ME:: U:G↑LY

As can be seen in the transcription, the recipient of the instigated story (Beyan) inserts herself
as a figure in Samina’s telling and cooperates in the negative characterisation of the third party
(i.e. the target Rana) (Goodwin 2011: 268). The affective response cry, ‘↑MEhh’, from the
recipient (Beyan) is marked by prosody, expressive amplitude, and high pitch. The self-refer-
ence ‘↑ME::’ is then recycled three times (lines 12, 14, 16). The instigated story is designed
by the teller (Samina) to place the recipient (Beyan) in the position of the one who has been
offended: ‘↑MEhh (.) did she call ↑ME:: (.) UG↑LY’, (line 16). The reported offence motivates
a shift in footing or change of social positions for Beyan from ‘victim of offence’ to a new
public identity – that of an ‘accuser’ that commits herself to confronting the target (Evaldsson
and Svahn 2012, 2017) directly to her face (Excerpt 2b lines 18–20).

Beyan

Samina

Figure 20.2 Beyan runs up the stairs, followed by Samina.

310
 Examining girls’ peer culture-in-action

Excerpt (2b)
18. BEYAN: ((Walks toward Rana followed by Samina))
19. VARFÖR KALLA DU ME (.) ↑FULING (.)
WHY DID YOU CALL ME (.) ↑UGLY PERSON (.)
20. för ja e ↑BÄTT– ja ser ↑BÄTTre ut än du gör
cause I’m ↑BETT– I’m ↑BETTer looking than you
21. RANA: ((Mumbles something, turns away))
22. BEYAN: ASS↑Åhh >↑VA inte så himla kon.hh↑sti<
Lik↑ehh >↑DOn’t be so darn wei.hh↑rd<
23. ((Turns around, starts walking down))

The person referenced in Beyan’s confrontational act WHY DID YOU CALL ME (.)
“↑UGLY PERSON” (.) (line 19) builds further animosity towards the target for the
reported insult of ‘talking badly behind the back’. The derogatory category used by the
target is contrasted with a description of herself as ‘good looking’: ‘cause I’m ↑BETT–
I’m ↑BETTer looking than you’. The contrastive person description shifts the meaning of
the reported insult, ‘ugly person’, from ‘being disloyal’ to one’s gender to indexing ‘good
looking’ as valued gender attributes. However, instead of responding to the upgraded
accusation Rana now turns her body away and leaves the scene (line 21). The withdrawal
is commented on in a somewhat disappointed manner by Beyan, the accuser: ‘Like↑hh
↑DOn’t be so darn wei.hh↑rd’ (line 22), transforming the acts of withdrawal into a sign of
weakness or ‘cowardice’.
As Sacks (1972) notes, category-bound actions point out the moral features of cat-
egory concepts and thus the very accountability of certain absences or failures to per-
form the expected category-bound actions. Thus, failing to perform the normative peer
group conduct of ‘fighting back’ warrants a degradation of the targeted girl’s member-
ship status.

Invoking gender categorisations in playful and stylised performances

The following analysis of the unfolding gossip dispute will further show how categorical
attributes such as ‘fighting back’, ‘standing up for oneself’, and ‘having a laugh’, which are
common masculine attributes for working-class boys (see for example Evaldsson 2002),
are contingently re-enacted as a gender appropriate conduct when the girls organise their
peer culture-in-action. The following analysis will focus attention on how shifts in footings
(Goffman, 1981) from serious to playful modes permit the girls to manipulate and play with
gender representations of themselves as ‘victims of offences’ in a deliberate stylised play-
acting (Goodwin 2011: 268; Goodwin and Alim 2010).
The use of the collective pro-term ‘we’ (Sacks 1992: 144–149) in Emine’s open-
ing statement ‘we can fi:ght with Rana’ (line 1), projects a collective membership of
a particular category, that of ‘fighting girls’. The following imperative ‘[come ↑o::n’,
provides further resources for taking a collective stance and to stage an imaginary social
space where the girls commit themselves to confront the target and to ‘fight back’ (see
figures below and Excerpt 3).

311
Ann-Carita Evaldsson

Figure 20.3a The girls commit themselves to confront Rana.

Figure 20.3b Azra and Yaasmiin perform a catfight.

Excerpt (3)
1. EMINE: *VI KAN BRÅ:KA MED Rana, [KO↑::M
WE CAN FI:GHT WITH RANA, [COME ↑O::N
[((Smiles))
2. (2.0)
3. AZRA: >ja hatar henne< jag kommer sl↑åss
>I hate her< I’ll fi↑ght

312
 Examining girls’ peer culture-in-action

4. YAASMIIN: *Ja har lust att bar↑a [så hä:ra om du var hon=
I feel like j↑ust [like thi:s if you were her=
5. [((Grabs Azra))
6. [=da:::ng da:::ng da:::ng
7. [((Pretends to knee Azra three times))
8. AZRA: om du va ho:n -om du var ho:n jag [skulle så hä↑r
if you were he:r –if you were he:r I [would like thi↑s
9. [((Pulls Yaasmiins ponytail))
10. YAASMIIN: ↑WHOA::::H↑
11. ((Starts laughing))
12. TARA: ca:↑t fight hon ska göra ca↑t fight
ca↑t fight she’ll do a ca↑t fight
13. (1.8)
14. YAASMIIN: *hon tar ca:t fight jag gör så
*she takes a ca:t fight I do like
15. här da:::ng da:::ng*
this da:::ng da:::ng*
16. BEYAN: ((Looks into the camera))
17. (hh.) ja:: (.) det h↑är e:: (.) galna ungdomar
(hh.) yea::h (.) th↑is i::s (.) crazy kids

Already in the following turn one of the girls, Azra, affiliates with the collective member-
ship. In verbalising a highly negative emotional stance, ‘I ha:te her’ (line 3), she upgrades
her negative attitude towards the target for acting in gender-inappropriate ways while she
commits herself to participate in a fight: ‘I’ll fight’ (line 3) (Evaldsson and Svahn 2017). The
animated use of the personal pronoun ‘I’ becomes gendered through the embodied conduct,
exaggerated affect displays, and talk-in-interaction (Jackson 2011: 32). The affective self-
reference ‘I’ in ‘I hate her’ and ‘I’ll fight’ index shared gender norms of conduct by virtue of
the strong subjective stance claims made for ‘standing up for oneself’ and ‘taking a fight’;
although the gendered character of the ‘I’ as a ‘fighting girl’ in contrast to the claimed iden-
tity for the target as a ‘coward’ is not explicitly used in the girls’ interaction.
The two girls’ highly emotional attitude towards ‘taking a fight’ and ‘fighting back’ sets
the stage for the girls in the peer group to co-participate in a hypothetical fight play. Yaasmiin
immediately aligns with the imaginary play framework staged by Azra, by embracing for
herself the normative gender conduct of ‘taking a fight’, in terms of ‘I feel like just’ (line
4). She then directs Azra to play the role of an imaginary target, ‘if you were her’ (line 4).
A combination of physical re-enactments, including pretend knee-kicks and a loud voice
animating the sound effects of actual hits ‘↑DA:NG ↑DA:NG ↑DA:NG’ (line 6), is used
here to produce a stylised fight performance in which certain gender characteristics are
re-enacted. In an embodied format tying, reiterating Yaasmiin’s earlier performance, Azra
stages a stylised performance in which she directs Yaasmiin to act as ‘if you were her,’ (line
8). Azra now treats Yaasmiin ‘as if’ she were the reported target, Rana, through pulling
on her ponytail (Figure 20.3b). The stylised performance, keyed through the use of high-
pitched prosody, excessive screams and laughter tokens (lines 10–11), demonstrates how
the excitement lies not only in ‘fighting back’ but also in ‘having fun’ and ‘putting down’
other girls who engage in gender inappropriate conduct.
What is noteworthy is how the girls’ stylised embodied performances of ‘taking a fight’
becomes recognisable as a culturally valued gendered activity for members of the girl group.

313
Ann-Carita Evaldsson

The fact that Tara playfully categorises Azra’s stylised hairpulling as a ‘cat fight’ (line 12)
warrants a contrastive category for the reported target as a ‘coward’. In direct connection,
Yaasmiin makes a stylised fight performance: ‘I do like this DA:NG’ (lines 14–15), which is
staged to her own advantage. The stylised embodied re-enactments intensify the derogatory
characterisations of the targeted girl as acting in ways that are, as Speer (2005: 119) notes,
‘transgressive’ for one’s gender. Through the girls’ collaboratively performed actions the tar-
geted girl, Rana, is cast as a flawed character and as someone who does not even know how
to fight. Although it is obvious to the present girls that the stylised fight performance does
not produce authentic gender identities (lines 16–17), the imaginary framework has a serious
side as it commits the co-participating girls to physically confront the target in the near future.

The multiplied use and consequentiality of a sexual abusive category ‘slut’

In this micro-drama, which culminates in directed telling, chasing, and open fights, the gender-
inappropriate behaviours and derogatory categories attributed to the offending party for the
reported insults are gradually transformed and upgraded (Evaldsson and Svahn 2017). The last
example focuses attention on the multifunctional valence of the sexual abusive category ‘slut’
in the reported insult ‘WHY ARE YOU CALLING ME A SL↑U::T’ (Excerpt 4, line 1). It will
be shown how reports about offences have a strong performative character and power in that
they position the target as accountable for the offences while they commit the person being
attacked behind her back to confront the offender. The girls now end up chasing the targeted
girl, Rana, across the schoolyard, with Samina taking the lead (see the figure below).

Rana
Samina
Esma

Figure 20.4 The girls end up chasing Rana.

Excerpt (4)
1. SAMINA: VARFÖR KALLAR DU MEJ H↑O::RA
(→Rana) WHY ARE YOU CALLING ME A S↑U::T
2. EMINE: ä::rligt du har gjort de’ förut också::
(→Rana) ho::nestly you’ve done it before too::
3. ESMA: Sarah håller käften
Sarah shuts up

314
 Examining girls’ peer culture-in-action

4. SAMINA: ingen annan går på:: henne::

no one else atta::cks he::r
5. RANA: SLUTA FÖLJA EFTER MEJ ↑HOROR (.) VAD S↑ÄGER NI?
STOP FOLLOWING ME ↑SLUTS (.) WHAT DO YOU S↑AY?
6. SAMINA: ((Starts walking faster towards Rana))
7. ESMA: [((Tries to hold Samina back))
8. (→Samina) [ski:t i henne
[never mi:nd her
9. SAMINA: KALLA ’RU OSS ↑HO::ROR?
(→RANA) AR’ YOU CALLIN’ US ↑SLU::TS?
10. ((Shakes off Esma’s hand and starts running towards Rana))

Through the girls’ chasing activity, the targeted girl, Rana, is held directly responsible for the
morally transgressive actions of sexually insulting another girl. The use of the sexual category
‘slut’ in the reported sexual insult portrays the target in a sexually derogatory ways (Brown
and Chesney-Lind 2005). Calling someone a ‘slut’ is a common form of sexual regulation that
circulates between girls in schools when they attack other girls for coming across as ‘bad girls’
(Brown and Chesney-Lind 2005). However, while the term ‘slut’ is both gendered and sexual-
ised on a symbolic level, the girls’ use of the pejorative category ‘slut’ indexes norm breakers
of all types, not just those seeking sexual attention
By accusing the target for the reported sexual insult ‘WHY ARE YOU CALLING ME
A SL↑U::T’ (Excerpt 4, line 1) Samina positions herself as a ‘victim of a sexual offence’ to
justify her rights to ‘fight back’ (cf. Shuman 1986: 172). At the same time the public event of
a reported sexual insult creates a micro-political environment of multiple parties who sup-
port the current accuser to oppose the offender. One of the girls, Emine, immediately aligns
with Samina to build further animosity towards the reported offender. Through her use of an
extreme case formulation (Pomerantz 1986), the targeted girl’s behaviour is cast as extreme
and longstanding: ‘you’ve done it before too::’ (line 2). The upgrading of the offence creates
a public event of all-against-one with all the girls’ now ending up chasing the target around
the schoolyard. At this moment the accused and chased party, Rana, turns the sexual abuse
towards her perpetrators to defend herself, ‘STOP FOLLOWING ME (.) SLU::TS’ (line
5). However Rana’s come-back is turned against her in Samina’s embodied response-work,
which makes not only herself but the collectivity of girls into ‘victims of a sexual offence’:
‘Ar’ you callin’ us slu::ts?’ (line 9).
Thus, rather than simply presenting oneself as a ‘successful aggressor’, the reports of a
sexual insult works to assert that the girls have been seriously offended (cf. Shuman 1986:
172). The accusation warrants ‘fighting back’ as justifiable feminine conduct for the girls
in the peer group. The reported offences position also specific participants such as Samina
with specific entitlement to take action and defend the moral order, ‘no one else atta::cks
he::r’ (lines 4, 7–10) (cf. Goodwin 1990). Indeed, what is evident is that the right to take
action against an offender by ‘fighting back’ is also charged with social consequences for the
accuser who may lose her social status if she fails to carry through the attack.

Concluding comments
In this chapter I have shown how an ethnomethodological conversation analytical approach
to membership categories in sequences (Stokoe 2006), combined with notions of stance and
footing (Goffman 1981; Goodwin and Goodwin 2004; Goodwin 1990, 2006) make observable

315
Ann-Carita Evaldsson

the emergent character of how young girls are doing gender through repeated stance perfor-
mances in courses of unfolding gossip dispute activities (see also Evaldsson 2007; Evaldsson
and Svahn 2012, 2017). It is in recognition of the multi-layered cultural meanings of catego-
ries, stances, and actions within specific contexts of use that an ethnomethodological approach
to gender and language provides an empirically grounded analysis of situated actions that
‘remains true to the participants’ orientations’ (Speer and Stokoe 2011: 25). What is note-
worthy in the sequential analysis is how the participants’ orientations to gender and norms of
conduct become recognisable through negative person descriptions, assessments, and cate-
gory-bound activities directed towards a target for failing to perform gender-appropriate con-
duct. The gender reference is also picked up in subsequent turns at talk and embodied conduct
by the co-participating girls, to whom the gender inappropriate behaviours do not apply, who
upgrade their group membership as they commit themselves to confront the target.
The ethnomethodological conversation analytical approach to membership categories
demonstrates the importance of analysing the emergent propriety of social interaction and
‘in-course’ accomplishment of gender (Goodwin, 2011; Speer and Stokoe 2011; Stokoe
2006); and how an analysis of ‘doing gender’ incorporates a sensitivity to the contingencies
that the participants themselves orient to in the temporal unfolding of actual events. Thus,
from members’ perspective what is a gender appropriate conduct become important to be
attentive to in-courses-of-actions. As shown, if a girl does not manage to master proper
interactional procedures her gendered identities and membership status in the group may
easily become subject to questioning and contestation.
As has been found in earlier research of girls growing up in working-class communities,
group members continuously strengthen alignments and commitments to shared norms of
femininity through linguistic practices such as gossip disputes (Eder 1995; Evaldsson and
Svahn 2012, 2017; Goodwin 1990, 2006; Shuman 1986). The sociolinguistic notion of stance
and footing as developed by Goodwin (2006, 2007, 2011; Goodwin and Alim 2010) demon-
strates the dynamic and performative nature of gendered identities and how the participating
girls not only manipulate their actual selves but create multiple representations of self and
others. For example, the girls’ stylised play actions of a ‘cat fight’ signal the multiple and
multi-layered character of actions, stances, and identities (Jaffe 2009); and how the form of a
given act can be transformed through the use of multiple modalities (talk along with non-vocal
performances and embodied conduct) (see Goodwin and Alim 2010) into something else.
The study of stance also demonstrates the importance of considering the more enduring
interactional and social purposes, and the subject positions taken up more broadly (Bucholtz
2009). At a symbolic cultural level the repeated stance performances of degradation, either
playful or serious, have enduring consequences for the positioning of the target. In the social
context of gossip disputes in-group members draw upon gendered norms of conduct in their
communities to cast particular girls as ‘cowards’ and as ‘disloyal’ to their gender, which in
turn may have severe social consequences for specific girls (cf. Goodwin and Alim 2010).

Future directions
In conclusion, worth drawing on from the findings are some theoretical and methodological
issues that are of relevance for future ethnographic research on gender, sexuality, and lan-
guage in interaction; and that both girls and boys social worlds and peer language practices
should be included in this endeavour.
Most important is the creative and agentive role children’s peer interactions play in per-
forming and simultaneously enacting a feminine or a masculine gender identity and group

316
 Examining girls’ peer culture-in-action

membership, and how in this process talk, voice, and the body, along with the material
environment, are used to perform multifaceted membership categories, identities, and social
relations (Goodwin and Kyratzis 2014 for an overview). Ethnographically based research
and close analysis of talk- and bodies-in-interaction are both essential for capturing the
routine local accomplishment of gender along with other social categories in children’s
everyday peer language practices; the more enduring social organisation and networks of
its production; and the various situated and cultural practices in which it occurs. As noted
by Goodwin and Kyratzis (2011) in their review of peer language research, it is ‘through
examining stance taking we can come to grips with the concerns that deeply animate par-
ticipants’. The ways in which girls across groupings manipulate and play with gendered,
ethnical and racial representations in peer interaction, point to the importance of exploring
how local identifiers and broader social categories are co-constructed through playful stance
taking and deliberate stylizations.
It is in this dialogical context that long-term ethnographic research of children’s peer
language practices becomes critical for making observable the situated and dynamic ways
in which girls and boys, in transnational, culturally hybrid and multilingual settings perform
social identities that are important to them in their social life with peers.

Transcription conventions
Transcripts are based on Jefferson notation (Jefferson 2004).

bu-u- a cut-off of the preceding sound

[] an overlap of speech
= no break or gap between turns
(.) brief interval (less than 0.2) within or between utterances
so:::rry a sound stretch,the number of colons indicates a prolongation
you underline indicates emphasis
↑ shifts into high pitch
↓ shifts into low pitch
DOG loud talk is indicated by upper case
hey? indicates a rising intonation
dog¿ a Spanish question mark indicates a substantial rise
here, a comma indicates a continuing intonation
did. a full stop indicates falling, final intonation
soft˚ softer, quieter sounds
.>quick< talk is speeded up
<slow> talk is slowed down
() the talk is not audible
(house) transcriber’s best guess for the talk
together! indicates an animated tone
((walking)) annotation of nonverbal activity

Note
1 The data in this chapter has been used in a coedited article with a different analytical focus (Evaldsson
and Svahn 2017).

317
Ann-Carita Evaldsson

Further reading
Goodwin, M. H. and Kyratzis, A. (2011) ‘Peer socialization’, in Duranti, A., Ochs, E., and Schieffelin,
B. (eds) The handbook of language socialization. Malden, MA: Wiley Blackwell, pp. 365–390.
This chapter provides an overview of central research on children’s peer language socialisation
across diverse cultures and settings.
Goodwin, M. H. and Kyratzis, A. (2014) ‘Language and gender in peer interactions among children
and youth’, in Erlich, S., Meyerhof, M., and Holmes, J. (eds) The handbook of language, gender,
and sexuality. New York: John Wiley and Sons, pp. 509–528.
This chapter presents an in-depth review of the different paradigms that have influenced and
continue to influence research on the language and gender of children and youths among peers.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context.
Cambridge: Cambridge University Press, pp. 335–358.
In this chapter, Ochs presents her pioneering work on indexing gender and arguments for an
indexical perspective on gender as mediated by stance.
Speer, S. and Stokoe, L. (2011) ‘An introduction to conversation and gender’, in Speer, S. and Stokoe,
L. (eds) Conversation and gender. Cambridge: Cambridge University Press, pp. 1–28.
This edited volume brings together a range of studies on gender using a conversation analytic
approach, examining children’s play, helpline calls, family interaction, social work, and mediation.
West, C. and Zimmerman, D. (2009) ‘Accounting for doing gender’, Gender and Society, 23(1), pp.
112–122.
In this article, West and Zimmerman reflect on their classical ethnomethodological work (West and
Zimmerman 1987) on ‘doing gender’ and how the concept has been used among other researchers
with different theoretical and methodological approaches.

Related topics
The accomplishment of gender in interaction: ethnomethodological and conversation analytic approaches
to gender; feminist conversation analysis: examining violence against women; the accomplishment of
gender in interaction: ethnomethodological and conversation analytic approaches to gender; gender and
sexuality normativities; applying queer theory to language, gender, and sexuality research in schools.

References
Benwell, B. and Stokoe, E. (2016) ‘Ethnomethodological and conversational analytical approaches to
identity’, in Preece, S. (ed) The Routledge handbook of language and identity. London: Routledge,
pp. 66–82.
Brown, L. M. and Chesney-Lind, M. (2005) ‘Growing up mean: covert aggression and the policing
of girlhood’, in Lloyd, G. (ed) Problem girls: understanding and working with the troubled and
troublesome. Abingdon: Routledge, pp. 74–86.
Bucholtz, M. (2009) ‘From stance to style. Gender, interaction and indexicality in Mexican immigrant
youth slang’, in Jaffe, A. (ed) Stance: sociolinguistic perspectives. Oxford: Oxford University
Press, pp. 146–170.
Eder, D. (1995) School talk: gender and adolescent culture. New Brunswick, NJ: Rutgers University
Press.
Evaldsson, A.-C. (2002) ‘Boys’ gossip telling: staging identities and indexing (non− acceptable)
masculine behavior,’, Textile, 22, pp. 1–27.
Evaldsson, A.-C. (2007) ‘Accounting for friendship: moral ordering and category membership in
preadolescent girls’ relational talk’, Research on Language and Social Interaction, 40(4), pp. 377–404.
Evaldsson, A.-C. and Svahn, J. (2012) ‘School bullying and the micro-politics of girls gossip telling’,
in Danby, S. and Theobold, M. (eds) Disputes in everyday life: social and moral orders of children
and young people. Bingley: Emerald, pp. 297–322.

318
 Examining girls’ peer culture-in-action

Evaldsson, A.-C. and Svahn, J. (2017) ‘Staging social aggression. Affective stances and moral
character work in girls’ gossip telling,’, Research on Children and Social Interaction, 1, pp. 1–27.
Garfinkel, H. (1967) Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.
Goffman, E. (1981) Forms of talk. Philadelphia, PA: University of Pennsylvania Press.
Goodwin, C. (2000) ‘Action and embodiment within situated human interaction’, Journal of
Pragmatics, 32(10), pp. 1489–1522.
Goodwin, C. and Goodwin, M. H. (2004) ‘Participation’, in Duranti, A. (ed) A companion to linguistic
anthropology. Malden, MA: Blackwell Publishing Ltd, pp. 222–244.
Goodwin, M. H. (1990) He-said-she-said: talk as social organization among black children.
Bloomington, IN: Indiana University Press.
Goodwin, M. H. (2006) The hidden life of girls: games of stance, status, and exclusion. Oxford:
Blackwell.
Goodwin, M. H. (2011) ‘Engendering children’s play: Person reference in children’s conflictual
interaction’, in Speer S. A. and Stokoe E. (eds) Conversation and gender. Cambridge: Cambridge
University Press, pp. 250–271.
Goodwin, M. H. (2007) ‘Participation and embodied action in preadolescent girls’ assessment
activity’, Research on Language and Social Interaction, 40(4), pp. 353–375.
Goodwin, M. H. and Alim, S. (2010) ‘Whatever (neck roll, eye roll, teeth suck) Therapie situated
coproduction of social categories and identities through stancetaking and transmodal stylization’,
Linguistic Anthropology, 20(1), pp. 179–194.
Goodwin, M. H. and Kyratzis, A. (2011) ‘Peer socialization’, in Duranti, A., Ochs, E., and Schieffelin,
B. (eds) The handbook of language socialization. Malden, MA: Wiley Blackwell, pp. 365–390.
Goodwin, M. H. and Kyratzis, A. (2014) ‘Language and gender in peer interactions among children
and youth’, in Erlich, S., Meyerhof, M., and Holmes, J. (eds) The handbook of language, gender,
and sexuality. London: Wiley-Blackwell, pp. 509– 528.
Jackson, C. (2011) ‘The gendered “I”’, in Speer, S. and Stokoe, L. (eds) Conversation and gender.
Cambridge: Cambridge University Press, pp. 31–47.
Jaffe, A. M. (2009) ‘Introduction: the sociolinguistics of stance’, in Jaffe, A. (ed), Stance sociolinguistic
perspectives. Oxford: Oxford University Press, pp. 3–28.
Jefferson, G. (2004) ‘Glossary of transcript symbols with an introduction’, Pragmatics and Beyond
New Series, 125, pp. 13–34.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking context.
Cambridge: Cambridge University Press, pp. 335–358.
Pomerantz, A. (1986) ‘Extreme case formulations: a way of legitimizing claims’, Human Studies
Design Studies, 9(4), pp. 219–229.
Sacks, H. (1972) ‘On the analyzability of stories by children’, in Gumperz, J. and Hymes, D. (eds)
Directions in sociolinguistics: the ethnography of communication. New York: Rinehart and
Winston, pp. 325–345.
Sacks, H. (1992) Lectures on conversation, vol I and II. Oxford: Blackwell.
Shuman, A. (1986) Storytelling rights: the uses of oral and written texts by urban adolescents.
Cambridge: Cambridge University Press.
Speer, S. (2005) Gender talk. Feminism discourse and conversation analysis. London: Routledge.
Speer, S. and Stokoe, L. (2011) ‘An introduction to conversation and gender’, in Speer, S. and Stokoe,
L. (eds) Conversation and gender. Cambridge: Cambridge University Press, pp. 1–28.
Stokoe, E. (2006) ‘On ethnomethodology, feminism and the analysis of categorical reference to gender
in talk-in-interaction’, The Sociological Review, 54(3), pp. 467– 494.
West, C. and Zimmerman, D. (1987) ‘Doing gender’, Gender and Society, 1(2), pp. 125–151.
West, C. and Zimmerman, D. (2009) ‘Accounting for doing gender’, Gender and Society, 23(1), pp.
112–122.

319
 Part V
Sociocultural and critical
approaches
 21
Language, gender, and sexuality
Reflections on the field’s ongoing critical
engagement with the sociopolitical landscape

Lia Litosseliti (Part V lead)

Introduction
I started writing this chapter at a particularly socially and politically turbulent time in 2020.
The world was in the grip of the COVID-19 global pandemic, which put many of us in
‘lockdown’ within our homes and cities, and soon started to expose our societies’ various
pressure points and inequalities. At the same time, the death of George Floyd at the hands
of the police in the USA sparked the beginning of a wave of anti-racism solidarity protests
around the world, which continue globally and locally as I write. During that time, I was
reminded by Michel deGraff, in his talk (aovivo.abralin.org – June 2020) on Black lives
and the politics of linguistics and education in post-colonies, of the Linguistic Society of
America’s statement that ‘given our distinct insights into communication and culture, lin-
guists are well positioned to contribute to social justice and equality’. It could not have been
a more suitable time, then, to reflect on equality and social justice, and specifically on the
focus of this section of this volume, on sociocultural and sociopolitical approaches within
the language, gender, and sexuality (LGS) field.
The field has been, since its inception, inherently concerned with the examination of lan-
guage and gender – and later, language, gender, and sexuality – as sociocultural phenomena,
explored within social settings, domains, contexts, and cultures. Additionally, LGS scholars
have developed extensive and rich analyses of the multi-faceted relationships between lan-
guage, gender, sexuality, and society, and of the effects of language use within societies and
vice versa. Excluding work in formal linguistics focusing exclusively on parts and features
of language such as grammatical gender or pronouns (and not on the use of grammatical
gender and pronouns), the vast majority of LGS work could be broadly described under
the umbrella term of sociocultural or socially oriented linguistic scholarship, in being con-
cerned with the above-mentioned social contexts, relationships, and effects. The work cited
in this chapter on sociocultural and sociopolitical approaches, accordingly, comes from all
the traditions included in the eight parts of this volume; and scholars in all of these traditions
have helped advance the field as a whole, including by engaging with many perspectives in
social and feminist theory and other disciplines.

323
Lia Litosseliti

Sociolinguistics is an apt example of a substantial body of work within sociocultural

linguistics that reflects the influence of other disciplines (linguistic anthropology, sociology,
psychology, communications, literary theory) and shows how linguistic, cognitive, ideo-
logical, stylistic, cultural, and social dimensions are inextricably intertwined – for example
in studies of the prestige of languages and language varieties. However, the increasingly
narrow association of the term sociolinguistics with quantitative studies of linguistic fea-
tures and their correlation to sociological variables, has led to a distinction between the body
of work in ‘sociolinguistics’ and the umbrella term of ‘sociocultural linguistics’ – the latter
denoting ‘the broad interdisciplinary field concerned with the intersection of language, cul-
ture, and society’ (Bucholtz and Hall 2005: 5). And further, within the diverse sociocultural
linguistics scholarship, some theoretical and methodological approaches will be distinct
from others in their particular emphasis on political dimensions (for example, these are at
the centre of critical discourse analysis, see below) and on cultural dimensions (which are
often key, for example, in anthropology and ethnography).
In this chapter, I map the terrain of the language, gender, and sexuality field’s ongo-
ing critical engagement with the sociopolitical landscape, as it expanded and became more
diverse over the decades. I have chosen to refer to the ‘sociopolitical’ landscape as a more
encompassing term than ‘social’ or ‘sociocultural’ and to highlight the field’s development
as inextricably linked with sociopolitical struggles – notably struggles for women’s, civil,
and LGBT+ rights. I start with the field’s long-standing close links with feminism and the
practice of feminist politics (first section). I then consider three broad areas where the field
has expanded in recent years – and which have involved/ necessitated and further led to
the field explicitly engaging with sociopolitical developments: attention to intersectionality
(second section); the inclusion of sexuality scholarship (third section); and the development
of critical methodologies, including through dialogue with other disciplines (fourth section).
In the fourth section I also draw from my own work in order to illustrate the importance of
a continuing and renewed engagement with feminist theory/methodology and other disci-
plines for our field. Finally, I introduce the four contributions which follow in this part of
the volume.
The chapter is not intended as a comprehensive review; rather I am being selective in
my references to existing studies, in order to illustrate key strands of LGS work and per-
spectives. In addition, I highlight texts on key developments and debates for those new to
the field (in this respect, excellent starting points are Bucholtz 2014, McElhinny 2014, and
Zimman and Hall’s 2016 bibliography on language, gender, and sexuality).

The feld’s feminist foundations

The feminist movement – a movement of political campaigning for reforms on different
sets of ideas such as women’s suffrage, access to education and employment, sexual vio-
lence, equal pay, and reproductive rights – is often described through a wave metaphor.
First-wave feminism is associated with the suffragette women’s movement and the strug-
gles for political equality in the late nineteenth and early twentieth centuries. Second-wave
feminism, from the early 1960s to the 1980s, promoted social equality for women, repro-
ductive freedom, and political resistance against women’s casual and systemic oppression
inside and outside the home. Third-wave feminism, from the early 1990s (though there is
disagreement about its chronology), can be described as a diffuse movement, embracing
intersectional feminism and marginalised groups and advocating for the multiplicity and
performativity of gender. Finally, there is considerable debate about whether the digitally

324
 Engaging with the sociopolitical landscape

driven, sex-positive, queer, #MeToo, inter-generational feminism of the last decade or so

can be described as a fourth wave.
Since the second wave, feminism has been a fundamental influence on language, gen-
der, and sexuality work. The different feminisms developed over time may have differ-
ent perspectives and emphases, but they share a commitment to addressing social/gender
inequality. In the same vein, feminist linguists share an explicit or implicit commitment to
connecting gendered-related linguistic phenomena to gender inequality or discrimination,
even though they take varying theoretical and methodological perspectives in fulfilling this
commitment.
In her 2014 review of the feminist foundations of LGS research, Bucholtz traces the the-
oretical roots of feminist linguistics back to the difference-based theories that emerged dur-
ing feminism’s second wave. From the early days of Lakoff (1975) and the ‘sexist language’
debates, to notions of gender ‘dominance’ of men over women through small-scale social
interactions, to the ‘(cultural) difference’ approach and theorisations of female talk, these
theories start from a typically binary ‘difference’ position (women’s/girls’ language use
being different from men’s/boys’) and with women’s experience at the centre. Their merits
and omissions have been extensively discussed in both texts (e.g. Eckert and McConnell-
Ginet 2013; Ehrlich et al. 2014; Litosseliti 2006) and LGS curricula over time, typically
treated as overlapping theories of deficit (liberal feminism), difference (cultural feminism),
and dominance (radical feminism) (Cameron 1995).
It is not my intention to reproduce the difference paradigms’ contributions and gaps
here. Rather, the point in the context of my discussion of the field’s long-standing engage-
ment with feminist politics and social inequality, is that these theories were a product of
their political time: a time of second-wave feminist consciousness-raising around civil
and women’s rights, action to increase women’s visibility and participation in the public
sphere, and campaigns against sexism/sex discrimination in all domains of social life. Put
differently, these paradigms, too, took shape in a particularly fertile environment for both
explicit critiques and activism around social and gender inequality. And as the environment
kept changing over time, so did the LGS field (more on which later). The field’s long-term
intimacy with feminism notwithstanding, not all LGS work (back then and today) can be
described as feminist linguistics work, because its stake ‘in feminism as a political move-
ment or theory may not be evident or explicit’ (Litosseliti, 2006: 22); and because ‘not all of
it shares a political commitment to social justice. [Rather, some of it] seeks simply to corre-
late language patterns with categories of gender and / or sexuality’ (Bucholtz 2014: 23–24).
Nevertheless, feminist linguists have been instrumental to the field’s development from the
start and continue to take the field in new sociopolitically committed directions (see below).
Not surprisingly, these directions are not necessarily or straightforwardly unified. A case
in point is material feminist linguistic approaches (Shankar and Cavanaugh 2012), which
arose within linguistic anthropology. These approaches are concerned with the material and
physical conditions of language use and their role on the social production of gender/gender
hierarchies – for example, the conditions that enable or restrict access to education, higher
or lower economic status, and workplace and career opportunities. These approaches cannot
be categorised in terms of perspective as easily as the ‘difference’ approaches can. As seen
in reviews of examples of this work (Bucholtz 2014; Zimman and Hall 2016), they include
an array of studies, from gender, class, and variation, to gendered linguistic labour in the
home, to language and gendered embodied practices. The material effects of language/dis-
courses of gender and sexuality are also discussed in several workplace studies within this
tradition of research: Hall’s (1995) study on the selling of stereotyped women’s language in

325
Lia Litosseliti

phone sex; Toerien and Kitzinger (2007) on emotional labour in a UK beauty salon; Holmes
(2006, 2014) on the gendering of requests and refusals in New Zealand factories; Cameron
(2000) on feminine communicative styles in UK call centres; and Ostermann (2003) on
affiliative strategies at an all-female police station and feminist crisis-intervention centre in
Brazil. I return to material feminist linguistic approaches and more recent work on language
and political economy towards the end of this chapter.
Since the early days, LGS research has developed to become wider and more diverse.
In the mid-1990s, under the influence of the ‘discursive turn’ in social theory and post-
structuralism (see Part VI, this volume), the conceptualisation of gender as something we
do or accomplish in discourse and gender identities being constructed in a context-situated
and fluid process of negotiation, appropriation, and restatement (see e.g. Litosseliti 2006;
McEntee-Atalianis 2019) has added greater nuance, context-specificity, and analytical
complexity to language, gender, and sexuality work. The fragmentation of feminism into
feminisms (liberal, radical, multicultural, postcolonial, etc.), and particularly the third-wave
emphasis on the multiplicity and performativity of gender, have also led to a problematisa-
tion of gender categories as fixed or stable – especially so when categories like race, class,
ethnicity, religion, or sexual orientation are also taken into account.
Over time, the field grew and benefitted from the involvement of new language scholars
from a range of fields, such as anthropology, communication, education, linguistics, psy-
chology, sociology, and gender studies. It also benefitted from a wider array of global and
cross-cultural perspectives, illustrated in studies mentioned later in this chapter; in examples
of work from Poland, Brazil, Japan, the Middle East, and North Africa (Ehrlich et al. 2014),
Japan (e.g. Okamoto and Shibamoto Smith 2004), sub-Saharan Africa (Atanga et al. 2012),
Greece (Canakis et al. 2010), etc.; and in Global South geopolitical perspectives that ques-
tion the dominance of Northern scholarship in our field (see Milani and Lazar 2017).
Moreover, contrary to the studies of the 1970s, 1980s, and early 1990s, LGS investiga-
tions since the mid-1990s have extended their focus beyond white, straight, middle-class
women, and have included a focus on men/masculinity as well as on sexualities (see below).
In these ways, the field became more wide-ranging and diverse, as well as more critical
(although see the fourth section later): while a single defined set of feminist goals became
less of a focus, LGS scholars began to interact with critical feminist work on race, colonial-
ism/post-colonialism, multiculturalism, femininities/masculinities and sexuality.
In what follows, I consider three key areas where, in addition to the foundational and
ongoing engagement with feminism, the LGS field has expanded in recent decades – and
which have involved and further led to the field explicitly engaging with the broader soci-
opolitical landscape: attention to intersectionality; the inclusion of sexuality scholarship;
and the development of critical methodologies, including through dialogue with other
disciplines.

Intersectionality
As the field expanded, it became more engaged with the ways in which gender intersects
(particularly evident in close analyses of interaction) with other social dimensions or forms
of identity, such as race, social class, age, ethnicity, religion, and sexuality. Intersectionality
(see also Leppänen and Tapionkaski, this volume) is a theoretical response by black femi-
nist scholars who sought to understand black women’s particular experiences (e.g. Collins
2000; Crenshaw 1991), but more generally it is a theoretical response that recognises that

326
 Engaging with the sociopolitical landscape

the interplay of different dimensions of social life plays an important – and neglected – role
in maintaining systems of oppression.
Intersectionality theory can be usefully disruptive. Most evidently, it questions the sec-
ond-wave feminist assumption of similarity of experience among women and its primary
focus on the concerns of white middle-class heterosexual women (Hancock 2007; Smooth
2011; Weldon 2008). In this sense, it is a useful tool for holding a movement, or academic
field, or research paradigm to account, by questioning some of its own hegemonies and
exclusions. In addition, intersectional perspectives further disrupt the idea of a linear sin-
gular identity (that there is one way of being a woman, or being black, or being gay, and so
on), thereby shifting the attention to non-normative categories.
Intersectional perspectives within the LGS field are integral to many sociolinguistic stud-
ies (e.g. see Levon 2015; Levon and Mendes 2016) and much of the work on the role of
language in the doing of racial and ethnoracial identity work (e.g. Bucholtz 2011 on white-
ness; Cashman 2018 and Mendoza-Denton 2008 on Latinx communities; Lanehart 2009 on
African-American women). In addition, LGS research on masculinities – as part of the post-
1990s language and masculinity scholarship (Benwell 2014; Cameron 1997; Coates 2003;
Johnson and Meinhof 1997) which started to examine men’s previously neglected gendered
lives as well as problematise hegemonic masculinity – have used an intersectional lens to
highlight the performance of hegemonic and non-normative masculinities across different
social groups (e.g. Kiesling 2005; Milani 2015). In the past five years, there has also been a
proliferation of intersectional perspectives in the field – for example, Candelas de la Ossa’s
2019 study on guidance for domestic abuse survivors, Baker and Levon’s 2016 examination
of racialised and classed masculinities in print UK media, Trechter 2014 on language and
ethnicity intersections, and several studies in the Gender and Language special issue on
intersectionality, language, and queer lives (Gray and Cooke 2018).
This body of work moves the field forward by showing the inclusions and exclusions
made possible at the (previously neglected) intersections, but also by adding to calls for
more, and more developed, LGS intersectional approaches. Within language studies,
Romero (2017) for example has argued the need to develop variable-with-variable intersec-
tional approaches (rather than a variable-by-variable approach to gender or race identity), to
facilitate more complex, nuanced understandings of identities. Levon (2015) has also argued
for the need to integrate intersectionality theory and analysis more fully in language, gender,
and sexuality research, if we want to increase our understanding of how social inequalities
are produced (and constructed and resisted) by interacting axes of social differentiation. At
the same time, it is important to consider that intersectional perspectives require new and
nuanced methodological paradigms that can deal with complexity (Celis et al. 2013). And
as importantly, scholars also caution against the ubiquitous use of intersectionality to draw
attention to the lived experiences of those in the margins without actually engaging with
the intersectional systems that produce and maintain oppression (Nash 2008). This broader
point also applies to scholarship on sexuality, to which I now turn.

Sexuality
Another way in which the LGS field became wider, as it continued to interact with differ-
ent social theories and the changing sociopolitical environment, has been through encom-
passing, in the 1990s, critical work on the discursive and material character of sexuality.

327
Lia Litosseliti

(Sexuality and gender are closely intertwined (see e.g. McElhinny 2014). I discuss sexuality
separately here for the purposes of this chapter’s structure.)
Similarly to the recognition of the historic lack of intersectional perspectives, language
and gender scholars increasingly came to recognise the persistence of gender binaries and
heteronormativity that had implicitly shaped previous work in the field (see Zimman et al.
2014), as they began to engage with the emergent field of queer theory/queer feminism
(Butler 1990). Queer theory is an interdisciplinary theoretical approach that draws on a
range of methodological paradigms in order to critique and destabilise taken-for-granted
notions of heteronormativity and of gender as an a priori category (see Sauntson, this vol-
ume). The LGS field’s shift towards incorporating these perspectives is reflected in the
inclusion of sexuality in how the field has been referred to since the mid-1990s and in the
establishment in 2012 of the Journal of Language and Sexuality. This shift is discussed by,
among others, Cameron and Kulick (2003), and Motschenbacher and Stegu in their 2013
special issue on queer linguistics for Discourse & Society.
The overarching aim of queer linguistics scholarship (see Bucholtz and Hall 2004;
Cameron and Kulick 2003; Motschenbacher 2010) is to bring a linguistic lens to the critical
examination of sexuality and the relationship between gender and sexual identities and prac-
tices – including heterosexual identities and practices. Motschenbacher and Stegu (2013:
522) define queer linguistics as ‘critical heteronormativity research from a linguistic point
of view’, with a focus on the interrogation of privileged and normative sexualities, and a
concern for exposing the resulting forms of discrimination. As part of such interrogation,
queer linguistics scholarship also acknowledges that language and gender research does
not fully capture the linguistic dimensions of desire, affect, and emotion (for debates, see
Kulick 2000 alongside Bucholtz and Hall 2004; also Queen 2014). Further, incorporating
and building on intersectionality perspectives, this body of work also acknowledges that
sexuality needs to be viewed as embedded within broader formations of gender, race, eth-
nicity, ability, economic status, etc. (Leap and Motschenbacher 2012).
Linguistic approaches to sexuality have broadly focused on the ways in which people
enact sexuality through language, as well as the ways in which sexual identities are rep-
resented linguistically in a range of settings. More precisely, they can be summarised as
investigations of three types of questions (Bucholtz 2014: 36): ‘linguistic aspects of the
social and political struggle of LGBT groups and individuals’, ‘the linguistic practices of
particular LGBT-identified groups’, and ‘discursive representations of LGBT identities by
both ingroup and outgroup members’.
For example, several sociolinguistic variation studies have examined the role of different
variables of sociolinguistic practice – and their interplay – in the negotiation of sexuality (see
e.g. Levon 2010). Such work combines micro-level linguistic analysis with a queer linguis-
tics concern with broader social dimensions of power, normativity, inclusion, and exclusion.
Queer linguistic studies often draw on a range of different methodologies (Sauntson, this
volume), but are united in their consistent effort to question gender binaries and heteronor-
mative/hegemonic gender and sexual subjectivities (see, among others, Barrett 1999; Jones
2012; Zimman and Hall 2010). In many cases, such questioning goes hand in hand with the
questioning of Western binaries, as illustrated in, among others, Besnier’s 2003 ethnography-
based study of transgendered men in Tonga, Kulick’s (1998) and Borba and Ostermann’s
(2007) studies on travesti identities in Brazil, Gaudio’s 2009 study on the linguistic strategies
used by the Nigerian ‘yan daudu and Hall’s 2005 work with transgender kotis in India. In
addition, recent research explores language use and transgender, intersex, and genderqueer
identities in different sociocultural contexts around the world (Zimman et al. 2014).

328
 Engaging with the sociopolitical landscape

In the final section that follows, I turn to the role of critical and cross-disciplinary meth-
odologies in the LGS field and discuss some recent and new directions.

Critical methodological frameworks and cross-disciplinary bridges

In addition to critical theorisations, the language, gender, and sexuality field’s attention to
sociopolitical issues is also reflected in its development of critical/feminist methodological
approaches. Over the last 20 years, we have witnessed developments in feminist conversa-
tion analysis (Kitzinger 2000; Speer 2005; Weatherall and Tennent this volume), feminist
pragmatics (Christie 2000), and feminist post-structuralist discourse analysis (Baxter 2008,
see also Mackenzie this volume). In addition, probably the most deliberate feminist effort to
unite scholarship with political struggle against systemic oppression and discrimination can
be found in feminist critical discourse analysis (FCDA) (Lazar 2005, 2007). FCDA, at the
nexus of critical discourse analysis and feminist studies, examines language use as situated
in the sociopolitical discourses that constitute and maintain power hierarchies, with a par-
ticular interest in hierarchically gendered social orders. FCDA – like other emerging critical
approaches to the study of the environment, technology, and the global political economy (I
touch on the latter towards the end of this chapter) – bring together materialist and discur-
sive perspectives; this synthesis has the potential to ‘influence the field as a whole, as well
as feminism more generally’ (Bucholtz 2014: 32).
The development of critical methodological frameworks in the field has opened up new
questions about the nature of LGS scholarship. Notably, it has given impetus to ongoing
debates on the strengths and limitations of ‘bottom-up’ and ‘top-down’ approaches that
focus on micro- and macro-level analyses respectively. Discussions of these dimensions
can be found in, among others, Litosseliti (2018) and in Harrington et al. (2008) – the lat-
ter focusing on micro/macro LGS analyses in sociolinguistics, corpus linguistics, discur-
sive psychology, CA, CDA, FPDA, and queer linguistics. For the purposes of this chapter,
I will only highlight some notable examples of micro–macro considerations here.
Critical discourse analysts, for example, typically take a materialist approach by looking
at how discourses in key social domains (media, politics, workplaces, education) construct
power relations, hierarchies, and ideologies (Chouliaraki and Fairclough 1999; Weiss and
Wodak 2003) – including those related to gender, sexuality, and heteronormativity. They
look beyond the text and the data for interdiscursive patterns and macro-processes situated
in broader sociocultural and historical contexts. They are often criticised (e.g. by conversa-
tion analysts) for having an ideological and politically motivated agenda, for cherry-picking
texts to suit their agenda (see Murphy and Palma-Fahey this volume), and ‘seeing individuals
as an effect of discourse, thereby granting them limited agency’ (McEntee-Atalianis 2019:
23). In contrast, ethnomethodologists and conversation analysts take an emic approach,
starting with a micro-level detailed examination of interaction and speakers’ orientations,
‘avoiding reference to cognitive processes, discourses or speaker intention. They in turn
have been criticised for taking a narrow and decontextualized focus on short extracts of talk
without attending to broader cultural, social or political issues’ (ibid.: 24).
However, while the questioning of inequalities, ideological hegemonic structures, and
their consequences (particularly for those less powerful or in the margins) is central and
explicit in the ‘top-down’ CDA approaches, other, self-described as ‘bottom-up’, approaches
are also interested in such questioning. As Hall and Davis (this volume) put it, ethnogra-
phers also engage with questions of power and broader social hierarchies; this engagement
helps produce more robust analyses and may also ‘persuade our colleagues in other socially

329
Lia Litosseliti

oriented fields that language, no matter how small, matters to societal organizations of gen-
der and sexuality’ (ibid.).
The development of critical feminist frameworks in particular in the field has also raised
questions about the field’s continuing and/or renewed engagement with feminist theory. I
now turn to the importance of such engagement, which, similarly to Bucholtz (2014) and
others, I argue that is necessary for combining different perspectives, developing new theo-
retical and methodological approaches, and linking more closely to advocacy work. The
example from our work that follows is situated within a broader call by other LGS scholars
for a continuing engagement with feminism and a dialogue with feminist scholarship in other
fields. Many have argued against the field losing sight of its political or ‘analytical activism’
orientation (Holmes and Meyerhoff 2003; Lazar 2007, 2014; Litosseliti 2006; McElhinny
and Mills 2007; Mills and Mullany 2011). Cameron (2006, 2009) notably has warned against
a diminished preoccupation with political collective action as a result of the field’s interest
in local linguistic practice, and with the wider increased emphasis upon questions of iden-
tity – the shift or ‘identity turn’ away from ‘what can be done’ to ‘who I am’ – also echoed
in Lazar’s (2009) discussion of the shift from ‘we-feminism’ to ‘I-feminism’. Cameron has
again recently (2020) argued that the field’s focus on social identity and performance has
shifted attention away from issues of power and real-world inequalities.
In the example of my work with Gill and Favaro (Litosseliti et al. 2019), we reaffirm the
importance of a renewed political voice and motivation for the LGS field, while respond-
ing to calls for new forms of feminist linguistic analysis in a postfeminist landscape (Mills
2012) – that is, analysis that helps us understand the various shifting postfeminist represen-
tations of feminism and femininity (Lazar 2014) vis-à-vis ‘the global neo-liberal discourse
of postfeminism’ (Lazar 2007: 154). In particular, we highlight the analytic value of the con-
cept of ‘postfeminism as a sensibility’ for the LGS field, where it remains under-explored,
in contrast with cultural, media, and gender studies, where it has become central to feminist
scholarship (Gill 2007, 2016; McRobbie 2009; Tasker and Negra 2007). Starting with the
work of scholars in these fields, who take postfeminism as their object of critical inquiry as
well as an analytic category for cultural critique, we illustrate how postfeminist themes or
tropes can play out in contemporary workplace policies. We focus on the analysis of one
text on the idea of ‘agile working’. Using this text, we demonstrate the relationship between
postfeminism and language-use that on the surface celebrates ‘choice’ and ‘diversity’ while
in fact co-opting these ideas for profit and erasing any notions of collective struggle for gen-
der equality (see Litosseliti et al. 2019 for full discussion). While this work takes a similar
approach to other discourse analytic work with an interest in the discursive resources that
structure how gender and gender relations are made sense of, and what inequalities are cre-
ated therein, it does so by allying fine-grained analysis of language with critical work out-
side linguistics on postfeminism as a cultural sensibility. We argue that there is both a need
and an opportunity for LGS scholars to engage with such cross-disciplinary work, in order
to understand current broader cultural patterns, such as how the mainstreaming of work-
place policies on gender diversity can, in a seemingly paradoxical way, have depoliticising
effects. By expanding our theoretical and analytical toolkit we can also understand ‘how
current policy agendas can create new hierarchies and inequalities in the current moment’
(Litosseliti et al. 2019: 17).
Arguably, in the current moment, the potential for our field to continue to learn (as it
always has) in interaction with other fields and disciplines is more important than ever. The
contours and practices of the current sociopolitical landscape certainly present opportunities,
as well as challenges, for critical LGS examination. Some fertile areas for such examination

330
 Engaging with the sociopolitical landscape

that have emerged in the field in recent years and are likely to intensify, include, among
others, those below.

Gender, globalisation, and political economy

Globalisation continues to influence the production of social meaning and the use of lan-
guage within shifting, and sometimes competing, global and local contexts. McElhinny’s
(2007) collection on language, gender, and globalisation has provided theorisations of the
effect of globalisation on language and gender in different national contexts and settings.
Other scholars have looked at the role of English as an index of global prestige, desire, and
sexual identity (e.g. Hall 2009; Leap 2010; Piller and Takahashi 2006). Much of the work
in this area falls under material feminist work in language and political economy, which
investigates the material conditions of language-use and the commodification of language,
in spaces such as workplaces, in late capitalism. For example, Piller and Takahashi (2010)
have explored the gendered work of transnational migrant women and the inequalities and
ideologies therein. As wealth and social inequalities widen, and as labour mobility, migra-
tion, immigration, borders, diaspora, and citizenship continue to be contested and contro-
versial issues, there is significant scope for more LGS work and critique in these areas.

Emerging feminisms/postfeminism, and neoliberalism

As discussed, LGS scholarship in the proliferation of feminisms and in particular post-
feminism vis-à-vis neoliberalism has been emerging (Glapka 2018; Lazar 2014; Litosseliti
et al. 2019), and linguistic analyses of the ways in which feminism is appropriated to serve
and legitimate neoliberalism are much needed. Examples of LGS studies addressing neolib-
eralism have included Inoue (2007) on the linguistic governmentality of neoliberalism and
analyses of gender and language in the neoliberal university (e.g. Goncalves 2019), among
others. LGS scholars are well placed to examine how neoliberalism, austerity, and the vari-
ous processes of de-democratisation (Prügl 2015; Verloo 2011) construct and affect women,
men, LGBT+, and trans communities. They can also contribute further to discussions of
the recent resurgence of interest in feminism, particularly through the use of new forms of
activism, such as social-media technologies (Banyard 2010); and of current issues such as
sexualised culture, domestic and sexual violence, and the anti-gender populist movements
currently becoming more vocal and mainstream (Borba et al. 2020).

Gender and politics/political participation; intersectional approaches to

inequalities
Recent scholarship on gender, language, and politics, gender in media coverage of politi-
cal campaigns, and gender equality policies (e.g. Cameron and Shaw 2016; Lombardo and
Forest 2012; Shaw 2020) illustrates the relevance of cross-disciplinary methods from lin-
guistics and political science for the analysis of women’s political representation. Gender
and politics scholars have been calling for more robust analyses of the role gender and sexu-
ality language/discourses play in constituting political actors and structures (e.g. Celis et al.
2013); they also point to the need to integrate accounts of regimes, institutions, and other
structural dimensions of gender and politics with issues such as identity, self-understand-
ing, and other subjective, micro-level phenomena (ibid.). Further, there is need for more

331
Lia Litosseliti

interdisciplinary intersectional approaches to inequalities, as highlighted by both linguists

(e.g. Charity Hudley 2015) and gender and politics scholars (e.g. Krizsan et al. 2012).

On a different and final note, intersectional approaches to inequalities, including health

inequalities, are also likely to grow in prominence in the post-pandemic landscape. It is early
days at the time of writing, but it is clear that the COVID-19 global pandemic and its manage-
ment are affecting groups differently at the intersections of gender, race, class, ethnicity, and
socioeconomic status; and that language and other practices in the workplace, home, educa-
tion, and public life are being altered in far-reaching ways. Language, gender, and sexuality
scholars are well placed – especially in collaboration with other disciplines – to contribute
new analyses on (to name a few) the language/discourses/narratives of the pandemic; public
health communications and policies; COVID-19 media discourses; interaction among scien-
tists, politicians, and the public; expert discourses in the sense-making and management of
the pandemic; discourses and policies on domestic violence during the pandemic; the effects
of the pandemic on LGBT+ and other marginalised groups; and the discursive and material
dimensions of caring, including shifts in gender roles during and after lockdown measures.

The four chapters in Part V

Sauntson’s chapter sees the self-proclaimed political motivations of queer linguistics as a
particular strength for investigating gender and sexuality identities in educational contexts,
where a social justice orientation is important. The chapter exemplifies how a ‘queer applied
linguistics’ (QAL) framework – as an extension of a critical applied linguistics concern
with inequalities – can highlight social inequality problems around gender and sexuality in
schools. In addition, Sauntson reinforces the argument that queer linguistics lends itself to
being combined with other established methodologies by applying QAL concepts (temporal-
ity, spatiality, normativity) to a critical discourse analysis of her interview and interactional
data with LGBT+ young people and educators. The chapter adds to our understanding of the
discursive construction of restrictive gender binaries and heteronormativity, and their con-
sequences within education settings; and, as importantly, it takes a critical set of tools and
methodologies and combines them to enable, in Sauntson’s words, ‘a greater focus on how
macro-discourses and ideologies around sexuality are embedded and inscribed within micro-
interactions’. In both these ways, it is an excellent illustration of how far the field has come
in engaging with, adapting, and advancing social theory concepts and critical approaches.
Ehrlich and Romaniuk’s chapter is also concerned with exposing larger patterns of gen-
der inequalities, in this case through the application of a ‘textual trajectory’ approach to dis-
course. Textual trajectories is an approach invested in making visible and denaturalising the
workings of social power, control, and inequality. It does this through analysis of the re-con-
textualisation and meaning transformation of texts across social space and time. Through
two examples of analyses in legal and media settings, the authors demonstrate how gen-
dered meanings are transformed and gender inequalities perpetuated as texts travel across
settings and are re-contextualised. In the first example, from a sexual assault trial, Ehrlich
and Romaniuk’s close analysis shows how a strategic act of submission was reconfigured as
a sexual consent signal; in the second example, textual trajectories show how the meaning
of Hilary Clinton’s laughter was transformed as it moved across media environments, from
an interactional strategy to a characteristic of her negative, gendered persona. Crucially,
these meaning transformations were captured by looking at the interplay of texts within a
trajectory, rather than single, one-time instances of texts. The chapter makes a convincing

332
 Engaging with the sociopolitical landscape

case for a method with the potential to identify the basis for damaging ideological assump-
tions in institutional settings, an area where politically engaged scholarship is much needed.
In the third chapter in Part V, Murphy and Palma-Fahey investigate the representation
of gay marriage and the Irish Mammy stereotype in the TV series Mrs Brown’s Boys. The
particular Irish sociocultural and sociopolitical context – especially the Irish Referendum on
marriage equality – is key to understanding the interplay between these representations. The
centrality of the changing sociopolitical environment around gender and sexuality in this
context guides the authors’ use of CDA for their analysis, which they combine with corpus
linguistics methods/the use of concordance lines to look at patterns in representation. The
chapter emphasises the merits of critically engaging with and combining these methodolo-
gies, to usefully bring together micro- and macro-perspectives. At the same time, the media
data presented capture some current political struggles in a rapidly changing society; they
offer an insightful snapshot of the intersections of language, gender, sexuality, and national
identity at the current moment, but which will certainly warrant further investigation as they
shift and evolve.
In the last chapter of Part V, Cameron turns to a different, important set of questions: how
could LGS scholars make their work accessible and applicable outside academia? What
knowledge should scholars seek to communicate, to what audiences, and for what purposes?
What are the benefits and costs of such advocacy and public engagement work? Those
of us who are invited to communicate LGS research to non-academic audiences will find
Cameron’s discussion invaluable. She first maps the current terrain of public engagement,
which is characterised by feminist and non-feminist agendas co-existing and competing in
a public marketplace of ideas; increased pressure on researchers to do public engagement;
a multitude of media platforms being available for this purpose; overall, increased oppor-
tunities for two-way communication between researchers and wider audiences. Cameron
then offers a thoughtful, as well as practical, discussion of some of the problems faced by
researchers engaging with wider audiences: the pressure to demonstrate impact and opti-
mise its evidence by presenting controversial or sensationalised findings that have greater
publicity potential; having their work misrepresented in the media, reinforcing myths and
stereotypes which the research was intended to challenge; and being silenced or attacked
for resisting such misrepresentation or for being perceived as ‘over-critical’ or politically
contentious. The chapter not only offers guidance to scholars for addressing some of these
problems – by suggesting ways for them to correct, contextualise, and complicate their find-
ings – but also puts forward an approach to public engagement that is informed, realistic,
and strategic. As such, it is a worthwhile read for all language, gender, and sexuality schol-
ars who navigate the terrain of advocacy and public engagement.
Together, these chapters offer excellent new contributions to the field’s ongoing engage-
ment with the sociopolitical landscape and opportunities for fruitful reflection on the role of
the LGS researcher in the current moment.

References
Atanga, L., Ellece, S., Litosseliti, L., and Sunderland, J. (eds.) (2012) Special issue: Gender and
language in sub-Saharan African contexts: Research agendas, Gender and Language, 6(1), pp.
1–20.
Baker, P. and Levon, E. (2016) ‘“That’s what I call a man”: Representations of racialised and classed
masculinities in the UK print media’, Gender and Language, 10(1), pp. 106–139.
Banyard, K. (2010) The Equality Illusion. London: Faber and Faber.

333
Lia Litosseliti

Barrett, R. (1999) ‘Indexing polyphonous identity in the speech of African American drag queens’.
In: Bucholtz, M., Liand, A. C., and Sutton, L. A. (eds.) Reinventing Identities. Oxford: Oxford
University Press, pp. 313–331.
Baxter, J. (2008) ‘FPDA – A new theoretical and methodological approach?’ In: Harrington,
K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds.) Gender and Language Research
Methodologies. Houndmills: Palgrave Macmillan, pp. 243–255.
Benwell, B. (2014) ‘Language and masculinity’. In: Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.)
The Handbook of Language, Gender, and Sexuality, 2nd edn. Malden, MA: Wiley-Blackwell, pp.
240–259.
Besnier, N. (2003) ‘Crossing genders, mixing languages: The linguistic construction of transgenderism
in Tonga’. In: Holmes, J. and Meyerhoff, M. (eds.) Handbook of Language and Gender. Malden,
MA: Blackwell, pp. 279–301.
Borba, R. and Ostermann, A. C. (2007) ‘Do bodies matter? Travestis’ embodiment of (trans)gender
identity through the manipulation of the Brazilian Portuguese grammatical gender system’, Gender
and Language, 1(1), pp. 131–147.
Borba, R., Hall, K., and Hiramoto, M. (2020) ‘Editorial: Feminist refusal meets enmity’, Gender and
Language, 14(1), pp. 1–7.
Bucholtz, M. (2011) White Kids: Language, Race, and Styles of Youth Identity. Cambridge, UK:
Cambridge University Press.
Bucholtz, M. (2014) ‘The feminist foundations of language, gender, and sexuality research’. In:
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.) The Handbook of Language, Gender, and
Sexuality, 2nd edn. Malden, MA: Wiley-Blackwell, pp. 23–47.
Bucholtz, M. and Hall, K. (2004) ‘Theorizing identity in language and sexuality research’, Language
in Society, 33(4), pp. 501–547.
Bucholtz, M. and Hall, K. (2005) ‘Identity and interaction: A sociocultural linguistic approach’,
Discourse Studies, 7(4–5), pp. 585–614.
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Cameron, D. (1995) Verbal Hygiene. Abingdon, UK: Routledge.
Cameron, D. (1997) ‘Performing gender identity: Young men’s talk and the construction of heterosexual
masculinity’. In: Johnson, S. and Meinhof, U. H. (eds.) Language and Masculinity. Oxford and
Cambridge, MA: Blackwell, pp. 47–64.
Cameron, D. (2000) ‘Styling the worker: Gender and the commodification of language in the
globalized service economy’, Journal of Sociolinguistics, 4(3), pp. 323–347.
Cameron, D. (2006) On Language and Sexual Politics. London: Routledge.
Cameron, D. (2009) ‘Theoretical issues for the study of gender and spoken interaction’. In: Pichler, P.
and Eppler, E. (eds.) Gender and Spoken Interaction. Basingstoke: Palgrave, pp. 1–17.
Cameron, D. (2020) ‘Language and gender: Mainstreaming and the persistence of patriarchy’,
International Journal of the Sociology of Language, 263(263), pp. 25–30.
Cameron, D. and Kulick, D. (2003) Language and Sexuality. Cambridge: Cambridge University Press.
Cameron, D. and Shaw, S. (2016) Gender, Power and Political Speech: Women and Language in the
2015 UK General Election. London: Palgrave Macmillan.
Canakis, C., Kantsa, V., and Yiannakopoulos, K. (eds.) (2010) Language and Sexuality (through And)
beyond Gender. Newcastle upon Tyne, UK: Cambridge Scholars.
Candelas de la Ossa, A. (2019) ‘Exceptionalising intersectionality: A corpus study of implied readership
in guidance for survivors of domestic abuse’, Gender and Language, 13(2), pp. 224–250.
Cashman, H. (2018) Queer, Latinx, and Bilingual: Narrative Resources in the Negotiation of Identities.
London and New York, NY: Routledge.
Celis, K., Kantola, J., Waylen, G., and Weldon, S. L. (2013) ‘Introduction: Gender and politics: A
gendered world, a gendered discipline’. In: Waylen, G., Celis, K., Kantola, J., and Weldon, S. L.
(eds.) The Oxford Handbook of Gender and Politics. New York, NY: Oxford University Press, pp.
1–19.

334
 Engaging with the sociopolitical landscape

Charity Hudley, A. H. (2015) ‘Language and racialization’. In: García, O., Flores, N., and Spotti, M.
(eds.) The Oxford Handbook of Language and Society. New York, NY: Oxford University Press,
pp. 381–402.
Chouliaraki, L. and Fairclough, N. (1999) Discourse in Late Modernity: Rethinking Critical Discourse
Analysis. Edinburgh: Edinburgh University Press.
Christie, C. (2000) Gender and Language: Towards a Feminist Pragmatics. Edinburgh: Edinburgh
University Press.
Coates, J. (2003) Men Talk: Stories in the Making of Masculinity. Oxford: Blackwell.
Collins, P. (2000) Black Feminist Thought: Knowledge, Consciousness, and the Politics of
Empowerment, 2nd edn. New York: Routledge.
Crenshaw, K. (1991) ‘Mapping the margins: Intersectionality, identity politics, and violence against
women of color’, Stanford Law Review, 43(6), pp. 1241–1299.
Eckert, P. and McConnell-Ginet, S. (2013) Language and Gender, 2nd edn. New York, NY: Cambridge.
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.) (2014) The Handbook of Language, Gender, and
Sexuality, 2nd edn. Malden, MA: Blackwell.
Gaudio, R. P. (2009) Allah Made Us: Sexual Outlaws in an Islamic African City. Malden, MA:
Wiley-Blackwell.
Gill, R. (2007) ‘Postfeminist media culture: Elements of a sensibility’, European Journal of Cultural
Studies, 10(2), pp. 147–166.
Gill, R. (2016) ‘Post-postfeminism? New feminist visibilities in postfeminist times’, Feminist Media
Studies, 16(4), pp. 610–630.
Glapka, E. (2018) ‘Postfeminism – For whom and by Whom? Applying Discourse Analysis in Research
on Body and Beauty (the Case of Black Hair)’, Gender and Language, 12(2), pp. 242–268.
Gonçalves, K. (2019) ‘“What are you doing here, I thought you had a kid now?” the stigmatisation
of working mothers in academia – A critical self-reflective essay on gender, motherhood and the
neoliberal academy’, Gender and Language, 13(4), pp. 469–487.
Gray, J. and Cooke, M. (eds.) (2018) Special Issue: Intersectionality, language and queer lives, Gender
and Language, 12(4), pp. 437–458.
Hall, K. (1995) ‘Lip service on the fantasy lines’. In: Hall, K. and Bucholtz, M. (eds.) Gender
Articulated: Language and the Socially Constructed Self. New York, NY: Routledge, pp. 183–216.
Hall, K. (2005) ‘Intertextual sexuality: Parodies of class, identity, and desire in liminal Delhi’, Journal
of Linguistic Anthropology, 15(1), pp. 125–144.
Hall, K. (2009) ‘Boys’ talk: Hindi, moustaches, and masculinity in New Delhi’. In: Pichler, P. and
Eppler, E. (eds.) Gender and Spoken Interaction. Basingstoke, UK: Palgrave Macmillan, pp.
139–162.
Hancock, A.-M. (2007) ‘When multiplication doesn’t equal quick addition: Examining intersectionality
as a research paradigm’, Perspectives on Politics, 5(1), pp. 63–79.
Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds.) (2008) Gender and Language
Research Methodologies. Basingstoke: Palgrave.
Holmes, J. (2006) Gendered Talk at Work: Constructing Gender Identity through Workplace Discourse.
Malden, MA: Blackwell.
Holmes, J. (2014) ‘Language and gender in the workplace’. In: Ehrlich, S., Meyerhoff, M., and
Holmes, J. (eds.) The Handbook of Language, Gender, and Sexuality, 2nd edn. Malden, MA:
Wiley-Blackwell, pp. 431–451.
Holmes, J. and Meyerhoff, M. (2003) ‘Different voices, different views: An introduction to current
research in language and gender’. In: Holmes, J. and Meyerhoff, M. (eds.) The Handbook of
Language and Gender. Oxford: Blackwell, pp. 1–17.
Inoue, M. (2007) ‘Language and gender in an age of neoliberalism’, Gender and Language, 1(1), pp.
79–91.
Johnson, S. and Meinhof, U. (eds.) (1997) Language and Masculinity. Oxford and Cambridge, MA:
Blackwell.

335
Lia Litosseliti

Jones, L. (2012) Dyke/Girl: Language and Identities in a Lesbian Group. Basingstoke: Palgrave
Macmillan.
Kiesling, S. F. (2005) ‘Homosocial desire in men’s talk: Balancing and recreating cultural discourses
of masculinity’, Language in Society, 34(5), pp. 695–726.
Kitzinger, C. (2000) ‘Doing feminist conversation analysis’, Feminism and Psychology, 10(2), pp.
163–193.
Krizsán, A., Skeije, H., and Squires, J. (2012) Institutionalizing Intersectionality: The Changing
Nature of European Equality Regimes. Basingstoke, UK: Palgrave Macmillan.
Kulick, D. (1998) Travesti: Sex, Gender, and Culture among Brazilian Transgendered Prostitutes.
Chicago, IL: University of Chicago Press.
Kulick, D. (2000) ‘Gay and lesbian language’, Annual Review of Anthropology, 29(1), pp. 243–285.
Lakoff, R. (1975) Language and Woman’s Place. New York, NY: Harper & Row.
Lanehart, S. (ed.) (2009) African American Women’s Language: Discourse, Education, and Identity.
Newcastle upon Tyne: Cambridge Scholars Publishing.
Lazar, M. M. (ed.) (2005) Feminist Critical Discourse Analysis: Gender, Power and Ideology in
Discourse. Basingstoke: Palgrave.
Lazar, M. M. (2007) ‘Feminist critical discourse analysis: Articulating a feminist discourse praxis’,
Critical Discourse Studies, 4(2), pp. 141–164.
Lazar, M. M. (2009) ‘Entitled to consume: Postfeminist femininity and a culture of post-critique’,
Discourse and Communication, 3(4), pp. 371–400.
Lazar, M. M. (2014) ‘Recuperating feminism, reclaiming femininity: Hybrid postfeminist I-dentity in
consumer advertisements’, Gender and Language, 8(2), pp. 205–224.
Leap, W. L. (2010) ‘Globalization and gay language’. In: Coupland, N. (ed.) The Handbook of
Language and Globalisation. Malden, MA: Blackwell, pp. 555–574.
Leap, W. L. and Motschenbacher, H. (2012) ‘Launching a new phase in language and sexuality
studies’, Journal of Language and Sexuality, 1(1), pp. 1–14.
Levon, E. (2010) Language and the Politics of Sexuality: Lesbians and Gays in Israel. Basingstoke,
UK: Palgrave Macmillan.
Levon, E. (2015) ‘Integrating intersectionality in language, gender, and sexuality research’, Language
and Linguistics Compass, 9(7), pp. 295–308.
Levon, E. and Mendes, R. B. (2016) Language, Sexuality, and Power: Studies in Intersectional
Sociolinguistics. Oxford: Oxford University Press.
Litosseliti, L. (2006) Gender and Language: Theory and Practice. London: Hodder Education.
Litosseliti, L. (ed.) (2018) Research Methods in Linguistics, 2nd edn. London: Bloomsbury.
Litosseliti, L., Gill, R., and Favaro, L. (2019) ‘Postfeminism as a critical tool for gender and language
(study)’, Gender and Language, 13(1), pp. 1–22.
Lombardo, E. and Forest, M. (2012) The Europeanization of Gender Equality Policies: A Discursive-
Sociological Approach. Basingstoke: Palgrave.
McElhinny, B. (ed.) (2007) Words, Worlds, and Material Girls: Language, Gender, Globalisation.
Berlin: Mouton de Gruyter.
McElhinny, B. (2014) ‘Theorizing gender in sociolinguistics and linguistic anthropology: Toward
effective interventions in gender inequity’. In: Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.) The
Handbook of Language, Gender, and Sexuality, 2nd edn. Malden, MA: Wiley-Blackwell, pp. 48–67.
McElhinny, B. S. and Mills, S. (2007) ‘Launching studies of gender and language in the early 21st
century’, Gender and Language, 1(1), pp. 1–13.
McEntee-Atalianis, L. J. (2019) Identity in Applied Linguistics Research. London: Bloomsbury
Academic.
McRobbie, A. (2009) The Aftermath of Feminism: Gender, Culture and Social Change. London: Sage.
Mendoza-Denton, N. (2008) Homegirls: Language and Cultural Practice among Latina Youth Gangs.
Oxford: Blackwell.

336
 Engaging with the sociopolitical landscape

Milani, T. M. (ed.) (2015) Language and Masculinities: Performances, Intersections, Dislocations.

New York, NY: Routledge.
Milani, T. and Lazar, M. M. (2017) ‘Seeing from the South: Discourse, gender and sexuality from
southern perspectives’, Journal of Sociolinguistics, 21(3), pp. 307–319.
Mills, S. (2012) Gender Matters: Feminist Linguistic Analysis. London: Equinox.
Mills, S. and Mullany, L. (2011) Language, Gender and Feminism. Theory, Methodology and Practice.
London: Routledge.
Motschenbacher, H. (2010) Language, Gender, and Sexual Identity: Poststructuralist Perspectives.
Amsterdam: John Benjamins.
Motschenbacher, H. and Stegu, M. (eds.) (2013) Special issue on queer linguistics, Discourse and
Society, 24(5), pp. 536–552.
Nash, J. (2008) ‘Re-thinking intersectionality’, Feminist Review, 89(1), pp. 1–15.
Okamoto, S. and Shibamoto Smith, J. (2004) Japanese Language, Gender, and Ideology: Cultural
Models and Real People. New York, NY: Oxford University Press.
Ostermann, A. C. (2003) ‘Communities of practice at work: Gender, facework, and the power of
habitus at an all female police station and a feminist crisis intervention centre in Brazil’, Discourse
and Society, 14(4), pp. 473–505.
Piller, I. and Takahashi, K. (2006) ‘A passion for English: Desire and the language market’. In:
Pavlenko, A. (ed.) Bilingual Minds: Emotional Experience, Expression and Representation.
Clevedon, UK: Multilingual Matters, pp. 59–83.
Piller, I. and Takahashi, K. (2010) ‘At the intersection of gender, language, and transnationalism’. In:
Coupland, N. (ed.) The Handbook of Language and Globalization. Malden, MA: Blackwell, pp.
540–554.
Prügl, E. (2015) ‘Neoliberalising feminism’, New Political Economy, 20(4), pp. 614–631.
Queen, R. (2014) ‘Language and sexual identity’. In: Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.)
The Handbook of Language, Gender, and Sexuality, 2nd edn. Malden, MA: Wiley-Blackwell, pp.
203–219.
Romero, Y. (2017) ‘Developing an intersectional framework: Engaging the decenter in language
studies’, Critical Inquiry in Language Studies, 14(4), pp. 1–27.
Shankar, S. and Cavanaugh, J. R. (2012) ‘Language and materiality in global capitalism’, Annual
Review of Anthropology, 41(1), pp. 355–369.
Shaw, S. (2020) Women, Language and Politics. Cambridge: Cambridge University Press.
Smooth, W. (2011) ‘Standing for women? Which women? The substantive representation of women’s
interests and the research imperative of intersectionality’, Politics and Gender, 7(3), pp. 436–441.
Speer, S. A. (2005) Gender Talk: Feminism, Discourse and Conversation Analysis. London: Routledge.
Tasker, Y. and Negra, D. (2007) ‘Introduction: Feminist politics and postfeminist culture’. In: Tasker,
Y. and Negra, D. (eds.) Interrogating Postfeminism: Gender and the Politics of Popular Culture.
Durham, NC: Duke University Press, pp. 1–26.
Toerien, M. and Kitzinger, C. (2007) ‘Emotional labour in action: Navigating multiple involvements
in the beauty salon’, Sociology, 41(4), pp. 645–662.
Trechter, S. (2014) ‘A marked man: The contexts of gender and ethnicity’. In: Ehrlich, S., Meyerhoff,
M., and Holmes, J. (eds.) The Handbook of Language, Gender, and Sexuality, 2nd edn. Malden,
MA: Wiley-Blackwell, pp. 335–352.
Verloo, M. (2011) ‘Gender equality policies as interventions in a changing world’, Keynote Lecture at
the Second Gender and Politics European Consortium for Political Research (ECPR) Conference,
Budapest, January 13.
Weiss, G. and Wodak, R. (eds.) (2003) Critical Discourse Analysis: Theory and Interdisciplinarity in
Critical Discourse Analysis. London: Palgrave.
Weldon, S. L. (2008) ‘The concept of intersectionality’. In: Mazur, A. G. and Goertz, G. (eds.) Gender
and Concepts. Cambridge, UK: Cambridge University Press.

337
Lia Litosseliti

Zimman, L. and Hall, K. (2010) ‘Language, embodiment, and the “third sex”’. In: Llamas, C. and
Watt, D. (eds.) Language and Identities. Edinburgh: Edinburgh University Press, pp. 166–178.
Zimman, L. and Hall, K. (2016) ‘Language, gender and sexuality’. In: Aronoff, M. (ed.) Oxford
Bibliographies in Linguistics. New York, NY: Oxford University Press, pp. 166–178.
Zimman, L., Davis, J., and Raclaw, J. (eds.) (2014) Queer Excursions: Retheorizing Binaries in
Language, Gender, and Sexuality. New York, NY: Oxford University Press.

338
 22
Applying queer theory to
language, gender, and sexuality
research in schools1
Helen Sauntson

Introduction
This chapter explores how queer theory, and the associated linguistic framework of queer
applied linguistics, can be used to fruitfully investigate the ways that different gender and
sexuality identities are enacted through language. I exemplify a queer applied linguistics
approach by applying it to the analysis of interview and interactional data taken from school
settings. Queer theory is largely seen as an interdisciplinary theoretical approach that uti-
lises methods from a range of other paradigms in order to problematise and destabilise
taken-for-granted, socially sanctioned notions of gender and sexuality. Language and sexu-
ality work from the 1990s onwards has been heavily influenced by aspects of queer theory.
The study of language and sexuality has grown from a fairly restricted focus on lexical
studies of gay men’s slang to a broader examination of the ways in which language can be
used to index sexuality in various contexts of use. Furthermore, the field has moved from
an almost exclusive focus on gay men to investigating a range of sexual identities. The
incorporation of elements of queer theory has involved a critical reconsideration of what is
meant by ‘sexuality’ and ‘sexual identity’ and their relationship to linguistic forms, context,
and use. In Livia and Hall’s (1997) landmark publication, Queerly Phrased, the term ‘queer’
is used to refer to non-normative or resistant identity performances which may or may not
be tied to a particular category of sexual identity. Conceptualisations of ‘identity’ are a key
concern of queer theory, as they are in other sociocultural and critical methodologies used
in the study of language, gender, and sexuality.
In the following sections, I describe the development of queer theory and its application
to the study of gender and sexuality within linguistics – queer linguistics. I consider some of
the oft-cited criticisms of queer theory and offer a development of a ‘queer applied linguis-
tics’ (QAL) framework which works to counter such criticisms. Within QAL, I consider the
key concepts of temporality, spatiality, and normativity and their usefulness in accounting
for the ways in which different gender and sexuality identities emerge in, and are enacted
through, language. I then illustrate the QAL framework by examining some linguistic data
(interviews with teachers and young people, and recorded extracts of classroom interaction)
taken from school settings. In analysing this data, I draw on critical discourse analysis as

339
Helen Sauntson

a method that can be used within an overarching QAL approach. In the final section of the
chapter, I outline other applications of queer theory to research on language, gender, and
sexuality beyond education and offer suggestions for how the approach may be developed
and utilised in the future.

Queer theory and queer linguistics

Queer linguistics is underpinned by queer theory, which takes ‘heteronormativity’ as its
main object of critical investigation. Heteronormativity is defined by Cameron (2005: 489)
as ‘the system which prescribes, enjoins, rewards, and naturalises a particular kind of het-
erosexuality – monogamous, reproductive, and based on conventionally complementary
gender roles – as the norm on which social arrangements should be based’. Queer theory
scholars such as Halperin (1993) have pointed out that, importantly, there is actually no such
thing as ‘queer theory’ in a singular form. Rather, queer theory consists of many different
approaches. But what all queer theory work has in common is its critical investigation of
heteronormativity and its resistance to presenting gender as an a priori category. Instead,
queer theory interrogates the underlying preconditions of gender and sexuality identity, and
how these may be enacted and formulated in discourse.
Within queer theory, Butler’s (1990, 1993, 2004) theories of performativity have been of
particular importance for questioning socially sanctioned concepts of normality in relation
to gender and sexuality. Queer theory presents a unified view of gender and sexuality in that
it recognises that cultural ideologies of gender normativity are bound up with assumptions
of heterosexuality. Butler (1990) develops this notion in her claims that heterosexuality is
naturalised by the performative repetition of normative gender identities. Butler surmises
that the categories of gender and sexuality have been ‘causally entangled in knots that must
be undone’ (1998: 225–226). Thus, the principle of queer theory that claims an integral and
definitional relationship between gender and sexuality is of central importance to queer
linguistics and its applications.
Queer linguistics draws on the principles of queer theory outlined above and applies
them to the study of language. Motschenbacher and Stegu (2013: 522) define queer lin-
guistics in concise terms as ‘critical heteronormativity research from a linguistic point of
view’. Most definitions and explanations of queer linguistics are based around the concept
of heteronormativity and use it as a theoretical and analytical starting point. Cameron and
Kulick (2003) assert, importantly, that queer linguistics can be applied to the critical inves-
tigation of heterosexual identities and desires as well as those that are sexually marginal-
ised. They note that research on language and sexual minorities tends to focus on analysing
linguistic manifestations of homophobia and other kinds of sexuality-based discrimination,
whilst queer linguistics more broadly encompasses an analysis of discursive formations of
all sexual identities, including heterosexualities. Part of this analysis involves exploring the
linguistic means by which heterosexuality comes to be seen as the assumed default sexual-
ity whilst other sexualities become marked as ‘non-normative’. Furthermore, it is certain
kinds of heterosexualities that are privileged and this is also a concern of queer linguistics
(also discussed by Leap and Motschenbacher 2012). What we can take from queer linguis-
tics is that there also needs to be more critical scrutiny of how privileged forms of hetero-
sexuality are discursively formed in applied contexts with a view to ultimately challenging
and changing such practices. With this in mind, intersectionality is an important concept
which has been developed within queer theory in order to acknowledge and understand

340
 Applying queer theory

how ‘sexuality’ can involve more than the hetero/homo continuum. For example, identities
and relationships may be discursively constructed as normal/not normal in relation to other
social dimensions of identity such as ethnicity, age, and social class.
Motschenbacher (2011) and Motschenbacher and Stegu (2013) argue that queer linguis-
tics lends itself well to an eclectic combination of linguistic analytical methods (or meth-
odological pluralism) in order to provide mutually qualifying positions. Leap (2018: 10)
has also referred to a ‘scavenger methodology’ as being particularly appropriate for queer
inquiry across a range of disciplines. In work which applies queer linguistics, various estab-
lished methods are therefore drawn on in order to analyse different types of language data.
These methods include corpus linguistics, critical discourse analysis, conversation analy-
sis, multi-modal discourse analysis, linguistic landscapes, and interactional sociolinguis-
tics (and this list is not exhaustive).2 Within the scope of this chapter, it is not possible to
review or exemplify applications of all of these methods to work on language, gender, and
sexuality. I therefore take one approach – critical discourse analysis (discussed later) – as
an illustration of how an established methodological and analytical framework can be used
within queer linguistics to explore how gender and sexuality identities are enacted through
language in school contexts.

Criticisms of queer linguistics and the development of QAL

Motschenbacher (2011) documents the main criticisms levelled against queer linguistics –
that it has restricted relevance and real-life empirical applications and that it raises problem-
atic issues concerning political agency. Motschenbacher counters such claims by arguing that
queer linguistics, contrary to the critical claims, does not focus exclusively on documenting
gay and lesbian aspects of language. Its focus is on how all sexual identities are construed
through discourse, but with particular attention paid to how heteronormativity becomes
materialised through discourse as the dominant sexuality. In other words, attention is paid
to how non-heteronormative identities are often marginalised through particular discursive
processes which are identified through the application of queer linguistic approaches. The
effects of such marginalisation can arguably be damaging for heterosexual-identified peo-
ple as well as members of sexual minorities. The political motivations underlying queer
linguistics are thus self-proclaimed and seen as a strength rather than a limitation. It has
been argued that there are tensions between queer linguistics as an identity-questioning
approach and the focus on lesbian, gay, bisexual, and transgender (LGBT) issues, which
instead suggests an identity-affirming theoretical background. However, counter-arguments
claim that the notion of identity does not need to be dispensed with if taking a queer lin-
guistic approach. In terms of its empirical applicability, Motschenbacher counters criticisms
by arguing that queer linguistics lends itself well to utilising linguistic methods that have
already been established within other areas of applied linguistics, such as those mentioned
in the previous section (and many of which are covered throughout this volume), and apply-
ing them to queer research purposes.
One such way forward may be to embed queer theory into the existing approach of
critical applied linguistics and apply it to analysing discursive constructions of heteronor-
mativity in specific contexts. Critical applied linguistics (CAL) has been defined as ‘the
practice of applied linguistics grounded in a concern for addressing and resolving prob-
lems of inequality’ (Hall, Smith, and Wicaksono 2017: 18). According to Hall et al., critical
applied linguistics is an approach to language study which addresses a specific problem for

341
Helen Sauntson

the benefit of a defined set of ‘clients’ or end-users. They argue that the identification of a
‘real-world’ problem should be informed by the people who experience it. But their defini-
tion does not address key theoretical issues which, I argue below, have emerged from queer
linguistics and are potentially useful to the field of applied linguistics more broadly. Data
discussed in this chapter identifies a set of social inequality problems related to gender and
sexuality in schools which can be investigated through a queer theory-informed analysis
of linguistic practices. We may loosely define ‘queer applied linguistics’ (QAL), then, as
CAL which is informed by queer theory/queer linguistics and which is applied to address-
ing social concerns with inequalities around gender and sexuality. Like CAL, QAL has a
social justice orientation. Drawing on Motschenbacher’s (2011) conceptual developments
of queer linguistic approaches, I argue that three key overlapping issues have emerged as
relevant to work which utilises a QAL framework. These issues are: temporality; space; and
normativity. ‘Real-world’ problems are always situated in time and space, and are therefore
temporally and spatially construed. These first two issues map onto what Motschenbacher
calls contextuality (spatiality) and fluidity (temporality) in his identification of dimensions
that a queer approach to applied linguistics must be able to grasp. What is considered a
‘problem’ is also at least partially constructed by culturally and socially situated notions of
‘normativity’ – a concept which, according to Motschenbacher, needs to be theorised more
fully in queer linguistics. I briefly consider these key QAL concepts in the remainder of this
section before applying them to the critical discourse analysis of school-based language data
in the later sections.
Leap (2020) indicates that the temporal signification of ‘before’ is of central importance
and should be critically interrogated in work which focuses on language and sexuality. Leap
proposes the approach of ‘queer historical sociolinguistics’ for engaging with issues of time
in relation to language and sexuality. Significantly, Leap argues that temporal narratives can
frame historical events in such a way that the narratives obfuscate problematic issues which
persist after the event in question. Leap’s work enables us to see how discriminatory lan-
guage practices around gender and sexuality are real-world problems which are historical
(temporal) constructs. Their temporal and contingent nature means that they are inherently
unstable and subject to narrative retellings which imbue them with diverse meanings and
effects in different temporal and spatial sites. Importantly, queer time can also function to
exclude, restrict, and erase, rather than to open up possibilities that lie beyond the ‘normal’.
Leap argues that narratives of language and sexuality that end in ‘triumph’ may actually
disguise homophobic (and other discriminatory) practices by focusing only on positive out-
comes and rendering invisible continuing struggles experienced by certain LGBT+ individ-
uals and communities. Examples include the ‘triumph’ narratives that circulate in popular
discourse around legal and social ‘successes’ relating to sexuality such as the passing of
same-sex marriage laws and decriminalisation.
This notion of temporality links to another important principle of queer theory discussed
by Hall (2013) – that heteronormativity itself is not stable across time and space. Hall (2013:
638) argues that ‘the social meaning granted to heteronormativity, even if its idealisation
persists, is always shifting across the interactions of those associated with it’. Although
temporality and spatiality are probably true of any so-called resistance movement, within
queer theory, the ways in which they have the potential to destabilise heterosexuality are
important.
Milani (2013) also draws attention to the need for temporal analysis in the discussion of
normativity (another key concept in QAL), asserting that normativity as a concept is never
stable and is always temporally and spatially contingent. It is from this position that Hall

342
 Applying queer theory

argues that temporality (especially in relation to heteronormativity) should be a key concern

of queer linguistics.
In another seminal queer theory work, In A Queer Time and Place (2005), Halberstam
argues that time and space are subjected to the same kinds of naturalisation processes as
discursive practices. Therefore, ‘queer time’ can both construct and resist normative identi-
ties. Drawing on Halberstam’s (2005) work and transferring it to the study of language and
sexuality, Leap and Motschenbacher (2012: 7) state:

To study language, sexuality and queer temporality is to ask questions about linguistic
and sexual practices that take place in spatial and temporal domains that lie outside of
the ordinary, the familiar, and the ‘normal’.

Such ideas have led to a recent incorporation of examinations of space and time in relation
to language and sexuality. Milani (2013), for example, examines multimodal (linguistic and
spatial) constructions of sexual identity in a university environment. He argues that using
multimodal methods of discourse analysis, and incorporating a semiotic analysis of sexual-
ity and space, could further our understanding of the relationship between language and
sexuality and how it operates in educational and other contexts.
A third key concept in QAL, and in queer theory more broadly, is that of normativity.
Motschenbacher (2014) has been critical of the lack of theorising around the term ‘normativ-
ity’ (including heteronormativity) in queer linguistics. He argues that much recent language
and sexuality work makes frequent reference to the concept of normativity without fully
explaining or theorising it. Motschenbacher argues that speakers have a tendency to orient
towards a shared notion of normativity in their language practices. However, normativity
itself is not stable and a way of theorising it is to view it as constantly shifting and relative to
spatio-temporal contexts. Therefore, the three QAL concepts identified here always overlap.
In sum, QAL is problem-focused and has a social justice orientation relating specifi-
cally to gender and sexuality issues. It takes critical heteronormativity analysis as its central
focus and recognises that key issues within contextualised examinations of heteronormativ-
ity need to take into account temporal and spatial understandings and realisations of nor-
mativity. I discuss these issues below with illustrations from linguistic data taken from UK
secondary school contexts.

Applications of the approach: language, gender, and sexuality in

schools
To date, education has been a key area of application of queer theory to the study of lan-
guage. Nelson (2012) asserts that queer linguistics is an appropriate approach to use to
examine the ways in which particular discourses of gender and sexuality are produced in
school contexts through the deployment of specific linguistic practices. In previous work
(Sauntson 2012, 2018), I illustrate how linguistic methods of analysis can be used along-
side queer theory to critically examine the discursive constructions of ‘normal’ and ‘queer’
gender and sexuality in school classrooms. I show how incorporating some of the principles
of queer theory into the types of analyses already used can help to uncover the ways in
which heterosexuality is naturalised and how other forms of sexual and gender identity are
‘queered’ in school contexts. QAL can therefore provide a helpful theoretical framework for
examining how normative and non-normative constructions of sexual identity are enacted
through and inscribed in language practices in schools, and how these language practices

343
Helen Sauntson

may effect particular discourses of sexuality. I incorporate the key conceptual elements of
temporality, spatiality, and normativity within QAL when researching language and sexual-
ity in schools. In practical terms, this means examining how and why speakers and writers
orient towards particular ideas of normativity through their language practices, and how
normative genders and sexualities are represented through the language used in school set-
tings. The approach also involves being aware of how the very notion of normativity may
shift in relation to spatial and temporal contexts.
Research in a range of international contexts identifies schools as sites that are charac-
terised by assumptions that students are heterosexual (heteronormativity) and by an atmos-
phere of homophobia, thus highlighting a ‘real-world’ problem. Despite recent legislative
changes in the UK and US, for example, research indicates that heteronormativity and
homophobia continue to pervade schools and that the effects of this are damaging for young
people identifying or perceived to be LGBT+ (Kosciw et al. 2015; McDermott et al. 2008;
Bradlow et al. 2017). However, little attention has been paid to the linguistic dimensions
of these problems. The application of queer theory to the study of language and education
using a QAL framework can arguably be particularly helpful for exploring the often com-
plex ways in which homophobia and heteronormativity are enacted in school contexts, as
illustrated in the sections that follow.

Methodology
In a project which explores the role played by language in constructing sexual identities in
schools, CDA is used within a QAL approach to uncover linguistic practices in the data-
sets of: 20 semi-structured interviews with LGBT+-identified3 young people (aged 13–25)
who attend or have recently attended school; 14 semi-structured interviews with educators
working in a range of 11 to 16 and 11 to 18 schools in two areas of the UK; 4 (x 1-hour)
Relationships and Sex Education (RSE) classroom interactions. In the interviews, all par-
ticipants discuss their perceptions of how issues around sexual diversity are handled in
their schools and what they think the key issues are. Additionally, the young people reflect
on their experiences of, and attitudes towards, school in relation to their LGBT+ sexual
identities.
In CDA, various types of linguistic analysis are used to uncover power relations and
ideologies. There are various formal linguistic features which can be focused on in applying
CDA, such as (but not limited to): lexical items; metaphors; evaluative language (e.g. seman-
tic fields and adjectives); intertextual references; grammatical and syntactic structures. In
the data presented in this chapter, I focus mainly on particular lexical items and intertextual
references which Pakula et al. (2015) have identified as ‘gender-triggering points’ (GTPs)
in texts. GTPs are based on Sunderland et al.’s (2002) concept of ‘gender critical points’
in classroom interaction in which specific references draw attention to gender and make it
relevant to the lesson in some way. A GTP, according to Pakula et al. (2015), happens when
gender is negotiated into relevance through the interaction taking place. Typical examples
identified by the authors might include: gender roles being ascribed to characters or social
actors; explicit linguistic instantiations of heterosexuality or heteronormativity; stereotypi-
cal or non-stereotypical representations of femininity and masculinity. I incorporate ‘sexu-
ality’ into this framework so that GTPs become GSTPs (‘gender and sexuality triggered
points’) in the current analysis. And I subsequently add any explicit or implicit linguistic
references to sexuality as indicators of GSTPs in the data (some examples are discussed in

344
 Applying queer theory

the next section). I also pay particular attention to GSTPs which also signalled anything
relating to temporality, spatiality, and normativity. Therefore, the critical analysis of GSTPs
in the data is used to explore the QAL concepts of temporality, spatiality, and normativ-
ity. Arguably, the use of this kind of CDA within an overarching QAL approach enables a
greater focus on how macro-discourses and ideologies around sexuality are embedded and
inscribed within micro-interactions.
After presenting the QAL-informed critical discourse analysis around temporality and
spatiality in the interview data, I then move to using examples from the RSE classroom
interaction data to explore the dimension of normativity as it manifests in the language used
in the data. This is done intentionally to give an idea of the different datasets used in the
study. The key QAL concepts of temporality, spatiality, and normativity emerge as salient
factors in the interviewees’ responses and narratives about their school experiences. These
concepts also emerge in the analysis of the RSE classroom interaction data. In the sections
that follow, I illustrate the importance of these three concepts and explore how they emerge
as salient in the interview and classroom interaction datasets. In doing so, I aim to exemplify
a QAL approach which draws on the specific analytic method of corpus-based discourse
analysis. As stated earlier, the concepts do overlap – they are presented in separate sections
in this chapter purely for illustrative purposes.

Temporality
In the interview data taken from the school-based project, the young people and teacher
participants in the study refer explicitly to school as a temporal (and spatial) state, as in the
examples below (temporal/historical indicators are underlined). In the first example, Lauren
(a history teacher) responds to a question about whether she had been given any training on
raising awareness about sexual diversity issues in schools. In the response, Lauren locates
her own stance towards teaching about sexual diversity in schools through a historical refer-
ence to the introduction of Section 28 in 1988.4

Lauren: I started teaching when clause 28 came in and so you know especially again
teaching in Brighton it was all you weren’t allowed to say it I mean I remember clause
28 was about not promoting it really it was bizarre it was all shite

Lauren reflects on how the introduction of the legislation had the effect of silencing talk
about same-sex relationships in schools and this reflection is accompanied by negative
evaluation. The very fact that Lauren raises and discusses this historical event suggests
that it continues to have salience for her in relation to sexual diversity issues in schools. It
is also an example of what Leap (2018) refers to as ‘spectrality’ in historical sociolinguis-
tics – past events and discursive formations ‘haunting’ and continuing to shape sexuality
in the present day. A similar phenomenon occurs in the next example in which Natalie (an
English teacher) makes a historical reference to ‘teachers of my age’ (i.e. teachers who
have many years’ experience in the profession) and their perceived temporally located
belief that open discussion about homosexuality continues to be illegal in schools. Natalie
reflects on her belief that the passing of Section 28 had the opposite effect of its intended
one of making teachers talk openly about sexuality more than they had done in the past.
She laments the lack of explicit talk about sexuality that she believes characterises the
present time.5

345
Helen Sauntson

Natalie: I was reading something recently there was an article in was it the TES6 and
somebody had done a big survey with this idea that teachers of my age thought you
weren’t allowed to talk about homosexuality […] I was reading this and thinking gosh
I wasn’t aware of that it almost made me do it more
[…]
I think there was a lot more awareness about that when I mean you know because we
think we mustn’t do that we have that discussion I don’t think as an English teacher we
talk about those sorts of things in the same way any more

However, Natalie does not take this as an indicator of young teachers’ regressive attitudes
towards LGBT+ identities, stating that she believes younger teachers to be ‘more liberal’
than some of their older counterparts. But, importantly, she perceives less of a willingness
to ‘take risks’ in the present school climate. Thus, open positive discussion about same-
sex relationships and LGBT+ identities is temporally located in the past as she indicates
here:

Natalie: I think they’re probably very much more liberal than perhaps thirty forty years
ago about homosexuality but in terms of text content they are less willing to take risks

The LGBT+-identified young people in the study likewise made numerous references to
temporality in relation to certain events and issues raised in the school context. In the first
example below, Josh reports a teacher in the present time referring to AIDS as a ‘gay dis-
ease’ located in the space of gay clubs. This spectral homophobic discourse of AIDS as a
gay disease is brought into the present in the spatio-temporal context of Josh’s recent experi-
ence of RSE lessons at school.

Josh: one thing that shocked me is that they were talking to us all about these dangers of
sex and stuff in the sex education part and then they’d take us to a disease section and
the teacher there she gets very into it and she talks about AIDS and that was the only
time when she talked about gays and stuff but she sort of talked about like AIDS was a
gay disease like she was saying this spread from gay clubs

Here, the temporal indicators are not explicit in that Josh does not directly refer back to an
earlier time period in the way that the teachers do. Rather, the temporal reference is implicit
and functions as another instantiation of the kind of spectral haunting discussed by Leap.
The reference to AIDS as a gay disease echoes the dominant AIDS discourses of the 1980s
– but here we see in Josh’s narrative that this discourse is still being reiterated in classrooms
in the late 2010s. Ashford also reports on a teacher relaying to a class that openly discuss-
ing LGBT issues and same-sex marriage is ‘illegal’ in schools and churches. This erroneous
information is another example of a spectral discourse from the past being construed in the
temporal present. In a similar way to the previous example, the temporal indicators are not
explicit, but are implied through spectrality: homosexuality and same-sex marriage were
illegal in the past and, even though this is no longer the case, the discourse from the past is
reiterated by the teacher in the present to produce a homophobic discourse in the classroom.

Ashford: I asked the teacher and she said we can’t do anything like LGBT and marriage
because it’s illegal to do it in the church and school

346
 Applying queer theory

For these young people, there are no available reporting mechanisms for spectral (past)
homophobic discourses being construed in the present time in schools. Thus, temporal spec-
trality (echoes of past homophobic discourse) becomes a legitimate means of perpetuating
homophobic discourse in school spaces.

Spatiality
The examples discussed in the previous section have shown how temporality emerges as
a QAL concept in the interview data and functions as a means of showing how some of
the young people and teachers understand how homophobia is produced in schools in the
present day. The second QAL concept of spatiality also emerges as salient in the interview
data and often overlaps with temporality and normativity. The young people interviewed for
the research often talk about school as a space (or a number of spaces) and frequently report
their school experiences as conflicting with their experiences of other spaces, such as the
home, social media, and the internet in relation to their sexuality. In this sense, the GSTPs
invoke spatiality as well as gender and/or sexuality. They often perceive different gender
and sexual norms as operating within the space of the school, meaning that ‘normativity’
itself becomes imbued with specific meanings which are contingent on school space. The
young people participants, in particular, also often report different norms operating in differ-
ent school spaces, such as specific subject classrooms, toilets, and changing rooms. Whilst
classrooms are often perceived to be ‘safer’ due to the presence of teachers, schools spaces
which are not policed by adults are reported by the young people to be experienced as unset-
tling and threatening and as spaces where non-normative sexualities and genders are likely
to be punished through physical and verbal violence. Thus, normativity itself shifts in rela-
tion to time and space. Several young people report on the spatial segregation of students
according to perceived binary sex in their schools, as in the example below.

Ashford: another thing that schools need to change is segregating or like putting people
in different things because of their gender
[…] girls’ changing room right next to it they’re on completely different sides of the
school and even our PE lessons are quite often segregated which is ridiculous

Some of the young people also constructed their school as an ‘unsafe’ space, lamenting the
lack of spaces within the school which offered safety and freedom from the strict gender and
sexuality policing perceived as characterising the school environment.

Ashley: I think that it would be good to implement discussion of sexuality and gender
identity in any kind of conversation about sexual health
I think that in terms of policy there should really have been some safer spaces
I feel like the solution they offered was so absurd that I should change with the female
staff that would have never worked

In the interviews, the students suggest that safe school spaces could, for example, take
the form of ungendered toilets and changing rooms and classrooms in which any form
of discriminatory language and behaviour is consistently and explicitly challenged. In
sum, the young people talked about their school-based experiences of gender and sexual
diversity as largely heteronormative and homophobic, and these discourses were located

347
Helen Sauntson

temporally and spatially through linguistic signifiers such as GSTPs. Moreover, these sig-
nifiers of temporality and spatiality overlap with the third key concept within QAL – that
of normativity.

Normativity
In the interview data, the young people identified SRE lessons as a key school-based site
for the reinforcement of heteronormativity and, therefore, as a temporal space in which they
repeatedly felt isolated and marginalised. This is revealed through the QAL-informed CDA
presented in the preceding sections. As explained earlier, I also recorded a sequence of four
(x 1-hour) SRE lessons as part of the research project. The lessons focused on the topics of
the risks associated with embarking on a sexual relationship, and making decisions about
relationships. In applying CDA to this dataset, I again used the method of critically analys-
ing GSTPs in relation to normativity.
A key finding which emerged was that ‘gender’ itself was frequently constructed in
binary terms and transgender issues are completely ignored, thus constructing a norm of
gender as biological and static. For example, there were frequent references to ‘penis’ (bio-
logical male) in relation to ‘boys’ (gender). The result is the emergence of normative con-
structions of gender as reducible to biological sex, and as a binary construct. Furthermore,
talk about biological body parts in relation to sex almost always occurred in relation to the
negative consequences of engaging in sexual activity, such as the transmission of sexually
transmitted infections (STIs), as illustrated in the following examples (in the data examples,
GSTPs which construct normativity are underlined):

1 T: this is thrush this is what a discharge would look

2 like boys the one below is what a thrush discharge would
3 look like for a girl

1 T: it’s genital warts and you can see warts on a boy’s

2 penis and warts on a girl’s vulval area as well

1 T: if you don’t get it treated then it can cause

2 infertility when you’re older which means it might stop
3 you from conceiving a baby in the natural way that a baby
4 is conceived

Another key finding is that particular and restricted normative discourses of heterosexual-
ity are (re)produced through the interaction as well as through the content of the lesson.
Monogamous heterosexual relationships are ideologically afforded a normative status and
sexual activity which takes places within such relationships is prioritised. Other possible
relationship and sexual activity options are notably absent from the discourse. This sup-
ports Motschenbacher’s (2010, 2011) argument that heteronormativity is ‘ubiquitous’ and
continually thriving in everyday talk. But there is also no diversity represented within het-
erosexuality – it is almost always constructed as two-person, monogamous, and involving
no physical sexual activity other than vaginal intercourse. Furthermore, sex is constructed
as risky and dangerous, even the vaginal intercourse that is presented as the primary activity
taking place within such heterosexual relationships. Sex is constructed in negative terms,

348
 Applying queer theory

and there is a predominant focus on ‘unwanted’ outcomes, such as pregnancy and the trans-
mission of STIs. In the example below, this coincides with the idea that sexual activity
amongst teenagers is expected to remain ‘hidden’ from parents and caregivers and that
engaging in sex is something that young people would not normally be expected to discuss
with their families. It is implied in this example that the hypothetical ‘mum and dad’ would
not have been involved in the young person’s decision to start engaging in sexual activ-
ity and to take measures to ensure protection. The notion of secrecy and lack of dialogue
amongst families is a theme which runs throughout the whole sequence of lessons. The
possibility of sex being a topic which is openly discussed in a positive way in the home is
absent from the discourse.

1 T: understand the impact that your decisions have on your

2 life and remember that there are other people involved in
3 that life and it can be for example pregnancy it can be
4 sexually transmitted diseases it can be something as
5 simple as your mum or dad have found condoms in your
6 bedroom or the pill

In examples such as these in which the teacher is discussing the negative consequences of
engaging in heterosexual activity, there are more serious consequences for girls than for
boys. The negative consequences include infertility, pregnancy, STIs (there are more exam-
ples discussed of girls having STIs than boys), being labelled, and having regrets. Thus,
‘regret’ is constructed as a norm for girls in this particular spatio-temporal context.
The actual SRE guidance document for England and Wales (2014) (which informs the
content of SRE in state schools) states that there should be ‘no promotion of sexuality’
despite the implicit focus on heterosexuality as exemplified by its concurrent emphasis on
reproduction, contraception, and ‘importance of marriage for bringing up children’.

[…] children should be taught about the nature of marriage and its importance for fam-
ily life and for bringing up children. The Government recognises that there are strong
and mutually supportive relationships outside marriage. Therefore, children should
learn the significance of marriage and stable relationships as key building blocks of
community and society […] teachers should be able to deal honestly and sensitively
with sexual orientation, answer appropriate questions and offer support. There should
be no direct promotion of sexual orientation. Sexual orientation and what is taught in
schools is an area of concern for some parents.
(SRE Guidance for England and Wales 2014: 11–13)

This tension between the edict to not promote sexual orientation and the routine ‘promo-
tion’ of heterosexuality is realised through the ideological assumptions made about nor-
mative gender and (hetero)sexuality in the classroom interaction data. In the two typical
examples below, the GSTPs invoke heterosexuality, and explicit reference is made by the
teacher to ‘mum and dad’, thus inferring that a heterosexual two-parent family structure is
the expected norm:

1 T: diseases it can be something as simple as your mum or

2 dad have found condoms in your bedroom or the pill

349
Helen Sauntson

1 T: and you’ve got your mum and dad there and you’re like
2 [singing noise] or your gran is even worse isn’t it

In other examples, the teacher makes reference to ‘the guy’ and ‘the girl’ when discussing
‘relationships’ in a general sense, thus reinforcing heterosexuality as the expected norm:

1 T: when we watched that ‘A to Z of Love and Sex’ there

2 was a guy on there that talked about his first intimate
3 relationship was with a girl it was her first time
4 T: just glide it out don’t just pull your penis out
5 because what happens is the condom will stay inside the
6 girl

In sum, the norms operating in the school environments of these young people were over-
whelmingly perceived to be discriminatory and intolerant in relation to gender and sexual
diversity.
The examples discussed above are intended to give a flavour of how a QAL approach,
drawing on the established linguistic analytical method of CDA in its application, can be
applied to the study of language, gender, and sexuality in school contexts. The examples
illustrate the centrality of the QAL concepts of temporality, spatiality, and normativity
within a queer approach and, in doing so, draw attention to real-world problems and chal-
lenges currently being faced by young people. What the QAL-informed research reported
on in this chapter has shown, for example, is that homophobia (and other gender and sexual-
ity-based discriminatory practices) in UK schools now largely operates at a discursive level
and is therefore very difficult to challenge. Moreover, it is evident that research on language,
sexuality, and education must move beyond a focus on homophobia towards broader queer
linguistic issues of gender and sexuality diversity. The context of the research suggests that
sexual equality (‘triumph’) is becoming an accepted linear ‘triumph narrative’ (Leap 2020)
which actually functions to occlude and obscure ongoing discriminatory practices in rela-
tion to gender and sexuality which continue to be discursively achieved.

Future directions
Future contributions to the development of queer theory and its application to language,
gender, and sexuality are also likely to involve the development of methods and analytical
frameworks. Milani (2013), for example, examines multimodal (linguistic and spatial) con-
structions of sexual identity in a university environment. He argues that using multimodal
methods of discourse analysis, and incorporating a semiotic analysis of sexuality and space,
could further our understanding of the relationship between language and sexuality and
how it operates in educational and other contexts. Queer linguistics has been applied to the
analysis of language and sexuality in virtual spaces (e.g. Hiramoto 2015; King 2011) and it
is likely that this will continue to be an important area of development.
The datasets and analytical frameworks used for illustrative purposes throughout this
chapter are by no means the only ones that may be used within a QAL approach. Indeed,
a key strength of QAL and queer linguistics more generally is its flexibility and adapt-
ability to context and purpose. A QAL approach itself is not intended to be presented as
a ‘finished product’. In fact, Milani (2014) notes that ‘uncertainty’ is what crops up in
many queer theoretical writings, and this includes research informed by queer linguistics.

350
 Applying queer theory

The queer dimension of the approach itself means that it should not be stable, but should
instead be constantly evolving and always open to contestation and critical development.
As an approach, QAL certainly needs more rigorous theorisation of spatiality, tempo-
rality, and normativity, as well as what Pennycook (2017) has recently termed ‘a queer
approach to materiality’ whereby greater attention is paid to discursive materialisations
of sexualities in relation to political economies. The constructs of spatiality, temporality,
and normativity may also come to be usefully applied to other areas of critical linguistic
theory, although, at the time of writing, their relevance is only being addressed in relation
to queer linguistics.
Furthermore, future projects that apply queer theory (and in particular QAL) to language,
gender, and sexuality will take place in different times and spaces and may address differ-
ent conceptualisations, understandings, and experiences of sexual and gender normativity.
But what they will have in common is a desire for social transformation, in which schools
become places where acceptance of diversity is the new norm, and a move towards greater
social justice.

Notes
1 Some of the material in this chapter is drawn from Sauntson (2016) and (2018).
2 See Sauntson (2019 – Chapter 3) for further discussion of how each of these methods has been
applied in empirical studies of language, gender, and sexuality.
3 ‘LGBT+’ (lesbian, gay, bisexual, and transgender +) is a shorthand term for a range of non-heter-
osexual and non–gender-conforming sexual and gender identities. The ‘plus’ sign (+) on the end
of ‘LGBT’ is an acknowledgement of the diversity of gender and sexuality identities, whilst at the
same time realising that it is not feasible to iterate or capture all of them when discussing gender
and sexuality issues. The indeterminacy of the ‘+’ is also an attempt to go some way towards
recognising that gender and sexual identities are fluid and difficult to define. For more informa-
tion, see www.stonewall.org.uk/help-advice/glossary-terms and www.stop-homophobia.com/lgbt-t
erms-and-definitions.
4 Section 28 of the Local Government Act 1988 made it illegal for homosexuality to be ‘promoted’ in
schools. Non-heterosexual relationships were described as ‘pretended family relationships’. It was
repealed by Britain’s Labour government in 2003.
5 Data extracts are transcribed for content only. Conversational features are not included as they did not
form part of the analysis.
6 Times Educational Supplement – a British newspaper.

Further reading
Baker, P. (2008) Sexed Texts: Language, Gender and Sexuality. London: Equinox.
This textbook explores the role of language in the construction of gender and sexuality identities
in a range of text types and contexts.
Cameron, D. and Kulick, D. (eds.) (2006) The Language and Sexuality Reader. London: Routledge.
This reader brings together work from a range of fields and disciplines which examines various
aspects of the relationship between language and sexuality.
Cashman, H. (2018) Queer, Latinx and Bilingual: A Critical Sociolinguistic Ethnography. London:
Routledge.
This book presents findings from an ethnographic study of LGBTQ Mexicans/Latinxs which
explores how participants’ ethnic and sexual identities are understood and enacted in relation to their
language practices.
Morrish, E. and Sauntson, H. (2007) New Perspectives on Language and Sexual Identity. Basingstoke:
Palgrave.

351
Helen Sauntson

This book, informed by queer theory and queer linguistics, presents a range of linguistic data to
explore aspects of the relationship between language and sexual identity with a specific focus on non-
heteronormative identities.
Nelson, C. (2009) Sexual Identities in English Language Education: Classroom Conversations.
London: Routledge.
This monograph draws on queer theory to explore the challenges and opportunities which arise in
addressing queer themes and perspectives in English language classrooms.

Related topics
Language, gender, and sexuality: reflections on the field’s ongoing critical engagement with the
sociopolitical landscape; gender and sexuality normativities; interactional sociolinguistics in language
and sexuality research: benefits and challenges; determining the impact of gender stereotyping on
patient feedback; investigating gendered language through collocation.

References
Bradlow, J., Bartram, F., Guasp, A. and Jadva, V. (2017) School Report: The Experiences of Lesbian,
Gay, Bi and Trans Young People in Britain’s Schools in 2017. London: Stonewall.
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Butler, J. (1993) Bodies That Matter: On the Discursive Limits of ‘Sex’. New York, NY: Routledge.
Butler, J. (1998) ‘Afterword’. In: Munt, S. (ed.) Butch/Femme: Inside Lesbian Gender. London:
Cassell, pp. 225–230.
Butler, J. (2004) Undoing Gender. New York, NY: Routledge.
Cameron, D. (2005) ‘Language, gender and sexuality: Current issues and new directions’, Applied
Linguistics, 26(4), pp. 482–502.
Cameron, D. and Kulick, D. (2003) Language and Sexuality. Cambridge: Cambridge University Press.
Department for Education (2014) Sex and Relationships Education Guidance. Crown Copyright. DfE..
Halberstam, J. (2005) In a Queer Time and Place. New York, NY: New York University Press.
Hall, C., Smith, P., and Wicaksono, R. (2017) Mapping Applied Linguistics: A Guide for Students and
Practitioners, 2nd edn. London: Routledge.
Hall, K. (2013) ‘Commentary I: “It’s a hijra!”: Queer linguistics revisited’, Discourse and Society,
24(5), pp. 634–642.
Halperin, D. (1993) ‘Is there a history of sexuality?’ In: Abelove, H., Barale, M., and Halperin, D.
(eds.) The Lesbian and Gay Studies Reader. London: Routledge, pp. 416–421.
Hiramoto, M. (2015) ‘Who’s really normal? Language and sexuality in public space’, Journal of
Language and Sexuality, 4(2), pp. 183–192.
King, B. (2011) ‘Language, sexuality and place: The view from cyberspace’, Gender and Language,
5(1), pp. 1–30.
Kosciw, J., Greytak, E., Giga, N., Villenas, C., and Danischewski, D. (2015) The 2015 National School
Climate Survey. New York, NY: GLSEN.
Leap, W. (2020) Language and Sexuality before Stonewall. Basingstoke: Palgrave.
Leap, W. and Motschenbacher, H. (2012) ‘Launching a new phase in language and sexuality studies’,
Journal of Language and Sexuality, 1(1), pp. 1–14.
Livia, A. and Hall, K. (eds.) (1997) Queerly Phrased: Language, Gender, and Sexuality. Oxford:
Oxford University Press.
McDermott, E., Roen, K., and Scourfield, J. (2008) ‘Avoiding shame: Young LGBT people,
homophobia and self-destructive behaviours’, Culture, Health and Sexuality, 10(8), pp. 815–829.
Milani, T. (2013) ‘Expanding the queer linguistic scene: Multimodality, space and sexuality at a South
African university’, Journal of Language and Sexuality, 2(2), pp. 206–234.

352
 Applying queer theory

Milani, T. (2014) ‘Sexed signs – Queering the scenery’, International Journal of the Sociology of
Language, 228(228), pp. 201–225.
Motschenbacher, H. (2010) Language, Gender and Sexual Identity: Poststructuralist Perspectives.
Amsterdam: John Benjamins.
Motschenbacher, H. (2011) ‘Taking queer linguistics further: Sociolinguistics and critical
heteronormativity research’, International Journal of the Sociology of Language, 212, pp. 149–179.
Motschenbacher, H. (2014) ‘Focusing on normativity in language and sexuality studies: Insights from
conversations on objectophilia’, Critical Discourse Studies, 11(1), pp. 47–70.
Motschenbacher, H. and Stegu, M. (2013) ‘Queer linguistic approaches to discourse: Introduction’,
Discourse and Society, 24(5), pp. 519–535.
Nelson, C. (2012) ‘Emerging queer epistemologies in studies of ‘gay’-student discourses’, Journal of
Language and Sexuality, 1(1), pp. 79–105.
Pakula, L., Pawelczyk, J., and Sunderland, J. (2015) Gender and Sexuality in English Language
Education: Focus on Poland. London: British Council.
Pennycook, A. (2017) ‘Discussion: Sexuality and applied linguistics: poststructuralist perspectives’,
Colloquium at American Association of Applied Linguistics Conference, Portland, Oregon, USA,
March 2017.
Sauntson, H. (2012) Approaches to Gender and Spoken Classroom Discourse. Basingstoke: Palgrave.
Sauntson, H. (2016) ‘Language, sexuality and education’. In: Wortham, S. and Kim, D. (eds.)
Encyclopedia of Language and Education: Discourse and Education. New York, NY: Springer.
Sauntson, H. (2018) Language, Sexuality and Education. Cambridge: Cambridge University Press.
Sauntson, H. (2019) Researching Language, Gender and Sexuality: A Student Guide. London:
Routledge.
Sunderland, J., Cowley, M., Abdul Rahim, F., Leontzakou, C., and Shattuck, J. (2002) ‘From
representation towards discursive practices: Gender in the foreign language textbook revisited’.
In: Litosseliti, L. and Sunderland, J. (eds.) Gender Identity and Discourse Analysis. Amsterdam:
John Benjamins, pp. 223–255.

353
 23
Text trajectories and gendered
inequalities in institutions
Susan Ehrlich and Tanya Romaniuk

Introduction
The relationship between language and gender has been investigated by discourse analysts
from a variety of traditions, many of which Bucholtz (2003) surveyed in the first edition of
the Blackwell Handbook of Language and Gender (Holmes and Meyerhoff 2003). Included
among the approaches Bucholtz described in this chapter is one she labelled, following
Silverstein and Urban (1996), a ‘natural histories of discourse’ approach. This approach,
rather than taking as its starting point the pre-existence of ‘bounded’ texts that are then sub-
ject to analysis, focuses its analysis on the very formation of texts as ‘autonomous objects’
as well as on their movement across contexts (Bucholtz 2003: 61). In 2003, when Bucholtz’s
chapter was published, she argued that this approach had ‘not yet been fully tapped for its
potential as a model for language and gender research’ (Bucholtz 2003: 61); 15 years later,
in 2018, we would suggest that it has remained largely ‘untapped’ for these purposes. That
is, while the investigation of textual formations and movements, what Maybin (2017: 417)
calls the ‘dynamic turn’ in discourse analysis, has informed recent studies on various types
of institutional discourse (see, for example, Lillis and Maybin 2017), very few of these
studies have involved the analysis of gender. Thus, in this chapter we attempt to show how
investigations of ‘movements … of spoken, written and multimodal texts … across social
space and time’ (Maybin 2017: 416)what we are calling ‘textual trajectories’ (following
Blommaert 2005) – can shed light on gendered inequalities in institutional settings in ways
that investigations of single instances of bounded texts cannot. We do this by presenting
two case studies, one involving the legal system and one involving the mainstream media.

A ‘textual trajectory’ approach

A ‘textual trajectory’ approach to discourse, by virtue of its concern with language, culture,
and society and its roots in linguistic anthropology and discourse analysis, is part of the
broad, interdisciplinary field that Bucholtz and Hall (2005: 586) have termed ‘sociocultural
linguistics’. Building on Bakhtin’s (1981, 1986) insight that all utterances, even monologic
ones, are dialogical in nature (i.e. responsive to previous utterances and anticipatory of future

354
 Text trajectories and inequality

utterances), work on textual trajectories has, as Maybin (2017: 419) says, ‘decenter[ed]
from specific interactions and bounded texts in order to address the dialogic chains stretch-
ing before and after’ (emphasis ours). As an example, consider a police interview with a
suspect: when a suspect is interviewed by a police officer that interview forms a ‘text’;
however, the chain of texts does not stop there; that interview is subsequently turned into
a written statement, a statement that can then be referred to, quoted, or used as evidence
in a trial by lawyers or judges. Likewise, a ‘text trajectories’ analysis would not stop with
an examination of the original police interview; it would investigate this interview along
with the chain of texts that ‘stretch … after’ it – for example, its written version produced
by the police, and the way that this written version might be discussed or cited or quoted
by legal professionals in a court of law. Indeed, Blommaert (2005: 67) argues that focus-
ing on a chain of interconnected texts – as opposed to ‘the unique, one-time’ instance of
a given text – provides insight into the institutionally consequential meanings that can be
ascribed to a text once it is removed from its original site of production and interpreted by
those not involved in the original production. This is important because, as Bauman and
Briggs (1990: 73–75) argue, ‘entextualization practices’ can have transformative effects:
once a stretch of talk is ‘lifted out of its interactional setting’, segmented, and turned into
a ‘text’ (what Bauman and Briggs call ‘entextualization’), it may bring something from its
earlier context, but may also take on different meanings as it is ‘recentred’ (i.e. recontextu-
alised) in a new context. Moreover, to the extent that participants in institutions may have
unequal access to and/or control over contextualising spaces (as is true in many institutional
settings), these transformations in meaning can be deeply implicated in larger patterns of
social inequality, including gendered inequalities. As Bauman and Briggs (1990: 76) say, ‘to
decontextualize and recontextualize a text is … an act of control’ given that participants in
institutions may have differential access to the creation and circulation of texts, differential
legitimacy in claims to texts, and differential competence in their use.
The focus on power, control, and inequality that characterises much work on textual tra-
jectories resonates with ‘critical’ approaches to discourse analysis and the assumption that
language (e.g. text creation, movement, and transformation) is implicated in the workings
of social power. For a critical discourse analyst such as Fairclough (2001: 230), the way that
‘language figures within social relations of power’ or ‘works ideologically’ is often opaque,
and thus it is only through analysis that the demystifying and denaturalising of these opaque
aspects of language is made possible. This is at least one of the ways in which critical discourse
analysis is ‘critical’ – it promotes an awareness of the naturalised dimensions of discourse (i.e.
those aspects of discourse that have come to seem commonsensical and inevitable), with the
view that such awareness may, in turn, help to dismantle the unequal social practices revealed
by analysis (Chouliaraki and Fairclough 1999). In a comparable way, the goal of much work
on textual trajectories is to expose and make visible how institutions can ‘shore up institutional
power and disempower individuals’ (Maybin 2017: 424) through the recontextualisation of
texts and their concomitant transformations in meaning. As Park and Bucholtz (2009: 498)
maintain, if ‘politically engaged scholarship’ is to address ‘the social inequalities deriving
from the hegemony of institutions’, then it needs to conduct a ‘close analysis’ of processes of
entextualisation and recontextualisation (i.e. text trajectories in institutions).
In what follows, we present two close analyses of textual trajectories, one involving
the legal system and one involving the mainstream media. In both cases, we attempt to
demonstrate how texts can be invested with gendered and/or sexist meanings as they are
transplanted into new contexts, reinforcing and perpetuating gendered inequalities. As
noted above, one of the distinguishing features of our approach is that we do not focus on

355
Susan Ehrlich & Tanya Romaniuk

‘one-time’ instances of texts but rather on a linked chain or linked series of texts. Bucholtz
(2003) characterises the ‘method’ we use in the following way:

If some approaches to discourse analysis emphasize oral discourse, and others focus on
written texts, then natural histories of discourse call attention to the interplay between
the oral and the written and between earlier and later versions of the “same” oral or
written discourse.
(Bucholtz 2003: 61)

Indeed, it is by examining the ‘interplay’ between ‘earlier and later versions of the ‘“same”
… discourse’ that our approach is able to reveal some of the ideological effects of textual
movement and, specifically, how texts can take on gendered and/or sexist meanings as they
move into new settings and contexts.

The legal system: strategic submission vs. consent in a sexual assault case:
Her Majesty the Queen v. Ewanchuk1
A salient feature of communication in the legal system involves the movement of texts across
contexts, for example, when audio or video recordings from previous depositions/interviews
are played in trials or when trial testimony is represented in the closing arguments of lawyers
or in the decisions of judges. In this section, we analyse a sexual assault legal case in which
a complainant’s account of sexual violence was transformed as it moved from the trial to
some of the case’s judicial decisions. In particular, as the complainant’s trial testimony was
recontextualised in the decisions of the trial judge and one of the appellate courts, what the
complainant characterised as a series of strategies to avoid more extreme violence (from the
accused) was interpreted by the judges as consensual sex. We argue that this meaning trans-
formation was grounded in damaging ideological assumptions about women’s sexuality –
specifically, the idea that women’s sexual passivity and silence can be construed as consent.
The data analysed in this section come from the Canadian criminal trial, Her Majesty
the Queen v. Ewanchuk. The accused in this case was charged with sexual assault and was
acquitted by the trial judge. Upon appeal, this acquittal was upheld by the Alberta Court of
Appeal (a provincial court) but upon further appeal to the Supreme Court of Canada, the
acquittal was overturned and a conviction was entered for the accused. The sexual assault
took place during a job interview between the accused and the complainant in the accused’s
van. According to the complainant’s testimony, the interview itself was conducted in a polite,
business-like fashion. Nonetheless, after the interview, the accused took the complainant to
the trailer behind his van, closed and locked the door of the relatively small enclosure and
initiated a number of acts of sexual touching that became increasingly more aggressive.
The complainant testified that she complied with many of the accused’s requests/demands
out of fear that resistance would prompt the accused to become more violent. When the
sexual touching progressed to the complainant’s breasts and her inner thigh and pelvic area,
however, she said ‘no’ twice. In spite of these expressions of non-consent, the accused per-
sisted. His aggression culminated with his grinding his pelvis into the complainant’s pelvis,
touching her vaginal area and placing his penis on the complainant’s pelvic area under her
shorts, at which point the complainant said ‘no’ for a third time. The accused stopped after
this third ‘no’ (after two and a half hours in the van with the complainant), opened the door
of the van at the complainant’s request and the complainant left the van. She later charged
the accused with sexual assault.

356
 Text trajectories and inequality

We begin the analysis of the data by presenting excerpts from the initial text in this text
trajectory – the 1995 trial in the Ewanchuk case. The following three excerpts from the com-
plainant’s trial testimony reveal that the complainant was afraid of the accused and agreed
to many of his requests/demands because she feared that not doing so could lead to greater
brutality and violence.

Excerpt 12
Q: And what happened then?
A: He told me that he felt very tense and that he would like to have a massage, and he then
leaned up against me with his back towards me and told me to rub his shoulders and I
did that.
(some intervening turns)
Q: Did you want to give him a massage?
A: No.
(some intervening turns)
Q: If you didn’t want to give him a massage at that point in time, why did you touch his
shoulders?
A: I was afraid that if I put up any more of a struggle that it would only egg him on even
more, and his touching would be more forced.

Excerpt 2
Q: And what happened then?
A: Then he asked me to turn around the other way to face him, and he said he would like to
touch my feet or he would like to massage my feet, so I did. And he was just touching
my feet.
Q: Did you want him to massage your feet?
A: No.
Q: Why did you turn around?
A: Because I guess I was afraid. I was frozen. I just did what he told me to do.
(some intervening turns)
Q: What happened after you turned around?
A: He was massaging my feet, but he didn’t stay there. He was moving up my leg more toward
my inner thigh, my pelvic area, and then he’d move back again, and then he’d move back
up again, and I just sat there, and I didn’t – I didn’t do anything. I didn’t say anything. I knew
something was going to happen, and I didn’t want to fight. I didn’t want to struggle. I didn’t
want to scream, because I felt that that would just egg him on more.

Excerpt 3
Q: And what happened when he reached to hug you?
A: He just did, and I, at this time, I was trying really hard not to cry. I had been wiping my
eyes when he was on top of me when he couldn’t see me, and I just – I just responded
by just lightly putting my arm on him when he hugged me because I was afraid that he
would think I was really scared, and that I would leave there telling people.
Q: And why were you worried about him thinking that?
A: Because I didn’t think that he would stop there, that it would get worse, and it would be
more brutal.

357
Susan Ehrlich & Tanya Romaniuk

In all of these examples, the prosecuting lawyer is asking the complainant why she complied
with the accused’s requests/demands: in Excerpt 1, why she gave him a massage, in Excerpt
2, why she allowed him to massage her, and in Excerpt 3, why she reciprocated his hug. In
response to these questions, the complainant says a variety of things (italicised above): that she
‘was afraid’ and ‘frozen’, that she didn’t ‘want to fight … struggle … scream … because [she]
felt that would just egg him on more’, ‘his touching would be more forced’, and ‘he would not
stop there … it would get worse … and would be more brutal’. Such responses reflect strate-
gies that many victims of sexual violence employ to prevent more prolonged and extreme
instances of violence. As researchers on violence against women have argued, submitting to
coerced sex or physical abuse is often ‘a strategic mode of action undertaken in preservation
of self’ (Lembert 1996: 281). That is, if physical resistance on the part of victims can escalate
and intensify violence, as some research shows (e.g. Dobash and Dobash 1992) and as many
women believe (or are instructed to believe), then submission to coerced sex can be the best
strategy for survival. Indeed, the italicised portions of Excerpts 1–3 indicate that the com-
plainant in the Ewanchuk case adopted just this kind of strategy: she submitted to much of the
accused’s sexual aggression in an attempt to avoid more serious violation and injury.
As stated above, the trial judge acquitted the accused of sexual assault and the Alberta
Court of Appeal upheld his acquittal. Perhaps not surprisingly, then, these judges repre-
sented the events under investigation in ways that were often at odds with the complain-
ant’s testimony. Consider, in particular, an excerpt from another text in this text trajectory
– the 1995 trial judge’s decision at a point when the judge is recounting the ‘facts’ of the
case.

Excerpt 4

1 B [the complainant] told A [the accused] that she was an open, friendly and
2 affectionate person; and that she often liked to touch people. A told B that he
3 was an open, friendly and affectionate person; and that he often liked to touch
4 people. A and B talked. They touched each other. They hugged. They were
5 sitting on the floor of the trailer and they were lying on the floor of the trailer.
6 A told B that he would like a body massage, and B gave A a body massage.
7 For the body massage, A sat in front of B so that B could massage A's back.
8 They later exchanged places so that A could give B a body massage. B later
9 lay on her back, and A gave B a foot massage. After the foot massage, A
10 massaged B's bare legs and he massaged her bare inner thighs. During this
11 period of two and one half hours, A did three things which B did not
12 like. When A was giving B a body massage, his hands got close to B's
13 breasts. B said "No", and A immediately stopped. When B and A were lying
14 on the floor, A rubbed his pelvic area against B's pelvic area. B said "No",
15 and A immediately stopped. Later on A took his soft penis out of his shorts
16 and placed it on the outside of B's clothes in her pelvic area. B said "No", and
17 A immediately stopped. During all of the two and one half hours that A and B
18 were together, she never told A that she wanted to leave. When B finally told
19 A that she wanted to leave, she and A simply walked out of the trailer.
(From Reasons for Judgement (Moore, J., C.Q.B.A.), 10th November 1995)

In general, this description of what transpired between the complainant and the accused
departs quite dramatically from the complainant’s testimony. That is, with the exception of

358
 Text trajectories and inequality

‘the three things’ that the accused did ‘which the [complainant] did not like’ (lines 11–12),
the picture that emerges from Excerpt 4 is one of consensual sex and reciprocity as opposed
to coercion and submission. For example, in line 4 of Excerpt 4, the trial judge states that the
complainant and the accused hugged using a reciprocal construction (i.e. ‘They hugged’)
whereas in Excerpt 3, above, the complainant says that she responded to the accused’s
hug out of fear, i.e. ‘because [she] was afraid that he would think [she] was really scared’
which in turn would lead to even greater ‘brutal[ity]’. Likewise, in line 6 of Excerpt 4, the
judge states simply that the complainant gave the accused a body massage in response to
the accused’s request (i.e. ‘A told B that he would like a body massage, and B gave A a
body massage’); yet in Excerpt 1, above, the complainant says that she agreed to massage
the accused only because she ‘was afraid that if [she] put up any more of a struggle that it
would only egg him on even more and his touching would be more forced’. Finally, we see
that, in lines 9–10 of Excerpt 4, the judge states that the accused massaged the complain-
ant’s bare legs and inner thighs, as if this were a consensual act (i.e. ‘After the foot massage,
A massaged B’s bare legs and he massaged her bare inner thighs’); but in Excerpt 2 above,
the complainant says that she submitted to the accused’s massages because she didn’t want
to ‘egg him on more’. In sum, as the complainant’s words were reframed within the trial
judge’s decision, crucial contextual details were erased – the complainant conveyed in her
testimony that she had little choice but to submit to the accused’s sexual advances, yet
the coercive dimension of these acts is absent in the trial judge’s recounting of the ‘facts’.
Indeed, what the complainant represented as submission and compliance in Excerpts 1–3
was recontextualised as freely given consent in Excerpt 4.
Consistent with the 1998 Alberta Court of Appeal’s decision to uphold the trial judge’s
acquittal of the accused, we see in the following excerpt (from another text in the text tra-
jectory) that the Alberta appellate court supported the trial judge’s rendition of the ‘facts’
of the case.

Excerpt 5

Yet, if review of the evidence that supports the trial judge’s doubts about consent
in this case is called for, it may be found in the following. The advances that are
now said to be criminally assaultive were preceded by an exchange of consensual
body massages, partially on the floor of the trailer, hugs and assurances of trust
and restraint … Beyond that (and somewhat inconsistent with an appellate profile
of Ewanchuk as a relentless sexual predator) every advance he made to her stopped
when she spoke against it.
(From Reasons for Judgement of the Honourable Mr. Justice McClung,
12th February 1998)

Specifically, like the trial judge, the italicised portion of Excerpt 5 shows that the Alberta
Court of Appeal understood the hugs and massages between the complainant and the accused
to be consensual and reciprocal. But, of course, the kinds of recontextualisations that we see
in both Excerpts 4 and 5 raise questions as to why the complainant’s own representations of
the events, that is, that she submitted to the accused’s wishes out of fear and intimidation,
did not receive uptake by the trial judge and the Alberta Court of Appeal.
Other parts of these two judicial opinions are revealing in this regard. Both courts
determined that the complainant had ‘implied consent’ through what they termed her
‘conduct’. That is, while they both found the complainant’s fear of the accused to be

359
Susan Ehrlich & Tanya Romaniuk

credible, they argued that she had not communicated this fear to the accused. Rather,
according to the trial judge, ‘she successfully kept all her thoughts, emotions and specula-
tions deep within herself’ and ‘she did not communicate most of her thoughts, emotions
and speculations’ (Reasons for Judgement (Moore, J., C.Q.B.A.), 19th November 1995).
Such a description of the complainant and of the kind of ‘conduct’ that the judges viewed
as ‘implying consent’ is interesting as it does not seem to depict a woman who is initiat-
ing sexual activity nor one who is responding in any active way to the accused’s sexual
advances. In other words, it is possible that normative ideas about women’s passive and
acquiescing sexuality (Gavey 2005) were ‘in play’ when the lower courts in the Ewanchuk
case ruled that a woman who ‘successfully kept all her thoughts, emotions and specula-
tions deep within herself’ implied consent. Put somewhat differently, the trial judge and
the Alberta Court of Appeal judges seemed to view sexual passivity as appropriately femi-
nine and, as a result, what the complainant described as ‘submitting to sex out of fear’ was
represented in the judicial opinions as the complainant ‘consenting to sex’, or at the least,
implying consent to the accused.3
In the Ewanchuk case, then, as the complainant’s trial testimony was recontextual-
ised in the lower courts’ judicial decisions, it underwent a fairly radical transforma-
tion: a strategy for preventing more extreme and prolonged instances of sexual violence
was transformed into a signal of consent. From a methodological point of view, it is
important to notice that the problematic interpretation of consent seen in these courts’
representation of the trial ‘facts’ would not have been evident if the trial or the judicial
opinions had been the exclusive object of analysis. That is, it was only by looking at
the ‘interplay’ of texts in the case’s textual trajectory that this meaning transformation
became evident.
Maybin (2017: 417) has argued, following Bauman and Briggs (1990), that a textual
trajectory approach to discourse analysis can elucidate the ‘ideological workings of textual
movements and transformations’ in institutions. Indeed, we believe that our analysis of the
‘textual movements and transformations’ in the Ewanchuk case has revealed how sexist ide-
ologies surrounding gender and sexuality can come to inform women’s accounts of sexual
violence. Thus, not only does a text trajectories approach illuminate meaning transforma-
tions in ways that analyses of single, bounded texts cannot, it also helps to identify and
expose the ideological basis of such transformations. In this particular case, for instance,
we have argued, following Gavey (2005), that problematic conceptions of female sexuality
made possible an interpretation of events such that an unresponsive, emotionless woman
could be understood as consenting to sex. We now turn to a second case study that adopts a
textual trajectories approach to the analysis of gender from another institutional setting, that
of presidential politics.

Politics and the mass media: laughter vs. cackling in media portrayals of
Hillary Clinton4
Like the legal system, mediated forms of public communication are another context in
which texts travel across settings and where the meanings of ‘post-hoc recontextualiza-
tions’ (Blommaert 2005: 46) of such texts can be enormously consequential, as we have
seen above with the judges’ recontextualisations of a rape complainant’s testimony. The
mass media’s coverage of politics is a quintessential example: political figures, especially

360
 Text trajectories and inequality

during leadership contests and political campaigns, engage in many forms of mediatised
speech events – interviews with journalists, debates with political opponents, and town hall
meetings with citizens, to name a few. At the same time, these speech events can be subject
to entextualisation/recontextualisation: they can become the topic of press releases, news
reports and political commentary (i.e. they can become ‘texts’) and, in the process, can take
on meanings that depart dramatically from those produced in the original acts of communi-
cation. In this section, we consider the meanings ascribed to Hillary Clinton’s laughter in the
context of her failed bid for the Democratic nomination for President of the United States
in 2007–2008. While Clinton’s laughter was not unlike the laughter produced by her (male)
political opponents during events such as broadcast news interviews, as it moved from its
‘original occasion of production’ (De Fina and Georgakopoulou 2012: 133) to various kinds
of media reports it became known as ‘The Clinton Cackle’. In other words, it took on sexist,
even misogynistic meanings, as it was recontextualised in other kinds of settings within the
mass media.
Consider, first, how Clinton used laughter in the kinds of broadcast news interviews (i.e.
the original occasions of production) that seemed to initially trigger her laughter’s char-
acterisation as ‘The Clinton Cackle’. In late September 2007, Clinton completed what is
known as a ‘full Ginsberg’ by appearing on all five Sunday morning political talk shows on
the same day – ABC’s This Week, CBS’s Face the Nation, CNN’s Late Edition, FOX’s Fox
News Sunday, and NBC’s Meet the Press. It was around this time that her laughter became
the subject of intense scrutiny by journalists, media commentators, and pundits, and was
dubbed ‘The Clinton Cackle’ as indicated by the following headlines (one form of ‘text’ in
the analysis of mass media):

Clinton cackles; lies about Iraq vote.

( The Rush Limbaugh Show, 13th September 2007)

That Clinton Cackle.

( The Boston Globe, 30th September 2007)

Democratic frontrunner can cackle all the way to the nomination.

( The Seattle Post, 2nd October 2007)

Hillary Clinton is trying to get the last laugh over her now-infamous cackle.
( Daily News, 4th October 2007)

The cold cackle of opportunism.

( The Washington Times, 5th October 2007)

The candidate’s ‘cackle’-like laugh.

( The Toronto Star, 30th December 2007)

Excerpt 6 comes from one of these five ‘Ginsberg’ interviews. (This excerpt, and the ones
presented in what follows, are representative of more general patterns identified in the data.)
It begins following a brief segment in which the interviewer (IR), Chris Wallace, showed
a short clip of an interview with Clinton’s husband, former President Bill Clinton, during

361
Susan Ehrlich & Tanya Romaniuk

which Clinton complained about Wallace’s interview questioning and accused Wallace of
doing a ‘nice little Conservative hit job’ on him. Following the clip, Wallace launches his
first question of the interview, beginning at line 1.

Excerpt 6

2007Sep24-Fox News Sunday-Hyper-partisan

IR: Chris Wallace; IE: Hillary Rodham Clinton

01 IR: Senator taw:lk about Conservative hit jobs, right-wi:ng

02 conspiracies,=wh↑y do you: and the President have such a
03 hyper-partisan view of politics.
04 (0.3)
05 IE: khe=Hah=Hah=ha:=hah=ha:=ha:=ha=.hh £well Chris if you
06 had uh walked even a day in our shoes over the last
07 fifteen years I’m sure you’d understand.£ .hh
08 but y’know the real goal for our country right now is to
09 get beyo:nd uh partisanship. a:nd uh I’m £sure trying to
10 do my part.£ .hh because we’ve got a lot of se:rious
11 problems thet uh we’re trying tuh deal with.
12 this week I rolled out my: American health choices plan. […]

Wallace’s question asks Clinton to account for her and her husband’s ‘hyper-partisan view
of politics’. The irony of this question, of course, lies in the fact that such a description of
the Clintons comes from a journalist on the Fox News Channel, which, despite its slogan,
‘Fair and balanced’, has been frequently accused of biased reporting (see, for example,
DellaVigna and Kaplan 2007). Thus, when Clinton begins her response to the question with
laughter (line 5), she seems to be treating Wallace’s question as laughable. While going
on to accept the presupposition of Wallace’s question (i.e. that the Clintons are ‘hyper-
partisan’) to some extent in lines 5–7, in lines 8–9 Clinton resists the IR’s ‘hyper-partisan’
description by refocusing the question on the ‘real goal’ for the country, namely, moving
‘beyond partisanship’. This resistance to Wallace’s question (and its presuppositions) is at
least one indication that Clinton’s laughter here, even though produced in response to a seri-
ous question, is challenging the question’s legitimacy, signaling that it is not in fact one that
should be taken seriously.
Excerpt 6 is just one example of a practice that Clinton engaged in systematically in the
broadcast news interviews during the 2007–2008 primaries (and most likely, more gener-
ally) – laughing at the completion of or during serious interviewer questions (see Romaniuk
2016). Such laughter seemed to act as a public display of disaffiliation, expressing disagree-
ment with critical or otherwise problematic commentary produced by an interviewer. For a
politician like Clinton, then, laughing in such contexts acted as a form of ‘damage control’;
that is, the laughter treated the preceding questions as laughable, and, in combination with
other embodied stance displays, it functioned to mitigate the potential damaging effects of
such questions on her or her campaign. A question that arises from this kind of analysis of
Clinton’s laughter is the extent to which it may be a generic interactional practice, deployed
by politicians more generally as they seek to ‘contain’ the damaging effects of journalists’

362
 Text trajectories and inequality

questions. Excerpt 7 is one indication that this is the case (see Romaniuk (2013) for further
evidence that this kind of laughter functions as a generic interactional practice). This excerpt
features David Gregory interviewing New York Democratic Senator Chuck Schumer.

Excerpt 7

2009Jan12-MSNBC-MeetThePress-1: Sarah Palin

IR: David Gregory; IE: Chuck Schumer

1 IR: Is Sarah Palin the future of the Republican Party,

2 IE: .hh hh=w(h)el(h)heh heh heh .hhuh I guess I shouldn't
3 judge and let them £f(h)ight among themselves.£
4 hn [hehhheh]
5 IR: [What do] you think though.=
6 IE: =.h[h
7 IR: [D’you think she’s qualified to be president?
8 IE: .hh Well y’know, I- I- I think the: American people
9 saw her. (0.2) and they saw:, (.) problems
10 in terms of preparation and knowledge of things,
11 but y’know. uh: four years away is a lo- three and
12 a half years away is a long time away so I'm not
13 gonna make a judgment. […]

Like Excerpt 6, the IR’s question (in line 1) is delivered as serious. That is, the IR does not
‘invite’ (Jefferson 1979) a laughing response in any hearable or visible way and, yet, it is
responded to with laughter (lines 2 and 4). As in Excerpt 6, Schumer resists answering the
question, and this resistance combined with his laughter work to characterise the question as
one that should not be taken seriously. Indeed, Excerpts 6 and 7 (in addition to many others
documented by Romaniuk 2016) serve to illustrate that the practice of laughing in response
to (or during the production of) an IR’s serious questions is not some idiosyncratic feature of
Clinton’s discursive repertoire, but rather a systematic activity that many politicians engage
in during news interviews with journalists. Nonetheless, despite the generic nature of this
interactional practice, when Clinton’s laughter became recontextualised in other kinds of
media discourse, it acquired not only a gendered meaning but a very negative one. That
is, there is much evidence to suggest that ‘cackle’ (as in ‘The Clinton Cackle’) has strong
indexical associations to ‘witches’.
Romaniuk (2016) showed that a collocational analysis of ‘cackle’ (an analysis that inves-
tigates the ways in which certain words habitually pattern together) demonstrates its nega-
tive semantic prosody and its strong association with ‘witches’. But, there is also ample
evidence of this association in the media representations themselves, some of which make
explicit the connection between Clinton, her laughter, and witches. Excerpt 8 is but one
example of this pattern; it is of particular interest for our purposes here because it inserts
video footage of Clinton laughing during a news interview (of the type excerpted in Excerpt
6) into Glenn Beck’s television commentary. In other words, it entextualises Clinton’s
laughter (i.e. draws boundaries around it and detaches it from its naturally occurring con-
text) and then recontextualises it in another setting.

363
Susan Ehrlich & Tanya Romaniuk

Excerpt 8

Glenn Beck, October 7, 2007

America’s starting to pay attention to the real issues that America faces, namely Clinton’s
laugh. I’ve never noticed it but critics have. They’ve called it, and I’m quoting ‘less of a
laugh and more of a cackle’. Some have been a little more cruel as to even compare her
laugh to the Wicked Witch of the West, which is just a little unfair ((laughingly)). Here
is the actual Wicked Witch of the Wizard of Oz. [((Video clip of the Wicked Witch of
the West is played))]. Okay, and here’s the wick- uh the junior Senator from New York.
[((Plays entextualised video clip of Clinton laughing during news interview))].

In this excerpt, Beck reproduces a description of Clinton’s laugh as a ‘cackle’ and then
refers to those who have likened it to the laugh produced by the ‘Wicked Witch of the West’.
Although claiming that this comparison is ‘just a little unfair’, Beck proceeds to invoke
the association anyway. Two video clips are then played: one featuring the laughter of the
Wicked Witch of the West and the second featuring Clinton’s laughter during a news inter-
view. (Note the strategic use of self-initiated repair when Beck introduces the video clip of
Clinton: ‘here’s the wick– uh the junior Senator from New York’.) Thus, Clinton’s laughter
is ‘framed’ as witch-like not only by Beck’s references to the Wicked Witch of the West, but
also by the witch herself, as seen and heard via a video clip. And, of course, this (re)contex-
tualisation dramatically reworks and reshapes the meaning of Clinton’s laughter relative to
its meaning in its original speech event: it is no longer understood as an interactional strat-
egy – laughing off the seriousness of a question – that politicians deploy in doing ‘damage
control’. Rather, this kind of recontextualisation describes Clinton’s persona and does so in
a particularly negative, gendered way – she is depicted as witch-like.
This witch trope is not uncommon as a characterisation of women who participate in
electoral politics (consider the late Prime Minister of England, Margaret Thatcher, for exam-
ple), and, thus, in Clinton’s case, it seems not unrelated to the fact that she was seeking the
Democratic nomination for President of the United States at the time that these kinds of rep-
resentations were circulating. As feminist scholars of politics have argued, the presidency
remains a ‘bastion of masculinity’ (Anderson 2002: 105) and any viable contender for the
presidency, who also happens to be a woman, is forced to run against the deeply entrenched
cultural image of ‘man as president’ (Carroll 2009). Clearly, the ‘cackle’ characterisation
of Clinton’s laughter, and its concomitant association with witches, is just one of the many
examples of sexism and misogyny that women political leaders encounter when they ‘run
up’ against this dominant/ideological cultural image. For our purposes here, however, what
is noteworthy about these particular representations is their genesis in source interactions
where Clinton’s laughter was not in fact a gendered practice nor a negatively evaluated one.
In demonstrating the way that the meaning of Clinton’s laughter was transformed as it
moved across media environments, this analysis, like the previous case study, makes a key
methodological point: it was only by looking at the ‘interplay’ of texts in a trajectory of texts
that this meaning transformation became evident. That is, to have looked only at the media
representations of Clinton’s laughter, with its sexist and misogynistic overtones, would
not have revealed that Clinton’s laughter was a generic interactional practice deployed by
both male and female politicians in broadcast news interviews. And, conversely, to have
looked only at the source interactions (i.e. the broadcast news interviews) would not have
revealed how negative, gendered meanings can be ascribed to practices that are not in and

364
 Text trajectories and inequality

of themselves gendered. We see here how a textual trajectory approach is indeed a ‘critical’
approach to gender and discourse analysis: it can expose the workings of power and ideol-
ogy in institutional discourse in ways that analyses of single instances of texts cannot.

Conclusion
As outlined above, work on textual trajectories has ‘decentred’ analysis from discrete, bounded
interactions in order to address the textual chains that ‘stretch before and after’ (Maybin 2017:
419), under the assumption that meaning-making is never exhausted in a text’s original occa-
sion of production. Rather, as Blommaert (2005: 76) says, as texts move along their trajecto-
ries, participants ‘not involved in the initial act of communication’ can ascribe meanings to
texts ‘often deeply different from the ones performed in the initial acts of communication’.
We have tried to show that the investigation of textual trajectories offers language and gender
scholars a method for capturing some of the gendered inequalities that textual movements
can produce in institutions – inequalities that may be harder to discern when single, one-time
instances of texts are examined. In focusing on the way in which key contextual details were
lost as texts travelled across settings in our two case studies, we have provided insight into
how new contextualisations can invest texts with new meanings. Specifically, we considered
how a strategic act of submission was reconstituted as a signal of sexual consent and how a
politician’s (generic) strategy for doing ‘damage control’ was reconfigured as characteristic of
her negative, gendered persona. Overall, then, we are arguing that the kinds of reshapings and
recastings that texts may undergo in their institutional trajectories are by no means innocent
acts – rather, they are informed by, and thus can cast light on, broader patterns of gendered
inequalities. In this way, we view a textual trajectory approach, at least in the context of insti-
tutions, to be a key method for ‘critical’ discourse analysts interested in language and gender.
As we have tried to demonstrate, the denaturalising of opaque aspects of language that reflect
and perpetuate sexist meanings requires, in many cases, an examination of chains of intercon-
nected texts – what we have called, following Blommaert (2005), textual trajectories.

Future directions
We suggest that future research in the area of language, gender, and sexuality would benefit
from the kind of approach we are advocating in this chapter given that there are a variety
of institutional domains and contexts in which a linked series of texts (i.e. a text trajectory)
is implicated in meaning-making. One potentially fruitful area of inquiry would be to con-
sider how texts are subject to transformations as a result of new and emerging technologies.
For example, a text trajectory approach could investigate gendered slogans and/or phrases
(e.g., #metoo) by analysing their occurrence across social networking sites such as Twitter,
Instagram, and Snapchat and, in particular, how the affordances of these emerging technolo-
gies may work to promote, negotiate, contest, or erase the gendered meanings evident in the
terms’ original occasions of production.

Notes
1 The analysis in this section is based on Ehrlich (2007).
2 The excerpts that come from official court transcripts or media reports have been represented in Times
New Roman. Excerpts 6 and 7 are transcriptions of interactions transcribed by the researcher, follow-
ing the conventions of conversation analysis despite appearing in Times New Roman (as opposed to
Courier).

365
Susan Ehrlich & Tanya Romaniuk

3 As noted above, the Supreme Court of Canada overturned the acquittal of the accused and, accord-
ingly, its decision represented the complainant as submitting to the accused’s aggression out of fear.
4 The analysis in this section is based on Romaniuk (2013, 2016).

Further reading
Hodges, A. (2011) The “War on Terror” Narrative: Discourse and Intertextuality in the Construction
and Contestation of Sociopolitical Reality. New York, NY: Oxford University Press.
In this book, Hodges follows the trajectory of the discursive term, ‘War on Terror’, from the Bush
White House to its circulation in mediatised forms of public communication to its use in the talk of
‘ordinary’ citizens.
Lillis, T. and Maybin J. (eds.) (2017) Text and Talk. Special Issue: The Dynamics of Textual Trajectories
in Professional and Workplace Practice, 37(4), pp. 409–551.
This co-edited special issue of the journal, Text and Talk, has two goals: it contributes to the
theoretical development of an approach to discourse analysis that investigates the projection of texts
across different spatiotemporal contexts and it illustrates this dynamic approach to discourse analysis
with empirically based studies from a number of institutional contexts: policing, social work, medical
surgery, and social housing.
Trinch, S. (2003) Latinas’ Narratives of Domestic Abuse: Discrepant Versions of Violence. Amsterdam
and Philadelphis, PA: John Benjamins.
In this book, Trinch follows the process by which US Latina women obtain protective orders (i.e.
restraining orders) that will prohibit their violent partners from making contact with them. Trinch is
interested in the omissions and distortions that the women’s oral narratives of abuse undergo as they
are reformulated as written affidavits for the courts.
Wortham, S. and Reyes, A. (2015) Discourse Analysis beyond the Speech Event. London: Routledge.
This book introduces an approach to discourse analysis that looks beyond bounded, fixed-speech
events in order to understanding how meanings emerge across speech events. The book provides
concrete methods for analysing texts as they move across spatiotemporal contexts.

Related topics
Anthropological discourse analysis and the social ordering of gender ideology; gender and sexuality
normativities; language, gender, and sexuality: reflections on the field’s ongoing critical engagement
with the sociopolitical landscape; language, gender, and the discursive production of women as leaders.

References
Anderson, K. V. (2002) ‘From spouses to candidates: Hillary Rodham Clinton, Elizabeth Dole, and the
gendered office of U.S. President’, Rhetoric & Public Affairs, 5(1), pp. 105–132.
Bakhtin, M. (1981) The Dialogic Imagination: Four Essays. Holquist, M. (ed.) Austin, TX: University
of Texas Press.
Bakhtin, M. (1986) Speech Genres and Other Late Essays. Emerson, C. and Holquist, M. (eds.)
Austin, TX: University of Texas Press.
Bauman, R. and Briggs, C. L. (1990) ‘Poetics and performance as critical perspectives on language
and social life’, Annual Review of Anthropology, 19(1), pp. 59–88.
Blommaert, J. (2005) Discourse: A Critical Introduction. Cambridge: Cambridge University Press.
Bucholtz, M. (2003) ‘Theories of discourse as theories of gender: Discourse analysis in language and
gender studies’. In: Holmes, J. and Meyerhoff, M. (eds.) The Handbook of Language and Gender.
Malden, MA: Blackwell, pp. 43–68.
Bucholtz, M. and Hall, K. (2005) ‘Identity and interaction: A sociocultural linguistic approach’,
Discourse Studies, 7(4–5), pp. 585–614.

366
 Text trajectories and inequality

Carroll, S. J. (2009) ‘Reflections on gender and Hillary Clinton’s presidential campaign: The good, the
bad, and the misogynic’, Politics & Gender, 5(1), pp. 1–20.
Chouliaraki, L. and Fairclough, N. (1999) Discourse in Late Modernity: Rethinking Critical Discourse
Analysis. Edinburgh: Edinburgh University Press.
De Fina, A. and Georgakopoulou, A. (2012) Analyzing Narrative: Discourse and Sociolinguistic
Perspectives. Cambridge: Cambridge University Press.
DellaVigna, S. and Kaplan, E. (2007) ‘The FOX News effect: Media bias and voting’, The Quarterly
Journal of Economics, 122(3), pp. 1187–1234.
Dobash, R. E. and Dobash, R. P. (1992) Women, Violence and Social Change. London: Routledge.
Ehrlich, S. (2007) ‘Legal discourse and the cultural intelligibility of gendered meanings’, Journal of
Sociolinguistics, 11(4), pp. 452–477.
Fairclough, N. (2001) ‘The discourse of new labour: Critical discourse analysis’. In: Wetherell,
M., Taylor, S., and Yates, S. (eds.) Discourse as Data: A Guide for Analysis. London: Sage, pp.
229–266.
Gavey, N. (2005) Just Sex? The Cultural Scaffolding of Rape. London: Routledge.
Holmes, J. and Meyerhoff, M. (eds.) (2003) The Handbook of Language and Gender. Malden, MA:
Blackwell Publishers.
Jefferson, G. (1979) ‘A technique for inviting laughter and its subsequent acceptance/declination’. In:
Psathas, G. (ed.) Everyday Language: Studies in Ethnomethodology. New York, NY: Irvington,
pp. 79–96.
Lembert, L. (1996) ‘Women’s strategies for survival: Developing agency in abusive relationships’,
Journal of Family Violence, 11(3), pp. 269–289.
Lillis, T. and Maybin, J. (eds.) (2017) Introduction: The dynamics of textual trajectories in professional
and workplace practice, Text & Talk Special Issue, 37(4), pp. 509–530.
Maybin, J. (2017) ‘Textual trajectories: Theoretical roots and institutional consequences’, Text & Talk,
37(4), pp. 415–436.
Park, J. and Bucholtz, M. (2009) ‘Introduction: Public transcripts: entextualization and linguistic
representation in institutional contexts’, Text and Talk, 29(5), pp. 485–502.
Romaniuk, T. (2013) ‘Interviewee laughter and disaffiliation in broadcast news interviews’. In: Glenn,
P. and Holt, E. (eds.) On Laughing: Studies of Laughter in Interaction. London: Bloomsbury Press,
pp. 201–220.
Romaniuk, T. (2016) ‘On the relevance of gender in the analysis of discourse: A case study from Hillary
Rodham Clinton’s Presidential bid in 2007–2008’, Discourse & Society, 27(5), pp. 533–553.
Silverstein, M. and Urban, G. (eds.) (1996) Natural Histories of Discourse. Chicago, IL: Chicago
University Press.

367
 24
‘I thought you didn’t accept
gay marriage Fr’
Combining corpus linguistics and critical
discourse analysis to investigate the
representation of gay marriage and the Irish
Mammy stereotype in Mrs Brown’s Boys

Bróna Murphy and María Palma-Fahey

Introduction
Our work comes at a time when researchers such as Bednarek (2010) and others (Baker
2005; Gregori-Signes 2017; Westman 2007) have begun to explore the impact of television
scripts in society and on the general public. As Wood (2007: 257) asserts, media products
not only transmit certain values but also have the potential to contest values regarding
social issues. One such example is Westman (2007) who looks at the production of values
and the representation of geek women and men in Gilmore Girls and the manner in which
societal gender expectations are contested in this show. Westman (2007: 22–24) points out
that the programme offers a different gendered representation of the term ‘geek’ which
is not only portrayed in terms of male characters but also female. This in turn offers up
new perceptions of women in technology which differ from traditional associations. Geek
women, for example, are not portrayed as masculine and antisocial but as beautiful and
brilliant, prompting Westman (2007: 26) to propose a redefinition of the concept which
reflects a more positive portrayal of geek women in both textual, virtual, and real contexts.
This example hints at the power of scripted fictional media to generate and disseminate
cultural values (Fairclough 2003) while also supporting or contesting ideologies that in
turn promote changes to views and stereotypes around gender (Bednarek 2010; Quaglio
2009). Baker’s (2005) study on Will and Grace, in turn, looks at representations of gay
sexuality and explores the manner in which sexuality and desire are portrayed. He asserts
that the sitcom emphasises the representation of a type of discourse that he calls ‘be your-
self’ (2005: 101) and argues that the focus on gay identities relies on characters’ behaviour
which serves to highlight the need for gay characters to come out in order to fully be
themselves. Baker asserts that the representation of homophobia is ‘complex and context
dependent’ in such a series where homophobic terms, such as ‘fag’, are often used not only

368
 The representation of gay marriage

to enhance humour, (2005: 125) but also to defuse it. He also acknowledges the writer’s
role in exploring uprooted views around gay identity and its underlying aim to educate and
enlighten its viewers, a point which we will also consider in our analysis. This more recent
interest in representing a more inclusive view of gender and sexuality is also reflected in
shows such as Queer as Folk (Channel 4, UK 2000–2005) and The L-Word (Showtime,
2004–2009) which all include new linguistic and behaviour-related roles which reflect
insights into concepts of gender and sexuality. Bednarek (2010) also alludes to the inclu-
sion of gay or bisexual characters in some episodes of popular, established contemporary
series such as Grey’s Anatomy (ABC 2005–) and House (FOX 2004–2012) amongst others.
This chapter explores the manner in which the concept of gay marriage is portrayed and
contested against an Irish sociocultural and political backdrop in the series Mrs Brown’s
Boys. It focuses, in particular, on the interplay between gay marriage and the Irish Mammy
stereotype. In doing so, it seeks to illustrate the power of combining corpus linguistics (CL)
and critical discourse analysis (CDA) to unveil insights into changing sociocultural values
around gender and sexuality in scripted fictional media. Although our focus is on gender
and sexuality, it is worth mentioning other portrayals which promote inclusivity and contest
stereotypes such as the character of the young male surgeon in The Good Doctor (ABC
2017–) who is portrayed as having autism and savant syndrome. To some extent, these
portrayals remind us of Hartley’s (1999) view of television as ‘the transmodern teacher’,
a term that is used when television tackles issues in a way which serves to enlighten its
audience and widen the breadth of understanding around specific sociocultural issues. As
a result, such shows could be viewed as semiotic sites which facilitate the negotiation and
construction of meanings and identities around gender and sexuality (Gledhill and Ball
2013: 351), a point which we will return to below.

Context, data, and methodology

Context
Our corpus is a 71,655-word dataset of televised semi-scripted fictional media taken from
the popular television series Mrs Brown’s Boys between 2012 and 2014. We also draw on a
YouTube clip featuring Mrs Brown speaking on gay marriage which went viral in 2015. The
series was created by Irish writer and performer Brendan O’Carroll, who plays the leading
role of the Irish matriarch (‘the Mammy’ also known as ‘Mrs (Agnes) Brown’). O’Carroll
has long been associated with the character who came to life in the 1990s featuring in books
such as The Mammy (O’Carroll 1994) and The Chisellers (O’Carroll 1995). She has also
been portrayed by O’Carroll in stage plays, on radio, and in 2014, in film. The show centres
around the character Mrs Brown who is a loud, vivacious, and foul-mouthed, widowed Irish
matriarch who cares for her grown-up family following the death of her husband, and who
takes an active interest in their lives. Mrs Brown is played by O’Carroll, a man, who pre-
sents himself dressed as a middle-aged traditional Irish housewife and who draws on traits
associated with the stereotype (discussed below). This portrayal is not only reflective of
Mrs Brown’s early incarnation in stage plays but also follows historical traditions in theatre
where it was common for men and boys to play female roles, a trend which dates from the
Ancient Greeks to Shakespeare and into the seventeenth Century. Theatre has always been
a safe space to view fluid gender identities where gender roles and sexual identity, as illus-
trated below, are questioned, mocked, or unsettled and where humour is often generated
from this fluidity (Ward 2016).

369
Bróna Murphy & María Palma-Fahey

Mrs Brown’s leading role and importance are reflected in the title of the show and also
emphasised by the fact that she begins and ends the show with a solo personal monologue
and concluding address. The show features her adult sons Mark, Rory, Dermot, and Trevor,
and daughter, Cathy, as well as her best friend, Winnie, and other minor characters. It is
worth highlighting that some members of the cast are close family members in real life. For
instance, Cathy is O’Carroll’s (Mrs Brown) wife and Winnie is his sister. The show which
is pre-recorded at BBC Glasgow on stage, in front of a live audience, adopts an informal
production style where mistakes and tomfoolery are left unedited, ad-libbing is common,
and where much of the humour is provided by and centres around Mrs Brown. In 2011,
O’Carroll was quoted as saying, ‘I love the lines that are not in the script’, which reflects
his attitude and approach to the show. He added that as well as being the writer and playing
the lead role, the producer allows him to have an opinion on the overall look and feel of the
show. He also emphasised the importance of the television series being filmed in front of a
live audience as it allows them to keep it as close to a stage show as possible and stated that
both himself and the actors rely on the live audience for the rhythm of their performances.
The close relationship between the cast members, the ad-libbing, live recording, and the
fact that O’Carroll has mentioned that while the Irish matriarch is not based on his mother,
she is informed somewhat by his mother and other Irish mothers’ traits, all add layers of
contextual background to the data (Beacom 2013). However, in doing so, they also make it
challenging to categorise Mrs Brown’s Boys under a specific media genre. While it shares
many commonalities with the genre of sitcoms including a continuing cast of characters and
sequence of episodes, it also presents features related to the genre of soap opera as it draws
on the audience’s shared knowledge of the stereotype, and seeks to engage the audience,
through humour, by including direct messages and subtle, nuanced meanings which convey
an appreciation of the main traditional traits associated to the stereotype. To some extent,
it also adopts some of the traits of a stage play in that it is still performed in front of a live
audience and includes audience laughter, scenes, and shots of the audience. This fuzziness in
terms of its categorisation makes it an interesting context to explore as it breaks the mould by
positioning itself as a new and emergent genre of its own (Murphy and Palma-Fahey 2018).
One of the main reasons for exploring gay marriage in Mrs Brown’s Boys is its role in
actively supporting gay rights and equality outwardly showing support for same-sex mar-
riage. For instance, the show includes gay characters and featured a gay marriage story line
involving Rory, Mrs Brown’s son. Her character also recorded a video on YouTube which
went viral, prior to the Irish same-sex marriage referendum (2015), where she promoted
same-sex marriage as a necessary and natural step into the future, urging voters to exercise
their right. Out of character, O’Carroll has also spoken out on gay rights, equality, and
acceptance. In addition, Mrs Brown’s role as a matriarch and an advocator of gay rights is
also relevant as historically women in Ireland were controlled by church and state, and, as
the Catholic Church does not accept or acknowledge gay sexuality, Catholic women (and
men) were expected to conform and follow the Church’s teachings. The concept of the Irish
Mammy stereotype, mentioned in this chapter, is something of a hangover from years of
social, cultural, political, and religious entrenchment where women, in Ireland, were con-
fined to a life within the home (Inglis 2005) and where their main role as ‘natural mothers’
was written into the Irish Constitution in 1937. The Irish Mammy was expected to care for
the well-being of her family and the stereotype therefore relates to exaggerated traits around
motherhood, home-keeping, and a lack of worldliness (see also Murphy and Palma-Fahey
2018 for a full discussion). This historical context is relevant to our analysis and will be
integrated into our discussion below.

370
 The representation of gay marriage

Data
The data (see Table 24.1) amounts to approximately 71,655 words, including 13 full epi-
sodes, a number of which were taken from Series 2, which was broadcast in 2012, the year
the show won three comedy awards. Each episode lasts approximately 30 minutes usually
airing after 9pm and is aimed at an adult audience. According to an Ofcom 2017 survey, it
was the fourteenth most popular show watched by adults aged between 55 and 64 in 2016.
While the dataset is small, it is nevertheless useful in providing a snapshot into a very
particular and specialised genre in terms of gender/cultural representation. O’Keeffe et al.
(2007: 198) state that small specialised corpora, such as this one, do not need to be as large
as more general corpora to yield reliable results. Koester (2010) adds that many of the limi-
tations of such a corpus, which emerge naturally, can be counterbalanced by reference to
the context which provides useful contextual clues and background data, which, in turn, can
lead to more meaningful and nuanced accounts. The scripts were accessed by downloading
them from the BBC website, initially as PDFs which were then converted to word docu-
ments and analysed using CL and insights from CDA. Six scripts unavailable online were
sent to us directly by the BBC. No changes were made to the files and stage directions and
other information were also taken into account in the analysis. In addition to the corpus, we
also draw on a one-off 2.14-minute YouTube video by Mrs Brown where she breaks away
from the programme but, still in character, urges the Irish population to vote in favour of
marriage equality. The clip went viral attracting over 2 million views and 68,000 shares on
Facebook alone and was an example of how fictional media can be used to reach wide audi-
ences to push sociocultural and sociopolitical agendas while using the popularity of shows
as a platform.

Table 24.1 The Mrs Brown’s Boys dataset

Title Year No. of No. of viewers

words (million)

1 Series 1, Episode 1 – The Mammy 2011 5935 2.64

2 Series 1, Episode 2 – Mammy’s secret 2011 5809 2.63
3 Series 1, Episode 3 – Mammy’s merchandise 2011 5819 2.68
4 Series 1, Episode 4 – Mammy rides again 2011 6283 2.99
5 Series 1, Episode 5 – Mammy of the groom 2011 5811 2.96
6 Series 1, Episode 6 – Mammy’s miracle 2011 6518 4.10
7 Series 2, Christmas Special – Mammy Christmas 2012 5073 11.68
8 Series 2, Christmas Special – The Virgin Mammy 2012 4943 10.72
9 Series 2, Episode 4 – Super Mammy 2012 5221 7.05
10 Series 3, Christmas Special – Buckin’ Mammy 2013 6851 11.52
11 Series 3, Christmas Special – Who’s a pretty Mammy then 2013 6729 11.27
12 Series 3, Episode 6 – Mammy swings 2013 846 9.10
13 Series 4, Christmas Special – Mammy’s tickled pink 2014 5823 9.85
Sub-total 71, 655
YouTube Clip – Mrs Brown’s speaking in support of 2015 340 2 million +
gay marriage prior to the 2015 Irish Referendum on
marriage equality
Sub-total 340
Total word count 71, 995

371
Bróna Murphy & María Palma-Fahey

Methodology
To analyse the corpus, we draw on a combination of CL and CDA in order to explore the
portrayal of gay marriage and the Irish Mammy stereotype in Mrs Brown’s Boys. In this sec-
tion, we will discuss our reasons for combining CL and CDA and will then provide insights
into initial trends emerging from the data which provide structure for our analysis.
In the last decade, there has been increasing evidence of interest in exploring the syner-
gies that exist between CL and CDA (e.g. Baker 2012; Cheng 2012; Flowerdew 2012; Orpin
2005) and it is clear that obvious strides have been made since McEnery and Wilson (1996)
and Biber et al. (1998) first suggested that the amount of corpus-based research in dis-
course analysis was small (Krishnamurthy 1996). Since, further research has tapped into the
potential of what the combination of these methods has to offer (e.g. Baker 2008; Mautner
2012; Partington 2008) and, in particular, has illustrated their impact on and contribution to
the exploration of language, gender, and sexuality (Baker and Levon 2015; Taylor 2013).
However, while much of the discussion to date has largely focused on how corpus tools and
methodologies can help to enhance and support CDA around its areas of weakness (Mautner
2012; Partington 2004), there are fewer studies which highlight the strengths of CDA in sup-
porting some of the shortcomings encountered in corpus research, especially where context
is key. In this chapter, we illustrate how CL can be used as one of the methods within CDA
to provide insights into gender and sexuality. We seek to demonstrate the effectiveness of
concordance lines as a way of investigating initial insights into representation. However,
when we go beyond the line, we draw then on qualitative analyses which take the wider
sociocultural and sociopolitical context into account. In this way, CDA offers a way of
moving from exploring the patterns (through CL) to a more in-depth social analysis which
provides insights into the relationship between language and other elements of the social
processes (adapted, Wetherell, Taylor and Yates 2001).
In line with Cheng (2012), we describe CL as an empirical method of linguistic analy-
sis and description which uses corpora as the primary data and starting point. It aims to
find ‘probabilities, trends, patterns, co-occurrences of elements, features or groupings
of features’ (Teubert and Krishnamurthy, 2007: 6) to arrive at generalisations about lan-
guage phenomena. In contrast, CDA views ‘language as discourse and as social prac-
tice’ (Fairclough, 2001: 21) and studies the relationship between language and ideology
(van Dijk 1997; Wodak 2001). Halliday’s (1985) approach posits that ‘language is firmly
rooted in its sociolinguistic context’ (Orpin 2005: 37) and is comprised of three stages:
description of text, interpretation of the relationship between text and interaction, and
explanation of the relationship between interaction and social context (Fairclough 2001:
21–22 cited in Cheng 2012). In considering its status, Baker et al. (2008) see CDA as
an academic movement, a way of doing discourse analysis from a critical perspective,
which often focuses on theoretical concepts such as power, ideology, and domination.
These are not just abstract concepts but concepts which can be performed within particu-
lar texts. In this chapter, we look at how they are performed focusing on how language
use reflects societal ideologies in Mrs Brown’s Boys. We also comment on the power of
scripted media to contest and push social agendas within an Irish sociocultural context.
Baker et al. (2008) state that CDA is not a method nor is it associated with a specific set
of methods and instead adopts methods such as CL, in our case, that is adequate in order
to achieve its specific aims. Likewise, they do not view CL as a single method, but rather
a collection of different methods (both quantitative and qualitative) which are related and
make use of technology as well as human input.

372
 The representation of gay marriage

Despite the growth of research in this area, most studies which draw on CL and CDA
tend to focus on how corpus approaches can help enhance and improve CDA research
(Koller and Mautner 2004; Mautner 2012; Partington 2004). Baker (2012), among others,
puts forward the proposal that CL can improve the validity objectivity of CDA research,
resulting in a more robust and valid set of findings. Mautner (2012) highlights the value of
accessing corpora to help boost the empirical credence of analyses which serves to coun-
teract the charge, frequently levelled at CDA, that individual texts are cherry-picked to suit
the researchers’ own political agenda (Orpin 2005). The merits of enriching CDA with a
corpus- driven–approach, Mautner (2012) claims, also lies in allowing researchers to look at
data from a different perspective, triangulating other forms of analyses and, in the process,
validating results. This particular focus is enlightened by recognising the epistemological
differences between both fields and the fact that corpus analyses, by virtue of their meth-
odological status, treat the text as a product rather than as an unfolding discourse. This
places emphasis on the integrity of the text, the process, and social action, which are central
concerns in CDA (Flowerdew 2012). It is clearly at the socio-analytical level that CDA
becomes more powerful as it allows more readily for the incorporation of relevant contex-
tual factors such as historicity and the interwoven relationship and sociocultural impact of
religion, politics, and gender stereotypes, as discussed below.
Being aware of the portrayal of gay marriage in the show, we were interested in explor-
ing how it was portrayed in terms of language use. We began by searching for relevant
items in the corpus through frequency, n-gram lists, and manual analysis and found items
such as marriage, gay marriage, wedding, gay wedding, gay couple, Rory, Deano + the
lemma marry. We also found items such as homosexual, lesbian, queer, and homosexuality
although these were not linked, in any way, to the portrayal of gay marriage when looked
at in the data. Therefore, our search focused in on only the forms that provided insight
into gay marriage (gay marriage, gay wedding, marry*1). The overall number of instances
for the items was low given that gay marriage was only addressed in one episode (Ep. 6).
Nonetheless, when the forms were looked at in context, they seemed to show a negative
discourse prosody2 (see Table 24.2) around how the construct of gay marriage was portrayed
in the show (lines 1, 5, 6, 7 refer to gay marriage and the Catholic Church, lines 2, 3, 4, 8

Table 24.2 Negative semantic prosody for gay marriage in Mrs Brown’s Boys

1 I thought you didn’t accept gay marriage Fr D. No Mrs Brown I said the
Church didn’t
2 Why does it have to be a gay wedding Why does it have to make a
statement?
3 ng as they’re okay with it being a gay wedding Mrs B: None of your uncles or
aunties so
4 just as long as they know it’s a gay wedding Mick: Oh they’re fine they’re broad
minded
5 Rory and Deano will never be married in the eyes of the church
6 Rory and Deano would need the pope’s married in a Church but they don’t need
permission to get anybody’s
7 thing is Rory and Deano got married This morning the two of them on
their own
8 White satin. La La: Boring it’s a gay wedding It has to make a statement
9 See the Church’s position on marriage Is that it can only exist between a
man and a

373
Bróna Murphy & María Palma-Fahey

refer to gay marriage as ‘different’). These trends were then used as a starting point for our
analysis.
The merits of combining CL and CDA in this way means that the qualitative analysis
avoids the accusation of ‘cherry-picking’ data as it is based on a trend generated from the
concordance lines. Likewise, CL avoids the limitations associated with a small dataset and
lack of contextual background in that we view each text as a semiotic entity which is embed-
ded in an immediate text-internal co-text, exploring collocates and other relevant words
and phrases in proximity, as well as an intertextual and sociopolitical context (Baker et al.
2008). It is this context that is useful in understanding underlying and often hidden mean-
ings within a sociocultural context. This is particularly important in fictional media where
there are scriptwriters involved who play a role in the reproduction of dominant/shared
knowledge. They also shape and present agendas and ideologies in society (Bednarek 2012)
that are all framed against specific cultural, social, and historical contextual backdrops.

Analysis and discussion

This analysis will focus on two main themes which emerged from the classification of the
concordance lines: gay marriage as ‘different’, and gay marriage and the Catholic Church,
and we have used these lines as a way of selecting the extracts we explore below. The
final section will explore language use by Mrs Brown in her YouTube clip about marriage
equality.

Gay marriage as ‘different’

Extract 1 appears early in the show when Rory starts planning his wedding and discussing
who to invite. This initial conversation between Rory and his mother serves to acknowledge
that a gay wedding is something that needs others’ consideration and approval (Turn 2) and
may not be accepted by everyone (Turn 3) and subsequently does not seem to be on an equal
footing to heterosexual marriage (Turn 1). Turn 3 implies potential prejudice and homopho-
bia, at least, on the part of Mrs Brown’s relatives who, she suggests, will not attend. What is
interesting, however, is that the script does not portray Mrs Brown in the same way, setting
her up instead as a potentially more open-minded character who is still seeking insights
from Rory as to how gay marriage works (Turn 1), questioning its differences regarding pro-
tocol knowing that not everyone will be accepting but still wishing to support her child even
if it is contrary to traditional social expectations. In this way, the extract brings to the fore
conflicting views around acceptance and equality which are expressed by both Rory and
Mrs Brown. It also raises awareness around societal attitudes while portraying Mrs Brown’s
lack of familiarity with gay marriage but also her willingness to understand and support her
son and thus imply support for marriage equality.

Extract 1 (Episode: ‘Gay marriage episode’/concordance line: 3)

1. Mrs Brown: How does this work with guests? I mean do we invite your uncles and
aunties?
2. Rory: As long as they’re okay with it being a gay wedding
3. Mrs Brown: None of your uncles or aunties so

Extract 3 takes place between Mrs Brown and La La, the wedding planner, and provides
further evidence of how gay weddings are viewed as being ‘different’. On the one hand, La

374
 The representation of gay marriage

La plays to stereotypical views of gay sexuality, which are often enhanced by celebrities
in the media, according to which gay weddings and behaviour have to be extravagant and
ostentatious (Turn 1). On the other hand, Mrs Brown questions and indeed challenges this
perception rejecting the notion that it should be ‘a circus’ in favour of seeing it as a celebra-
tion of love by attempting to normalise it. This further reinforces the Irish matriarch’s grow-
ing understanding and acknowledgement of gay marriage as it shows her efforts to dispel
and disregard stereotypes by contesting and challenging any limiting views in her bid to
push an agenda of equality. It is in stark contrast to conversations involving Mrs Brown in
early episodes which portray her as being uneducated around the topic of gay sexuality (see
Extract 2) and plays to the more traditional traits of the Irish matriarch stereotype.

Extract 2 (Series 1, Episode: 5)

1. Mrs Brown: He must be a lesbian too!
2. Winnie: Mick is a homo?
3. Mrs Brown: Winnie – the book says you’re not allowed to say that – it’s not politi-
cally correct.
4. Winnie: Oh what do you say?
5. Mrs Brown: I dunno, queer I think

Extract 3 also shows her move away from the power of the Catholic Church and their
opposition to gay marriage which is revealing given that historically women in Ireland had
to comply with the Church’s domination. We see Mrs Brown’s attempt again to normalise
gay marriage and in doing so perhaps dispel the stereotypes associated with it. It could also
be said that she uses the script to challenges views in order to encourage her audience to
reflect on what gay marriage is. This is in line with Hartley’s (1999) view of television as ‘a
transmodern teacher’ and also O’Carroll’s style as a writer who, in the past, has integrated
and addressed relevant social issues in his work.

Extract 3 (Episode 6, Series 3/concordance line: 8)

1. LaLa (Wedding Planner): Boring, it’s a gay wedding it has to make a statement
2. Mrs Brown: Ah
3. LaLa: What?
4. Mrs Brown: Why does it have to be a gay wedding? Why does it have to make
a statement? Why can’t it just be that Rory and Deano are in love and commit
themselves to that love together for the rest of their lives. We don’t need a circus,
we don’t need wet suits and feckin’ oxygen tanks

Gay marriage and the Catholic Church

This section looks at the second theme where Extract 4 takes place between Mrs Brown
and Fr Damien, a local Catholic priest. It shows Mrs Brown challenging Fr Damien on the
Church’s stance on gay marriage (Turns 5, 7, 9, 11) and Fr Damien’s discomfort as he fails
to respond or provide justification for the Church’s position (Turns 8, 10). This leaves him
silenced, deflated, and weak in the face of her questioning (Turns 8, 10). She focuses, in
particular, on unpicking the Church’s views on love, one of the basic tenets of their teaching,
and is shown to portray the Church as lacking and contradictory in so far as they appear to
only support, recognise, and acknowledge certain kinds of love. It is interesting to note the

375
Bróna Murphy & María Palma-Fahey

importance and relevance of the script’s focus on the Catholic Church. This could be said
to be a deliberate effort by scriptwriters to challenge and deconstruct the Church’s archaic
views towards gay marriage in light of the 2015 referendum on marriage equality. In doing
so, the script taps into the historical role of the Church in Ireland as a potential stumbling
block for equality around gay marriage. It also targets the show’s popularity with an older
demographic (55–64) (Ofcom 2017), who traditionally would have grown up following
the Church’s restrictive teachings, in a bid to perhaps educate or provide alternative views
to what they may believe. This extract also emphasises Mrs Brown’s separation from the
Church’s teachings as she overtly challenges Fr Damien. She uses her voice for justice and,
to some extent, is an embodiment of the changes that women have gained in Ireland since
the 1970s (e.g. access to contraception, permission to work, right to pensions (Inglis 2005)).
Her interaction with the priest also signals, to some extent, the diminishing power of the
Church over society in Ireland which came through her argument as well as her use of taboo
language (Turn 11). This was later evidenced in Ireland’s decision to become the first coun-
try to legalise same-sex marriage by popular vote (2015) and overturn abortion laws (2018).
This analysis has shown that going beyond the concordance lines allows access to more
holistic insights, meanings, and interpretations which only surface when the wider context
is accessed (Antaki et al. 2003).

Extract 4 (Episode 6, Series 3/concordance line: 9)

1. Mrs Brown: Fr Damien I’m glad you called around I wanted … to have Rory’s
marriage blessed in the church …
2. Fr Damien: Mrs Brown Rory and Deano will never be married in the eyes of
the church
3. Mrs Brown: I know that ha ha no I’m just looking for a blessing of some sort …
4. Fr Damien: Look the church won’t bless something not only that it doesn’t rec-
ognize but is against
5. Mrs Brown: Against what is it against?
6. Fr Damien: The church’s position on marriage is that it can only exist between
a man and a woman
7. Mrs Brown: I know that but what is the church’s position on love?
8. Fr Damien: Sorry eh ah umm well there are many different types of love I mean
like love for one’s family is different from love for one’s let’s say pet
9. Mrs Brown: Really and what’s the difference? Explain the difference between
those two loves
10. Fr Damien: Well it’s like eh
11. Mrs Brown: Look Fr I know Rory and Deano would need the pope’s permission
to get married in the church but they don’t need anybody’s permis-
sion to fall in love and as for the ceremony I don’t give a shite if they
jump over a brush but you show me in any bible anywhere where
Jesus Christ refuses to sanctify love
12. Trevor: Would you like me to show you out Fr?
13. Fr Damien: Yes I think so

This portrayal of gay sexuality in relation to the Church is taken further in Extract 5.
Here we see another interaction between Mrs Brown and Fr Damien who turns up for the

376
 The representation of gay marriage

post-wedding celebrations. When questioned by Mrs Brown as to why he had attended,

Fr Damien is shown to disassociate himself from the Church, identifying as an individual
as opposed to a representative of the Catholic institution. He says ‘I said the Church was
against it, I’m happy to celebrate love in whatever shape it comes’ (Turn 4). In light of the
context and historical background as well as our argument above, this is a powerful state-
ment as he disengages from the Church’s views to promote equality and acceptance and
asserts his support for gay marriage (Turn 2).

Extract 5 (Episode 6, Series 3/concordance line:1)

1. Mrs Brown: Fr Damien, what are you doing here (post wedding celebrations)?
2. Fr Damien: Like the rest of you I came to celebrate the happy couple
3. Mrs Brown: I thought you didn’t accept gay marriage
4. Fr Damien: No Mrs Brown I said the Church was against it I’m happy to cel-
ebrate love in whatever shape it comes

A clear advantage of exploring this small specialised dataset from a CL-CDA perspective
is that it enables us to situate the interpretation of the findings within a wider multi-level
understanding of the broader ideological context which, in turn, facilitates access to more
subtle social and linguistic patterns (Taylor 2013). However, as Baker and Levon (2015)
point out, there are shortcomings to this kind of approach. The first is that the findings
may not be generalisable beyond the text and the second is that the analysis runs the risk
of being overly subjective. To account for this, in our study, and in order to validate our
interpretation of the sociopolitical agendas mentioned above, in the next section, we look
at O’Carroll’s interaction with marriage equality beyond the show focusing, in particular,
on a YouTube video he made while in character as Mrs Brown. This dataset serves as a
means of triangulating the data in order to affirm synergy and validity of our analysis from
the show.

Gay marriage beyond the show: Mrs Brown’s YouTube clip

Through social media, Mrs Brown’s character was used to produce a YouTube clip promot-
ing marriage equality in the lead up to the Irish referendum in 2015 referring to it as ‘A
message from the Mammy in chief’. Accompanied by her son Rory who is gay, Mrs Brown
addresses the Irish nation urging them to vote in favour of same-sex marriage and, to some
extent, cements her role as an advocate for policy change around marriage equality which
clearly moves her agenda beyond the show. In Turn 1, Extract 6, she attempts to normalise
same-sex marriage, a point she had already made in the show, and questions why it should
be an issue. She draws on previous historical moments in Irish society such as marriage
between Protestants and Catholics to show how attitudes change and taps into people’s fear
of change, claiming ‘the world didn’t end’. In this extract, we see the intersection between
fiction and reality where she acknowledges the difficulties of change but draws on shared
historical sociocultural and sociopolitical background, aligning herself with the audience by
claiming ‘we can do it, we’ve done it before’. The Irish matriarch again shows her evolution
by leaving the domestic stereotype which has long characterised her by claiming for herself
a public identity and taking an active sociopolitical role in the promotion of justice. This is
reflective, to some extent, of O’Carroll’s own mother, whose character he has said he draws

377
Bróna Murphy & María Palma-Fahey

on, as she was the first female member of the Irish parliament in the 1950s and campaigned
largely for social justice. The extract once again reflects O’Carroll’s use of the media as ‘a
transmodern teacher’ (Hartley 1999), extending its use beyond the show, as he educates the
audience, reminding them of the importance of embracing change: ‘we all have to grow up
a bit now … every generation gets its change to make a big change … so go on do it’.

Extract 6 (YouTube Clip)

1. Mrs Brown: Any two people who feel in love enough should be allowed to get
married, what’s the fecking fuss?
2. Rory: Well some people believe that if you allow gays and lesbians to get
married it might change the meaning of marriage and family
3. Mrs Brown: … when I was a young girl there was a big hoo-ha about mixed mar-
riages you know Catholics marrying protestants and black people
marring white people but … They still went and got married (laugh)
and the world didn’t end, no (eh eh eh) and we all grew up a little bit
and you know we all have to grow up a little bit now. Marriage isn’t
easy changing the law isn’t easy and changing attitudes is even hard-
er, but we can do it, we’ve done it before and the world didn’t end
…. You know Rory, there was a time when woman weren’t allowed
to vote (laugh). I know, you see that’s the thing, every generation gets
the chance to make a big change and you’re going to get your chance
on May the 22nd. So go on do it.

Conclusion and future directions

Combining CL and CDA has allowed us to unveil and discuss the interplay between
patterns in the script and current sociocultural and political issues within a wider Irish
context. We show the potential of CL in accessing insights into discourse prosody but also
show where it falls short and a more qualitative analysis becomes important. This analy-
sis, where wider social context is key, provides a more holistic exploration of relevant,
political, cultural, and historical background, which is not usually considered in corpus
analyses (Baker et al. 2008). We emphasise the importance of combining methodolo-
gies and also critical engagement with different methods in order to acknowledge their
benefits and understand their limitations. This closer level of scrutiny allows us to look
at how language functions in constituting and transmitting knowledge through fictional
media and its power in organising social and political agendas. By going beyond the show
and extending our analysis to social media, we seek to triangulate our data in order to
improve the validity of our insights. Our study relies on scripts and social media clips but
stops short of including insights from media professionals which would be an important
future direction in terms of engaging them in the potential impact of language analysis
in their work. Alongside this is the importance of audience perception and opinion which
is becoming increasingly more relevant as many fictional and semi-fictional programmes
are now being accompanied by live commentary and feedback through Twitter and other
means. This new relationship between the television and audience participation has much
potential in terms of taking forward how we, as viewers, consume television shows like
this and actively engage with key issues of gender and sexuality presented and/or emerg-
ing in the shows.

378
 The representation of gay marriage

Notes
1 * indicates the lemma.
2 Discourse prosody: ‘a form of meaning which is established through the proximity of a consist-
ent series of collocates’ (Louw 2000: 57). The primary function of discourse prosody is to express
speaker/writer attitude or evaluation (Louw 2000: 58). It can be positive or negative.

Further reading
Harrington, K. (2008) ‘Perpetuating difference? Corpus linguistics and the gendering of reported
dialogue’. In: Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds.) Gender and
Language Research Methodologies. Basingstoke: Palgrave McMillan, pp. 85–102.
This chapter raises awareness of issues around corpus analysis and the interpretation of corpus data
and is useful in providing a strong methodological account of corpus linguistics in studies on language
and gender. It is useful to read both this and the previous chapter by Kosetzi (2008) in order to see the
different ways gender is approached by these two ‘methodologies’.
Kosetzi, K. (2008) ‘Harnessing a critical discourse analysis of gender in television fiction’. In:
Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J. (eds.) Gender and Language
Research Methodologies. Basingstoke: Palgrave, pp. 227–239.
This chapter is useful as it explores the importance of using CDA to analyse gender representation
in fictional media. It provides a robust methodological discussion and deals with issues of validity in
CDA.

Related topics
Perception of gender and sexuality; the impact of language and gender studies: public engagement
and wider communication; language, gender, and sexuality: reflections on the field’s ongoing critical
engagement with the sociopolitical landscape; determining the impact of gender stereotyping on
patient feedback; an ethnographic approach to compulsory heterosexuality.

References
Antaki, C., Billig, M., Edwards, D., and Potter, J. (2003) ‘Discourse analysis means doing analysis: A
critique of six analytic shortcomings’, Discourse Analysis Online, 1(1), pp. 1–24.
Baker, P. (2012) ‘Corpora and gender studies’. In: Hyland, K., Huat, C. M. and Handford, M. (eds.)
Corpus Applications in Applied Linguistics. London: Continuum, pp. 100–116.
Baker, P. (2008) ‘“Eligible” bachelors and “frustrated” spinsters: Corpus linguistics, gender and
language’. In: Sunderland, J., Harrington, K., and Sauntson, H. (eds.) Gender and Language
Research Methodologies. London: Palgrave, pp. 73–84.
Baker, P. (2005) Public Discourses of Gay Men. London: Routledge.
Baker, P. and Levon, E. (2015) ‘Picking the right cherries?: A comparison of corpus-based and
qualitative analyses of news articles about masculinity’, Discourse and Communication, 9(2), pp.
221–336.
Baker, P., Gabrielatos, C., KhosraviNik, M., Krzyzanowski, M., McEnery, T., and Wodak, R. (2008)
‘A useful methodological synergy? Combing critical discourse analysis and corpus linguistics to
examine discourses of refugees and asylum seekers in the UK press’, Discourse and Society, 19(3),
pp. 273–306.
Beacom, B. (2013) The Real Mrs. Brown: The Authorised Biography of Brendan O’Carroll. London:
Hodder & Stoughton.
Bednarek, M. (2012) ‘Construing “nerdiness”: Characterisation in the Big Bang Theory’, Multilingua,
31, pp. 199–229.
Bednarek, M. (2010) The Language of Fictional Television. London and New York, NY: Continuum.

379
Bróna Murphy & María Palma-Fahey

Biber, D., Conrad, S., and Reppen, R. (1998) Corpus Linguistics: Investigating Language Structure
and Use. Cambridge: Cambridge University Press.
Cheng, W. (2012) Exploring Corpus Linguistics: Language in Action. London: Routledge.
Fairclough, N. (2003) Analyzing Discourse and Text: Textual Analysis for Social Research. London:
Routledge.
Fairclough, N. (2001) Language and Power, 2nd edn. London: Longman.
Flowerdew, L. (2012) Corpora and Language Education. Basingstoke: Palgrave Macmillan.
Gledhill, C. and Ball, V. (2013) ‘Genre and gender: The case of the soap opera’. In: Hall, S.,
Evans, J., and Nixon, S. (eds.) Representation, 2nd edn. London: Open University Press, pp.
335–390.
Gregori-Signes, C. (2017) ‘“Apparently, women don’t know how to operate doors”. A corpus-based
analysis of women stereotypes in the TV series 3rd Rock from the Sun’, International Journal of
English Studies, 17(2), pp. 21–43.
Halliday, M. A. K. (1985) Spoken and Written Language. Oxford: Oxford University Press.
Hartley, J. (1999) Uses of Television. London and New York, NY: Routledge.
Inglis, T. (2005) ‘Origins and legacies of Irish prudery. sexuality and social control in modern Ireland’,
Eire-Ireland, 40(H3&4), pp. 9–37.
Koester, A. (2010) Workplace Discourse. London: Continuum.
Koller, V. and Mautner, G. (2004) ‘Computer applications in critical discourse analysis’. In: Coffin,
C., Hewings, A., and O’Halloran, K. (eds.) Applying English Grammar: Corpus and Functional
Approaches. London: Arnold, pp. 216–228.
Krishnamurthy, R. (1996) ‘Ethnic, racial and tribal: The language of racism?’ In: Caldas-Coulthard,
C. and Coulthard, M. (eds.) Texts and Practices: Readings in Critical Discourse Analysis. London:
Routledge, pp.129–149.
Louw, B. (2000) ‘Contextual prosodic theory: Bringing semantic prosodies to life’, in Heffer, C.,
Sauntson, H. and Fox, G. (eds) Words in Context: A Tribute to John Sinclair on his Retirement.
Birmingham: ELR, pp. 48–94.
Mautner, G. (2012) ‘Language, space and the law: A study of directive signs’, International Journal
of Speech, Language and the Law, 19(2), pp. 189–217.
McEnery, T. and Wilson, A. (1996) Corpus Linguistics. Edinburgh: Edinburgh University Press.
Murphy, B. and Palma-Fahey, M. (2018) ‘Exploring the stereotype of the Irish Mammy in “Mrs
Brown’s Boys”: Making and breaking the stereotype’, Pragmatics and Society, 9(2), pp. 297–328.
O’Carroll, B. (1995) The Chisellers. Dublin: O’Brien Press.
O’Carroll, B. (1994) The Mammy. Dublin: O’Brien Press.
Ofcom (2017) ‘Adults’ media use and attitudes 2017 Annex: Top 50 websites visited and Top 20
programmes viewed’. Available at https://www.ofcom.org.uk/__data/assets/pdf_file/0018/102771/
adults-media-use-attitudes-annex-2017.pdf (Accessed: 28th May 2018).
O’Keeffe, A., McCarthy, M. J., and Carter, R. (2007) From Corpus to Classroom: Language Use and
Language Teaching. Cambridge: Cambridge University Press.
Orpin, D. (2005) ‘Corpus linguistics and critical discourse analysis: Examining the ideology of
sleaze’, International Journal of Corpus Linguistics, 10(1), pp. 37–61.
Partington, A. (2008) ‘The armchair and the machine: Corpus-assisted discourse research’. In: Taylor
Torsello, C., Ackerleyand, K., and Castello, E. (eds.) Corpora for University Language Teachers.
Bern: Peter Lang, pp. 95–118.
Partington, A. (2004) ‘Corpora and discourse, a most congruous beast’. In: Partington, A., Morley, J.,
and Haarman, L. (eds.) Corpora and Discourse. Bern: Peter Lang, pp. 11–20.
Quaglio, P. (2009) Television Dialogue: The Sitcom Friends vs. Natural Conversation. Amsterdam
and Philadelphia, PA: John Benjamins.
Taylor, C. (2013) ‘Searching for similarity using corpus-assisted discourse studies’, Corpora, 8(1),
pp. 81–113.
Teubert, W. and Krishnamurthy, R. (2007) ‘General introduction’. In: Teubert, W. and Krishnamurthy,
R. (eds.) Corpus Linguistics: Critical Concepts in Linguistics. London: Routledge, pp. 1–37.

380
 The representation of gay marriage

van Dijk, T. A. (1997) ‘What is political discourse analysis?’ In: Blommaert, J. and Bulcaen, C. (eds.)
Political Linguistics. Amsterdam: John Benjamins, pp. 11–52.
Ward, N. (2016) ‘A brief history of the pantomime – And why it’s about so much more than “blokes in
dresses”’, The Conversation. Available at: https://theconversation.com/a-brief-history-of-the-pant
omime-and-why-its-about-so-much-more-than-blokes-in-dresses-69683 (Accessed: 7th April
2018).
Westman, K. E. (2007) Beauty and the geek: Changing gender stereotypes and the Gilmore girls’.
In: Inness, S. A. (ed.) Geek Chic: Smart Women in Popular Culture. New York, NY: Palgrave
MacMillan, pp. 11–30.
Wetherell, M., Taylor, S., and Yates, S. (eds.) (2001) Discourse as Data: A Guide for Analysis. London:
SAGE.
Wodak, R. (2001) ‘The discourse historical approach’. In: Wodak, R. and Meyer, M. (eds.) Methods
of Critical Discourse Analysis. London: SAGE, pp. 63–94.
Wood, J. T. (2007) Gendered Lives: Communication, Gender, and Culture, 7th edn. Belmont, CA:
Thomson Higher Education.
YouTube clip (2016) ‘Mrs Brown votes “Yes” for equality’. Available at https://www.youtube.com/w
atch?v=9YtKtcedZzY (Accessed: 4th April 2016).

381
 25
The impact of language and
gender studies
Public engagement and wider
communication

Deborah Cameron

Introduction
Today, researchers in many parts of the world are exhorted by government agencies, fund-
ing bodies, and the institutions which employ them to maximise the impact of their work
beyond their own academic community. Exactly how ‘impact’ is defined and measured
varies widely across systems and institutions. What different definitions have in common,
however, is the principle that researchers should contribute not only to the advancement of
knowledge in their own academic field, but also to the larger economic and social good, by
making the knowledge they produce accessible and relevant to non-academic constituen-
cies which have a stake in it, a use for it, or an interest in it. For researchers in the field
of language and gender studies this is not a new development: there is a long tradition of
engagement with both feminist and more ‘general’ publics outside the academy. But the
conditions for this kind of engagement have changed significantly over time, presenting
researchers with new opportunities, but also new challenges. What these are, and how they
can be negotiated, are the main questions addressed in this chapter.
There are a number of ways in which researchers may pursue impact, not all of which
will be considered in detail in this chapter. Some research is designed from the outset to have
applications outside the academy: it may be commissioned by, or carried out in collabora-
tion with, a non-academic partner (e.g. a government department, a charity, or a commercial
company), or it may be undertaken independently, but with the aim of informing policy or
practice in a domain such as healthcare or education. Some language and gender research
falls into this category. There is, for instance, a well-established tradition of research on edu-
cational issues like boys’ disengagement from foreign language learning (Carr and Pauwels
2006), or the classroom dynamics that tend to marginalise girls (e.g. Graddol and Swann
1988), and this work has often addressed policymakers and practitioners as well as the
researchers’ own academic community. The same is true of some research on gender-linked
patterns in other kinds of institutional discourse, including customer service interactions
(Stokoe et al. 2017), political speech in legislative assemblies (Shaw 2013), and judicial

382
 The impact of language and gender studies

proceedings relating to sexual violence (Ehrlich 2001). An interesting recent example is

the DOOM (‘Discourses of Online Misogyny’) project undertaken by corpus linguists at
Lancaster University (Hardaker and McGlashan 2016), which investigated rape threats and
other misogynistic abuse addressed to women online. This project had academic goals and
was funded by one of the UK’s research councils, but it was also intended to help companies
like Twitter devise more effective ways of dealing with the problem.
Research which is not ‘applied’ in the sense just discussed may nevertheless be designed
to benefit the research subjects and/or the community they belong to, adopting what Cameron
et al. (1992) label an ‘advocacy’ approach. This has been a common practice in sociolinguis-
tic and linguistic anthropological research, where the best-known examples have involved
researchers advocating on behalf of disempowered minority language-users and speakers
of stigmatised nonstandard dialects (see e.g. Lawson and Sayers 2016). However, as Baugh
(2006) points out, there is also an element of advocacy in some work on language and
gender, where researchers have felt a similar responsibility to counter ill-founded stereo-
types and damaging popular myths by putting counter-arguments and counter-evidence in
the public domain. This kind of public engagement may be narrowly targeted (e.g. where
researchers advocate for an individual or a community in judicial or quasi-judicial proceed-
ings), or it may be much more broadly conceived, a matter of addressing a heterogeneous
‘general public’ via the print, broadcast, and online media. The second, ‘broad’ form of
public engagement will be the main focus of the following discussion.

History
The tradition of public engagement mentioned above goes back to the early 1970s, when the
study of language and gender first emerged as a visible and coherent academic enterprise. It
is often said that this new field of inquiry was influenced by the contemporaneous Women’s
Liberation Movement, but this arguably fails to capture the closeness of the connection.
Retrospective accounts (e.g. Lakoff 2004; McConnell-Ginet 2011) suggest that the creation of
the field was an active and conscious expression of feminist political commitment. As Lakoff
(2004: 18) recalls in her introduction to the thirtieth anniversary edition of Language and
Woman’s Place (a book which was one of the earliest and most influential examples of public
engagement by a language and gender scholar): ‘It was recognized early that language was
important, that there were consequences when grown women were “girls” and the masculine
pronoun was “normal” to refer to everybody’. The deleted agent of ‘recognized’ is ‘feminists’:
it was the discussions that were happening in feminist political groups which prompted aca-
demics like Lakoff, who had been trained in linguistics or an adjacent discipline, to build on
the insight that ‘language was important’ by systematically analysing the ways in which eve-
ryday language-use both reflected and reproduced women’s subordinate status.
In relation to this early phase of the field’s history it is arguably misleading to sepa-
rate the processes of academic knowledge production and public dissemination in the way
present-day models of impact do. Scholars were bringing feminist ideas into the academy
in order to create knowledge which they would then share with feminists and others outside
the academy, and both processes were seen as contributing to the larger struggle for equality.
For this generation of (mainly women) language and gender researchers, the academy was
far from being an ‘ivory tower’ which kept them at a distance from the ‘real-world’ prob-
lems they analysed. The institutional sexism of universities, displayed in everything from
their hiring practices to the curricula they prescribed for students, was a real-world problem

383
Deborah Cameron

in its own right (for an account of the conditions facing women academics in the US in the
early 1970s, see McConnell-Ginet 2011). In these circumstances, developing courses on
‘women and language’, or publishing analyses like Lakoff’s Language and Woman’s Place
(which originally appeared in 1973 as an article in the journal Language in Society, before
being published as a general interest book two years later) was an overtly political act. No
less than the more practical initiatives in which some researchers later became involved,
such as producing guidelines for non-sexist writing (e.g. Frank and Treichler 1988; King
1991), it was an intervention aimed at changing both the academy and the wider culture.
Though the term ‘impact’ in its present-day sense had not yet entered mainstream aca-
demic usage, it is clear that much early language and gender research was intended to be
accessed and used by non-academic feminist audiences. But the conditions in which the
field first emerged were specific to their time and place, and inevitably they did not persist
unchanged. By the late 1980s, academic feminism was losing its previous organic connec-
tion to grassroots political activism as the latter began to decline. Language and gender
research was gradually becoming more visible, more organised, and more academically
‘respectable’; it was still seen by many practitioners as a ‘feminist’ area of inquiry, but
as time went on there was less consensus on what that meant. Feminist perspectives were
themselves increasingly diverse, and there were also some researchers who argued that the
field’s historical association with feminist politics and feminist theory had become an obsta-
cle to scientific progress (for an example see Preisler 1998; a feminist counter-argument
was made by Cameron 1999). Differing ideas about the field’s relationship to feminism as a
political enterprise also affected attitudes to public engagement – what kinds of knowledge
researchers should seek to communicate, to what audiences, and for what purposes. Conflict
on this issue erupted dramatically at the beginning of the 1990s, as the research community
contemplated a spectacularly successful example of a language and gender scholar making
expert knowledge accessible and relevant to a mass audience: Tannen’s book You Just Don’t
Understand (1990), which became an international bestseller and made its author a house-
hold name. Its impact was indisputable, but it was also highly controversial.
There were two main reasons for this controversy. Tannen had used a popular generic
formula (that of the self-help or advice book) as the vehicle for her argument that gender dif-
ferences in language-use arose from the differing experiences of boys and girls in same-sex
childhood peer groups, and that in later life these differences led to systematic miscommu-
nication between individuals of different sexes. This ‘difference’ or ‘two cultures’ approach
was favoured by some researchers in the 1980s (e.g. Maltz and Borker 1982; Sheldon 1990)
while others criticised it for being inattentive to the effects of power and status inequalities
on cross-sex interaction. Tannen’s use of it (and the fact that she did not address the objec-
tions others had made to it) was one of the things with which her critics took issue, in her
academic as well as her popular work. But in the case of the popular work there was another
line of criticism, which focused more directly on what some saw as Tannen’s retreat from, or
even repudiation of, the field’s tradition of feminist political engagement. In a critical review
of You Just Don’t Understand, Troemel-Ploetz (1991) accused Tannen of depoliticising lan-
guage and gender research by harnessing it to the project of individual self-improvement
rather than social and political change. For another critic, Freed (1992), Tannen’s ‘differ-
ent but equal’ message both reflected and contributed to what Faludi (1991) had recently
dubbed the ‘backlash’, a new mood of cultural resistance or hostility to feminism.
Not all researchers shared these strong reservations about You Just Don’t Understand,
and Tannen has continued to contribute to the scholarly literature on language and gender
(see e.g. Tannen 2014). However, her critics were not wrong to see the popular reception of

384
 The impact of language and gender studies

certain ideas about men and women as symptomatic of a new gender conservatism. In the
first decade of the twenty-first century, ‘difference’ accounts of male and female commu-
nication styles would be reiterated approvingly in a series of popular science books which
more clearly exemplified Faludi’s concept of the anti-feminist ‘backlash’: their argument
was that feminism had reached its natural limit, and any remaining inequalities were the
inescapable effects of ‘hard-wired’ sex differences. That certainly was not Tannen’s view,
but other writers who wanted to advance it (see e.g. Baron-Cohen 2003) found it easy to
combine her generalisations about language (often rendered in simplified terms) with the
account of sex-differentiated human nature made available by evolutionary psychology, and
findings from the newly influential current of research on the human brain that Fine (2011)
would later dub ‘neurosexism’. It was the success of this popular science genre, together
with the continuing popularity of self-help books like Men are from Mars, Women are from
Venus (Grey 1992), which prompted me to make my own contribution to the tradition of
feminist writing about language and gender for non-academic audiences, The Myth of Mars
and Venus (Cameron 2007). In some ways this text harks back to the explicitly political/
polemical writing of the 1970s; but it reflects the mood of its moment by being a more
defensive intervention, an attempt to counter the ‘post-feminist’ biological essentialism that
seemed to be in the ascendant 30 years later.
That moment has also passed: since around 2010 there has been a resurgence of feminist
political activism, particularly among young women, which has created a new audience for
the kind of politically engaged writing that was pioneered by the first generation of language
and gender researchers. At the same time, the non- or anti-feminist forms of writing which
became popular in the 1990s and early noughties have not faded away. Today, feminist and
non-feminist presentations of the subject of language and gender co-exist, and indeed com-
pete, in an increasingly crowded public ‘marketplace of ideas’.

The impact agenda: competition and confict in the contemporary

media marketplace
One recent development which has contributed to the rise of this marketplace is the advent
of the ‘impact agenda’ itself (i.e. the formal or informal adoption of impact as a goal for
researchers and the institutions which employ or fund them). In the past, communicat-
ing with non-academic audiences was something individual academics might choose to
do, but there were few professional rewards for doing it, or, conversely, penalties for not
doing it. Today there is a much stronger expectation that researchers will take or create
opportunities for public engagement, and more pressure on them to develop the necessary
skills and habits. Although formal regimes for assessing impact do not necessarily rank
the dissemination of expert knowledge to general audiences as highly as, say, applied
research or consultancy (in the UK’s Research Excellence Framework public engagement
is defined as a ‘path to impact’ rather than a form of impact in itself), some research sug-
gests that in practice assessors in humanities and social sciences have been unwilling to
discount or downgrade it (Watermeyer and Chubb 2019), and it is also valued for other
reasons (explored further below) by the institutions that employ academic researchers.
Many universities now offer training sessions on subjects like ‘talking to journalists’,
‘using social media’, or ‘making a podcast’; doctoral students and early career academics
are routinely advised to promote themselves and their work by, for instance, maintaining
an active Twitter account, starting a personal blog, or contributing regularly to an estab-
lished group blog.

385
Deborah Cameron

These new practices of professionalisation reflect another key development: the prolifer-
ation of media platforms which can be used to engage with wider publics. In the pre-digital
era only a small and select group of academics could gain access to a mass audience, and
to do so they depended on the support of professional gatekeepers (such as the editors who
commissioned ‘general interest’ books or opinion pieces in newspapers, and the producers
who selected experts to contribute to radio and TV broadcasts). Since the advent of Web
2.0, however, which enables anyone with an internet connection to become a ‘content crea-
tor’, it has become possible for academics to bypass the gatekeepers and communicate with
non-academic publics directly. Though most academics who blog, vlog, or podcast will be
addressing relatively small online audiences, the potential exists for them to attract a large
global following – and if they do, there is every chance that the mainstream media gatekeep-
ers will also show an interest. The mainstream itself has expanded to fill the space available
online, and academe is among the sources digital media producers have turned to in order
to meet their need for a continuous supply of new content. It is still relatively difficult for
an academic to be commissioned to write a trade book or a column in a printed national
newspaper, but it is not so difficult to secure a slot for an opinion piece on the same news-
paper’s website, or in an entirely virtual publication like Slate or the Huffington Post. In the
context of the impact agenda this arrangement suits both parties: researchers and scholars
gain access to non-academic audiences, while the media gain access to a dependable (and
cheap or unpaid) pool of content-providers.
As a result of these developments it has become easier than it was in the past for aca-
demic researchers to communicate with wider audiences, and for non-academics to access
expert knowledge. Overall, that must surely be regarded as a positive thing. But some com-
mentators have drawn attention to what they see as the less positive effects. For instance,
Roelofs and Gallien (2017) suggest that the use of quantitative measures as evidence of
impact has created a perverse incentive for academics to produce ‘clickbait’, seeking to
optimise the statistical evidence of impact (number of page views, downloads, citations,
comments, shares) by deliberately taking up controversial topics, making provocative
arguments, and presenting research findings in sensational terms. Roelofs and Gallien’s
comments were prompted by a case in point – an article defending Western colonialism
which appeared in Third World Quarterly in 2017. The article’s highly provocative argu-
ment prompted criticism of its scholarship, complaints about the editorial process that had
allowed it to be accepted by a scholarly journal, and a petition calling for it to be retracted.
It was later withdrawn by the publisher. In the meantime, however, and as a direct result of
the outrage it provoked, the article had become, according to widely used measures such as
the ‘Altmetric Attention Score’, the most ‘impactful’ the journal had ever published. While
this may be an extreme case, Roelofs and Gallien argue that it exemplifies a rather general
trend: ‘Academia’, they write (2017: np), ‘is replicating the structure of the mass media’,
and ‘the result is to dilute the idea of impact to simply publicity’.
The US sociologist McMillan-Cottom (2015: np) observes that under current conditions
academic institutions may well see publicity as one of the benefits of public engagement:
‘attention can be equated with a type of prestige, [and] prestige is a way to shore up institu-
tions when political and cultural attitudes are attacking colleges and universities at every
turn’. But what institutions want is positive publicity, and they are often unprepared to deal
with the negative variety. If a researcher is attacked or threatened, s/he may not always be
able to rely on her institution to protect, defend, or support her. McMillan-Cottom believes
that there is too little institutional awareness of the potential costs of public engagement,
especially for women and minority scholars. As she bluntly notes, in most fields – especially

386
 The impact of language and gender studies

those concerned with politically contentious issues such as racism and gender inequality –
‘public scholarship means pissing people off’. Individual researchers are too often expected
to bear the costs alone, while conversely, institutions take it for granted that they are entitled
to claim a share in any benefits.
Language and gender researchers are well-positioned to take advantage of opportuni-
ties for wider communication via mass media, because the subject they study does gener-
ate both interest and controversy. But precisely for that reason, researchers are vulnerable
to the problems mentioned above. Feminist researchers may find themselves under attack
for political reasons: while that is not a new phenomenon – in the 1990s a journalist once
devoted a whole column in the local paper to denouncing me and my research – it is a
different kind of experience in the age of social media, which make it easy to generate
public outrage (one of the most important currencies of what McMillan-Cottom calls
the ‘affective attention economy’), and to engage in organised harassment. The newspa-
per column attacking me provoked one letter of complaint to my employer, which was
ignored; today the university might receive hundreds of emails. Another problem for lan-
guage and gender researchers is having their work presented by the media in ways that
misrepresent it, reinforcing myths and stereotypes which the research was intended to
challenge. Conversely, researchers may be denied media attention because their work
resists this kind of framing.
Some of these points can be illustrated by looking at the way language and gender
research has been covered in The Conversation, an online publication launched in 2011 to
showcase the work of academic researchers and scholars in a form accessible to non-spe-
cialists. Staffed by professional editors and run on a not-for-profit basis, it publishes short,
topical pieces under a Creative Commons licence which allows its content to be republished
by other media outlets. Not only are contributors to The Conversation not paid, universi-
ties must pay a subscription fee to make their employees eligible to write for it. This could
be compared to the funding model for open-access scholarly publishing where the cost of
publication is paid by authors, typically using money from either a research grant or the
institution they work for, and its successful adoption by a publication whose purpose is to
facilitate wider communication shows how important the impact agenda has become. What
The Conversation offers is access to a sizeable global audience, not only via its own site, but
also through other sites which republish content from it, and this is now a commodity for
which institutions are willing to pay.
A search of the articles published on the site over a 12-month period (October 2016 to
September 2017) identified eight items which presented or discussed some kind of language
and gender research. Their titles – presumably composed by staff editors rather than aca-
demic authors – are listed below in chronological order of publication:

1. ‘He’ versus ‘she’ in Australian media coverage: what the language of news tells us
about gender imbalance;
2. Trump vs. Clinton is a chance to think more clearly about gender and leadership;
3. Improving gender equality is the key to tackling Britain’s male teacher shortage;
4. Hidden figures: how Black women preachers spoke truth to power;
5. Do men and women really find different words funny? Here’s what the research says;
6. You can tell if someone is attracted to you by their voice;
7. Female doctors show more empathy than male doctors;
8. AI can predict whether your relationship will last based on how you speak to each
other.

387
Deborah Cameron

The Conversation’s brand is built on the idea of combining ‘academic rigour [with] journal-
istic flair’, but it might be asked whether some of these titles lend credence to the argument
that the impact agenda incentivises ‘clickbait’. Items (6), (7), and (8), in particular, might
invite the criticism that they ‘sex up’ the research discussed in the body of the article, pre-
senting its findings in attention-grabbing but reductive and potentially misleading terms. In
some cases it is not just the title which invites this criticism. For instance, the writer of (7)
ends his piece by suggesting that if you are looking for an empathetic doctor you should
choose a woman. That conclusion does not follow from the research findings he presents (it
was not the case that every female doctor in the sample showed more empathy than every
male doctor, only that there was a statistically significant difference between group aver-
ages): rather it exemplifies a kind of reasoning which is implicated in stereotyping and sex
discrimination.
The items listed above represent a range of academic approaches to, and perspectives on,
language and gender. Three of them ((1), (3), and (5)) were written by linguists, while the
authors of the other five are specialists in business (2), theology (4), speech processing (6),
philosophy (7), engineering and psychology (8). In all the pieces by linguists, and two of the
others ((2) and (4)), the perspective adopted could be described as (explicitly or implicitly)
‘feminist’, in that the author aims to draw attention to sexist biases or challenge gender
stereotypes (e.g. that men’s distinctive communication style makes them better suited to
leadership roles (2) and less well-suited to early years teaching (3)). However, the remaining
three items (all taken from the science and medicine sections of the site) reproduce assump-
tions which feminist research has questioned, presenting what might be characterised as
folk-stereotypes (e.g. that women are ‘naturally’ more empathetic communicators) or spec-
ulations (e.g. about the evolution of male–female differences in voice-pitch) as neutral and
objective scientific facts. Overall, it might be concluded, The Conversation’s ‘marketplace
of ideas’ operates as a free market, fettered only by the requirement that contributors must
have reputable academic credentials. Different kinds of expertise and conflicting perspec-
tives on the same subject simply sit alongside one another, leaving it up to individual readers
to judge what is valid, useful, enlightening, or interesting.
The Conversation’s eclecticism reflects the relatively large role academic contributors
themselves play in shaping the site’s content: while access to it does depend on gatekeepers
(some contributors are approached by editors, others pitch ideas to them – and some may
be identified as potential contributors by the press and PR offices of their own institutions),
there is no particular editorial line. By contrast, gatekeepers in the traditional media tend to
be more directive. While it is possible for academics to pitch their own ideas to editors and
producers, it is more typical for them to be approached by media professionals, who will
often be looking not for an open-ended conversation about the subject of a planned story
or programme, but for someone with ‘expert’ credentials to endorse the argument they are
already committed to.
I know this from personal experience: despite being the author of a book entitled The
Myth of Mars and Venus, I am regularly contacted by media producers asking me to con-
tribute to a story or feature which presupposes the truth of some well-worn Mars-and-Venus
generalisation (e.g. ‘men can’t express their feelings’ or ‘women apologise too much’). If
I question this, what most often happens is that the producer decides she has picked the
‘wrong’ expert; after thanking me for my input she will proceed to look for a different one
who does not reject the basic premise. This is not necessarily a reflection of her personal
views on the subject: it has more to do with her professional judgement of what will make
a ‘good’ programme for a particular audience. For some audiences, a ‘good’ programme

388
 The impact of language and gender studies

may be one whose approach is compatible with my own goals as an academic researcher
(e.g. it presents the relevant research evidence fairly and allows space for key points to be
explained); in other cases, however, what the producer wants (e.g. a polarised ‘debate’ on
something the evidence is clear about, or a deliberately unchallenging ‘light’ feature) may
be in conflict with my goals. Either way, it will not be my place to dictate to the producer:
on her professional turf it will be her judgement that prevails.
This is a point which might merit more attention in the media training offered to academ-
ics. Such training tends to presuppose that access to media audiences is desirable in and of
itself, and therefore to focus on ‘how to’ questions: how to present or pitch your work so
that gatekeepers will be convinced of its relevance and interest, and how to communicate it
effectively to an audience once the gatekeepers have given you that opportunity. But there
are other questions that need to be considered. What are you hoping to achieve by presenting
ideas or research findings to this audience? Will the terms on offer allow you to achieve your
goals, or will they stop you from communicating your knowledge in a way you consider
intellectually and politically defensible? In the latter case, should you participate at all?
My own practice, which has evolved over time and in the light of experience (both good
and bad), is to decline to participate in media discussions if the answers to my questions
about the format, participants, and angle suggest that the discussion will simply reproduce
myths and stereotypes which I consider harmful; I also decline requests to contribute to fea-
tures which trivialise the subject by treating it as light entertainment (a writer or producer’s
use of the word ‘light’, or ‘light-hearted’, is always a red flag). These ‘red lines’ reflect
my own personal commitments, and I am not suggesting they should be adopted by every
language and gender researcher. What I am suggesting, though, is that all researchers need
to think about their own red lines. It is commonly supposed that the main thing that causes
problems for academics in dealing with the media is a lack of skills, but arguably an even
more significant problem is lack of control: academics do not define, and often have little
power to influence, the media’s priorities. That is why it is important to have a sense of what
you are not prepared to do – and to be willing to say no if those boundaries are not respected.
As I noted earlier, though, in the age of digital media it is possible to bypass the gatekeep-
ers and become not only the creator but also the editor and publisher of your own online
content (among academics the main kinds of self-published content are blogs and podcasts,
for which there are numerous free and easy-to-use hosting platforms). The obvious benefit
this offers is control: you decide what to present to an audience and how. The disadvantage
is that, compared to any site with a large pre-existing subscriber base, most academic blogs/
podcasts will only ever attract a small audience. Once again, though, it could be argued that
thinking in these terms ‘dilutes the idea of impact to simply publicity’, and that ‘real’ impact
has more to do with the quality than the quantity of attention. It is one thing to reach people
and another to influence them: for the latter purpose, a small but very committed reader-
ship (or one deliberately targeted on the basis that your work is directly relevant to their
concerns) may be preferable to a mass audience with no particular investment in the subject.
My own decision to start a blog (Cameron 2015b) reflected my desire to communicate
about language and gender to a particular non-academic audience, namely the new gen-
eration of active feminists I mentioned earlier. Considering their age and social location,
I believed I was more likely to reach these readers through a blog than through the mass
media (I also started a Twitter account so that I could advertise what I published to my target
audience, which is well represented on Twitter). In the event, my blog also attracted read-
ers who were not in my target group (e.g. other researchers, teachers, and, through them,
students, of language and gender). Of course I am happy to have these readers, but I would

389
Deborah Cameron

not be unhappy if I did not have them, and I continue to design the blog’s content around the
concerns of its intended, feminist audience.
In one respect, having a larger and more varied audience than I initially expected has
been a problem rather than a bonus. My original aims included facilitating discussion among
feminists via the comments section of the blog, but as traffic increased I found myself deal-
ing with a growing number of hostile and abusive comments from readers who obviously
were not feminists. The targeting of feminists (and more generally, women) online is, of
course, a well-known phenomenon (for a comprehensive discussion see Jane 2017), but I
had assumed it would not be a serious problem for an academic blogger seeking to engage
a small feminist public in discussion of a ‘niche’ subject. If the audience had remained very
small that assumption might have held up, but even a modest level of visibility – say, a post
that gets more than 1,000 shares on Facebook – can bring you to the attention of people who
will try to cause you problems. On a personal blog you can screen comments and decline to
publish those you deem inappropriate, but if the volume of comments is high this becomes
a very time-consuming task. Eventually I stopped allowing comments. I also refuse the
requests I occasionally receive to republish content from my blog in mainstream media
sources, because their threshold for deeming comments unpublishable is too low for me.
However, the fact that I receive such requests (and am sometimes quoted in the press, or
interviewed on the radio about things I have posted on my blog) shows that a ‘niche’ product
designed for a specific and limited public can sometimes influence the mainstream, through
people who have links to both.

Entering public debate: language and gender in the news

One of the roles academics play in the contemporary media marketplace is that of commen-
tators on current events and news stories, and in a world of multichannel, multiplatform,
24-hour news the demand for this kind of commentary has increased, along with the ratio of
comment to news reporting. In addition to being asked to do it, academics who self-publish
and/or are active on social media often choose to respond to news events or to the public
debate they provoke. It might be thought that rather few news stories have any connection
with the concerns of language and gender researchers, but in fact a surprising number do.
A high-profile example was the US Presidential election of 2016: both candidates’ ways
of using language came under scrutiny during the campaign (Clinton’s voice and speaking
style, in particular, were frequently discussed in highly gendered terms), and in October
2016 the release of the ‘Hollywood Access’ tape, on which Trump was heard to engage in
what he later called ‘locker room banter’, prompted an extraordinary volume of media com-
mentary. In a case like this, where pundits of all kinds are competing to offer their own ‘hot
takes’, what do researchers with specialist knowledge have to offer?
My own answer would be that academic experts can offer three interrelated things, which
I will call correction, contextualisation, and complication. Correction (as in Labov’s (1982)
‘principle of error correction’, formulated in his classic discussion of the ethical responsibili-
ties of sociolinguists to the communities they study) means trying to counter biases, miscon-
ceptions, and dubious claims. During the presidential election campaign, it was suggested,
on the basis of statistical analysis of word frequencies, that Trump talked ‘like a woman’.
The logical and technical problems with this argument were pointed out by the linguist and
data analyst Schnoebelen (2016). Contextualisation involves drawing attention to the larger
patterns which are exemplified by a currently newsworthy case. For instance, criticisms of

390
 The impact of language and gender studies

Hillary Clinton’s ‘shrill’ voice prompted charges of sexism from her supporters, and argu-
ments from her opponents that the criticism had nothing to do with her sex. In response, the
linguist Subtirelu (2016) published a chart visualising corpus evidence showing that ‘shrill’
(along with ‘screech’ and ‘shriek’) was significantly more likely to be used about the speech
of women and girls than men and boys. Other researchers put the criticism of Clinton into
the larger historical context of discourse on women’s public speech, relating the way she was
represented in media campaign coverage to the treatment of other female politicians past and
present. Complication means using expert knowledge to tell a more complex or nuanced story.
For example, Trump’s ‘locker room banter’ was generally presented by critics as an expres-
sion of his sexist and predatory attitudes to women, while his defenders maintained that it was
‘just talk’. But as someone who had studied other examples of ‘banter’ (e.g. Cameron 1997), I
wanted to point out that it is also a mechanism for reinforcing the homosocial, fraternal bonds
among men which play an important part in maintaining their social dominance. Banter thus
has effects in the world whether or not the behaviour participants talk about (e.g. in this case,
‘grabbing women by the pussy’) really happened, and whether or not the attitudes they express
in this context reflect their ‘true’ feelings.
Blogging, maintaining a social media presence, and talking to journalists who contact
you (particularly if this becomes a regular occurrence) are all activities that demand time
and effort, and since they do not usually involve any formal partnership with a non-academic
stakeholder, they may not be counted by academic institutions as part of a researcher’s ‘offi-
cial’ workload. In these conditions it is reasonable to wonder whether the time they take up
is time well spent: are you actually making a difference through your contributions to public
debate, or are you being recruited to serve the needs of an industry whose real interest is in
entertainment rather than public understanding? Arguably, the answer in most cases will be
‘a bit of both’. It is unrealistic to expect a blog post, an op-ed article, or a radio interview to
make an immediate and visible difference to public opinion or public perception, let alone
to alter the course of real-world events. But if enough commentators make the same point or
argument often enough, over an extended period of time, the effect can be to shift the bal-
ance of discourse on a particular issue, and potentially to alter the views of people who are
dealing with the issue ‘on the ground’.
An example in the field of language and gender is the emergence during the last few years
of a relatively sustained public counter-discourse critiquing the kind of ‘verbal hygiene for
women’ which, while it has existed for decades (see Cameron 1994), has become a sig-
nificantly larger and more visible enterprise in the age of digital media. Women, especially
young ones, are endlessly castigated for a series of old and new ‘verbal tics’ which are
allegedly undermining their authority and their career prospects: these include using uptalk
or vocal fry (Wolf 2015), hedging requests with ‘just’ (Leanse 2015), ‘over-apologizing’
(Lerner 2015), saying ‘I feel like’ instead of ‘I think’ (Worthen 2016), and decorating emails
with girly emoji (Best 2017). But the same conditions which have enabled other self-iden-
tified ‘communication experts’ to flood the web with this deficit discourse also make it pos-
sible for language and gender researchers to correct, contextualise, complicate, and criticise
it, and many have taken opportunities to do so. One fairly recent and striking example
occurred in 2015, when a spate of popular media articles criticising women’s speech – in
particular, their use of uptalk and vocal fry – prompted a number of experts to make public
interventions (e.g. Cameron 2015a, Madill 2015, NPR 2015; this case is discussed further
by Lawson 2016). Their arguments were also picked up and recycled by some non-aca-
demic commentators (e.g. Friedman 2015; Marcotte 2015). The expert critique of deficit

391
Deborah Cameron

approaches to women’s language-use may not have achieved the status of received wisdom,
but it does now have a place in the conversation.

Future directions
What I have referred to in general terms as ‘the impact agenda’ is unlikely to disappear in
the foreseeable future. In some contexts its importance is clearly increasing: in the UK,
for instance, the weight given to impact in the Research Excellence Framework, which is
one mechanism for allocating research funding, has been increased since this requirement
was originally introduced. In 2017 the government announced plans to introduce a par-
allel framework for evaluating and rewarding ‘knowledge exchange’: the wording of the
announcement suggested a particular interest in partnerships between academe and busi-
ness. As McMillan-Cottom (2015) notes, the ambivalence of institutions towards potentially
controversial forms of public engagement is already a challenge for researchers in some
politically contentious areas of inquiry. In future, researchers may also have to deal with
pressure to abandon or downplay critical perspectives in order to exploit the commercial
potential of expert knowledge.
In this chapter I have paid attention to the potential costs as well as the benefits associated
with public engagement. Like McMillan-Cottom (2015), I think the issue of cost or risk is
often neglected, not only by institutions (which have their own motives for encouraging the
pursuit of impact), but also among academics, who are often enthusiastic about the project
of democratising knowledge and using it for the public good. What I am trying to advocate
is not an oppositional stance towards the impact agenda itself (though there is undoubtedly
scope for criticism of the way it is implemented in particular contexts), but an approach
to public engagement and wider communication that is: (1) informed – able to draw on a
critical understanding of the marketplace in which academics must operate (developing
this understanding should be as much a goal of media training as equipping researchers
with practical and technical skills); (2) realistic – not overestimating what can be achieved
through public engagement, or underestimating the potential problems (everyone needs red
lines); and above all (3) strategic – if we regard public engagement as more than just another
box professional academics have to tick, we need to be clear about our goals in doing it,
and make decisions about how, when, and where to do it in the light of what we are trying
to achieve.

Further reading
Cameron, D., Frazer, E., Harvey, P., Rampton B., and Richardson, K. (1992) Researching Language:
Issues of Power and Method. London: Routledge.
This book predates the current ‘impact agenda’, but it continues to be an important reference point
for discussions of the power dynamics of linguistic research, and the case studies it includes are of
particular interest to researchers whose goals include sharing expert knowledge with research subjects.
Jane, E. (2017) Misogyny Online: A Short (and Brutish) History. London: SAGE.
This is a qualitative study of a phenomenon which is both of interest to language and gender
scholars (in that the form of rhetoric it analyses is highly gendered) and of practical relevance to those
involved in public engagement online.
Lawson, R. and Sayers, D. (eds.) (2016) Sociolinguistic Research: Application and Impact. Milton
Park: Routledge.
This edited collection discusses the general notion of ‘impact’ and uses a selection of specific cases
to illustrate the various kinds of impact sociolinguistic research may generate.

392
 The impact of language and gender studies

Related topics
Perception of gender and sexuality; language, gender, and sexuality: reflections on the field’s ongoing
critical engagement with the sociopolitical landscape; feminist conversation analysis: examining
violence against women; non-binary approaches to gender and sexuality; language, gender, and
sexuality: sketching out the field.

References
Baron-Cohen, S. (2003) The Essential Difference: Men, Women and the Extreme Male Brain. London:
Allen Lane.
Baugh, J. (2006) ‘Participatory research and advocacy.’ In: Encyclopedia of Language and Linguistics,
2nd edn. Elsevier. Available at: http://www.sciencedirect.com/science/article/pii/B0080448542
015030 (Accessed: 21st September 2017).
Best, S. (2017) ‘Never use emoji in work emails: Smiley face emoticons make you seem incompetent,
claim scientists’, Mail Online. Available at: www.dailymail.co.uk/sciencetech/article-4788940/Sm
iley-emoji-work-emails-imply-incompetence.html (Accessed: 9th July 2018).
Cameron, D. (1994) ‘Verbal hygiene for women: Linguistics misapplied?’, Applied Linguistics, 15(4),
pp. 382–398.
Cameron, D. (1997) ‘Performing gender identity: Young men’s talk and the construction of
heterosexual masculinity’. In: Johnson, S. and Meinhof, U. H. (eds.) Language and Masculinity.
Oxford: Blackwell, pp. 47–64.
Cameron, D. (1999) ‘Feminist linguistics: A response to Bent Preisler’s review article: Deconstructing
“feminist linguistics”’, Journal of Sociolinguistics, 3(1), pp. 121–139.
Cameron, D. (2007) The Myth of Mars and Venus: Do Men and Women Really Speak Different
Languages? Oxford: Oxford University Press.
Cameron, D. (2015a) ‘A response to Naomi Wolf’, Language: A Feminist Guide. Available at: https://
debuk.wordpress.com/2015/07/26/a-response-to-naomi-wolf/ (Accessed: 9th July 2018).
Cameron, D. (2015b) Language: A Feminist Guide. Available at: http://www.debuk.wordpress.com
(Accessed: 25th May 2020).
Carr, J. and Pauwels, A. (2006) Boys and Foreign Language Learning: Real Boys Don’t Do Languages.
Basingstoke: Palgrave-Macmillan.
Ehrlich, S. (2001) Representing Rape: Language and Sexual Consent. New York, NY: Routledge.
Faludi, S. (1991) Backlash. New York, NY: Crown.
Fine, C. (2011) Delusions of Gender: The Real Science of Sex Differences. London: Icon Books.
Frank, F. and Treichler, P. (1988) Language, Gender, and Professional Writing: Theoretical Approaches
and Guidelines for Nonsexist Usage. New York, NY: Modern Language Association.
Freed, A. (1992) ‘We understand perfectly: A critique of Tannen’s view of cross-sex communication’.
In: Hall, K., Bucholtz, M., and Moonwomon, B. (eds.) Locating Power Proceedings of the
Second Berkeley Women and Language Conference (Vol. 1). Berkeley, CA: Berkeley Women and
Language Group, pp. 144–152.
Friedman, A. (2015) ‘Can we just, like, get over the way women talk?’ The Cut. Available at: https://www
.thecut.com/2015/07/can-we-just-like-get-over-the-way-women-talk.html (Accessed: 9th July 2018).
Graddol, D. and Swann, J. (1988) ‘Gender inequalities in classroom talk’, English in Education, 22(1),
pp. 48–65.
Grey, J. (1992) Men Are from Mars, Women Are from Venus. New York, NY: HarperCollins.
Hardaker, C. and McGlashan, M. (2016) ‘“Real men don’t hate women”: Twitter rape threats and
group identity’, Journal of Pragmatics, 91, pp. 80–93.
King, R. (1991) Talking Gender: A Guide to Non-Sexist Communication. Toronto: Copp Clark Pitman
Ltd.
Labov, W. (1982) ‘Objectivity and commitment in linguistic science: The case of the Black English
trial in Ann Arbor’, Language in Society, 11(2), pp. 165–201.

393
Deborah Cameron

Lakoff, R. (2004) Language and Woman’s Place: Text and Commentaries. New York, NY: Oxford
University Press.
Lawson, R. (2016) ‘A different drum: Social media and the communication of sociolinguistic
research’. In: Lawson, R. and Sayers, D. (eds.) Sociolinguistic Research: Application and Impact.
Milton Park: Routledge, pp. 171–191.
Leanse, E. P. (2015) ‘Google and Apple Alum says using this word can damage your credibility’,
Business Insider. Available at: http://www.businessinsider.com/former-google-exec-says-this-
word-can-damage-your-credibility-2015-6?IR=T (Accessed: 9th July 2018).
Lerner, H. (2015) ‘Why women over-apologize: How and why to stop it!’, Psychology Today.
Available at: https://www.psychologytoday.com/us/blog/the-dance-connection/201511/why-wom
en-over-apologize-how-and-why-stop-it (Accessed: 9th July 2018).
Madill, C. (2015) ‘Keep an eye on vocal fry—It’s all about power, status and gender’, The Conversation.
Available at: https://theconversation.com/keep-an-eye-on-vocal-fry-its-all-about-power-status-
and-gender-45883 (Accessed: 9th July 2018).
Maltz, D. and Borker, R. (1982) ‘A cultural approach to male-female miscommunication’. In:
Gumperz, J. J. (ed.) Language and Social Identity. New York, NY: Cambridge University Press,
pp. 196–216.
Marcotte, A. (2015) ‘The war on female voices is just another way of telling women to shut up’, The
Daily Dot. Available at: https://www.dailydot.com/via/vocal-fry-99-percent-invisible-womens-voi
ces/ (Accessed: 9th July 2018).
McConnell-Ginet, S. (2011) Gender, Sexuality and Meaning: Linguistic Practice and Politics. Oxford:
Oxford University Press.
McMillan-Cottom, T. (2015) ‘Everything but the burden: Publics, public scholarship and institutions’,
Tressiemc Blog. Available at: https://tressiemc.com/uncategorized/everything-but-the-burden-pub
lics-public-scholarship-and-institutions/ (Accessed: 7th October 2017).
N[ational] P[ublic] R[adio] (2015) ‘From upspeak to vocal fry: Are we “policing” young women’s
voices?’ Fresh Air. Available at: https://www.npr.org/2015/07/23/425608745/from-upspeak-to
-vocal-fry-are-we-policing-young-womens-voices?t=1531152352125 (Accessed: 9th July 2016).
Preisler, B. (1998) ‘Deconstructing “feminist linguistics”’, Journal of Sociolinguistics, 2(2), pp.
281–295.
Roelofs, P. and Gallien, M. (2017) ‘Clickbait and impact: How academia has been hacked’, LSE
Impact Blog. Available at: http://blogs.lse.ac.uk/impactofsocialsciences/2017/09/19/clickbait-and
-impact-how-academia-has-been-hacked/ (Accessed: 21st September 2017).
Schnoebelen, T. (2016) ‘Trump does NOT talk like a woman (BREAKING NEWS: Gender continues
to be complicated and confusing)’, Medium. Available at: https://medium.com/@TSchnoebelen
/trump-does-not-talk-like-a-woman-breaking-news-gender-continues-to-be-complicated-and-con
fusing-4c0d28b41d7 (Accessed: 12th October 2017).
Shaw, S. (2013) ‘Gender and politics in the devolved assemblies’, Soundings: A Journal of Politics
and Culture, 55(55), pp. 82–94.
Sheldon, A. (1990) ‘Pickle fights: Gendered talk in preschool disputes’, Discourse Processes, 13(1),
pp. 5–31.
Stokoe, E., Sikveland, R. O., and Huma, B. (2017) ‘Entering the customer’s domestic domain:
Categorial systematics and the identification of “parties to a sale”’, Journal of Pragmatics, 118,
pp. 64–80.
Subtirelu, N. (2016) ‘Bashing Hillary Clinton’s voice: “screeching”, “shrieking” and “shrill”’,
Linguistic Pulse. Available at: https://linguisticpulse.com/2016/02/08/bashing-hillary-clintons-v
oice-screeching-shrieking-and-shrill/ (Accessed: 12th October 2017).
Tannen, D. (1990) You Just Don’t Understand! Men and Women in Conversation. New York, NY:
Morrow.
Tannen, D. (2014) ‘Gender and family interaction’. In: Ehrlich, S., Meyerhoff, M., and Holmes, J.
(eds.) Handbook of Language, Gender, and Sexuality, 2nd edn. Malden, MA: Wiley-Blackwell,
pp. 491–508.

394
 The impact of language and gender studies

Troemel-Ploetz, S. (1991) ‘Review essay: Selling the apolitical’, Discourse and Society, 2(4), pp.
489–502.
Watermeyer, R. and Chubb, J. (2019) ‘Evaluating impact in the UK’s Research Excellence Framework
(REF): Liminality, looseness and new modalities of scholarly distinction’, Studies in Higher
Education, 44(9), pp. 1554–1566.
Wolf, N. (2015) ‘Young women, give up the vocal fry and reclaim your strong female voice’, Guardian.
Available at: https://www.theguardian.com/commentisfree/2015/jul/24/vocal-fry-strong-female
-voice (Accessed: 9th July 2018).
Worthen, M. (2016) ‘Stop saying “I feel like”’, New York Times. Available at: https://www.nytimes.
com/2016/05/01/opinion/sunday/stop-saying-i-feel-like.html (Accessed: 9th July 2018).

395
 Part VI
Poststructuralist approaches
 26
Poststructuralist research on
language, gender, and sexuality
Bonny Norton (Part VI lead)

Introduction
My interest in poststructuralism in relation to language, gender, and sexuality arises from
my work as a language educator and applied linguist, and my desire to better understand
the relationship between identity and language learning (Norton 2013, 2014; Norton and
Morgan 2020). In this introduction to ‘poststructuralist methodology’ I address the complex
landscape of poststructuralism from a range of perspectives, including theories of language,
subjectivity, identity, and investment. I argue that poststructuralism constitutes a set of theo-
retical stances that offers compelling insights about identity, knowledge, and power, all of
which are highly relevant to research on language, gender, and sexuality. My discussion
incorporates insights from the five contributions to Part VI, which address poststructuralist
research in the United Kingdom, Scandinavia, the Middle East, and Australia. I examine
the research topics under investigation, the methodologies the researchers draw on, and the
insights they offer to our understanding of language, gender, and sexuality. I conclude with
reflections on future directions in poststructuralist research.

Poststructuralist theories of language

Characteristic of the ‘linguistic turn’ in contemporary thought, poststructuralist theories
assign conceptual and analytic prominence to language and all forms of meaning-making
(Derrida 1978). In poststructuralist theory, language is seen as central to the circulation of
discourses, which are characterised as systems of power/knowledge that define and regulate
our social institutions, disciplines, and practices (Foucault 1972). In poststructural terms,
discourses normalise the personal and collective possibilities we are capable of imagining
in place and time.
Structuralist theories of language, often cited as originating with the work of the Swiss
linguist, Saussure (1966), emphasised the study of the linguistic knowledge (competence)
that allowed the idealised speaker/hearer to use and understand a language’s stable patterns
and structures. Saussure’s (1966) distinction between speech (parole) and language (langue)
was an attempt to provide a way of recognising that despite geographical, interpersonal, and

399
Bonny Norton

social variations, languages have shared patterns and structure. For structuralists, the build-
ing blocks of language structure are ‘signs’ that comprise the ‘signifier’ (or sound-image)
and the ‘signified’ (the concept or meaning). Saussure notes that it is the interrelationship
between signs within a specific linguistic system that guarantees their meaning, and that
each linguistic community has its own set of signifying practices that give value to the signs
in a language. From this perspective, actual instances of language usage (performance),
which could be affected by memory lapses, fatigue, slips, errors, and so on, were not seen
as revealing of idealised patterns, and thus were of little interest in the scientific study of
language.
Poststructuralists both build on and critique Saussure’s linguistic insights. One of the
criticisms levelled at his notion of language is that structuralism cannot account for struggles
over the social meanings that can be attributed to signs within a given language. As author
A’ali (this volume) indicates, for example, the signs /Muslim/, /woman/, and /leadership/,
can have different meanings for different people within the same linguistic community. Thus
while structuralists conceive of signs as having arbitrary meanings and linguistic commu-
nities as being relatively homogenous and consensual, poststructuralists take the position
that the signifying practices of societies are sites of struggle, and that linguistic communi-
ties are heterogeneous arenas characterised by conflicting claims to truth and power. For
poststructuralists, language is not conceived of as a neutral medium of communication, but
is understood with reference to its social meaning, in a frequently inequitable world. It is
this conception of language that poststructuralists define as ‘discourse’, ‘practices which
systematically form the objects of which they speak’ (Foucault 1972: 49).
Poststructuralist theories of language, proposed by many, but particularly by the Russian
literary theorist Bakhtin (1981), saw language not as a set of idealised forms independent
of their speakers or their speaking, but rather as situated utterances in which speakers, in
dialogue with others, struggle to create meanings. For Bakhtin, language had no independ-
ent existence outside of its use, and that usage was social. He used the metaphor of speech
communication being a ‘chain’, an ongoing conversation that new speakers (e.g. children or
newcomers to speech communities) strive to join. While structural theories might see lan-
guage learning as a gradual individual process of internalising a set of rules, structures, and
vocabulary of a standard language, Bakhtin saw language learning as a process of struggling
to use language to participate in specific speech communities. Using language meant using
a tool others had used before, and Bakhtin saw speakers as constrained by those past usages.
However, he also saw speakers as able to use language to express their own meanings.
The work of the French sociologist, Bourdieu, directly addresses the poststructural-
ist study of language and power (Bourdieu 1977, 1991). While poststructuralists are not
the only theorists interested in language and power, Bourdieu explicitly drew attention to
the importance of power in structuring discourse, with interlocutors seldom sharing equal
speaking ‘rights’. For Bourdieu, ‘legitimate’ and ‘illegitimate’ speakers were distinguished
by their differential ‘rights to speech’ or their ‘power to impose reception’ (1977: 648). A
group of workers, for example, might listen more carefully to a talk given by a senior direc-
tor of a company than they would to a talk given by a junior member of the human resources
department. For Bourdieu, using language is a social and political practice in which an
utterance’s value and meaning is determined in part by the value ascribed to the person
who speaks. Recognising that the value ascribed to a person or group can vary depending
on circumstances or contexts (in Bourdieu’s terms, ‘fields’), he saw linguistic discourse as
‘a symbolic asset which can receive different values depending on the market on which it is

400
 Poststructuralist research

offered’ (1977: 651). He further noted that dominant usage is associated with the dominant
social class.
As the contributors to this section of the Handbook indicate, the poststructuralist the-
ory of language as discourse provides a framework in which to examine the relationship
between language, gender, and sexuality. If we take the position that linguistic communities
are not homogeneous and consensual, but often heterogeneous and conflicted, we can better
investigate and understand gendered relationships between individuals, communities, and
nations, and the ways in which power is implicated in struggles for human possibility.

Poststructuralism, language, and subjectivity

Poststructuralists take the position that every time we speak, we are negotiating and rene-
gotiating our sense of self in relation to the larger social world, and reorganising that rela-
tionship across time and space. The intersectional relationship between gender, race, class,
ethnicity, religion, and sexual orientation, among other characteristics, are all implicated
in this negotiation of identity. Feminist poststructuralists, in particular, argue that relations
of gender and sexuality are by no means static, natural, or inevitable (Weedon 1997). They
make the case that relations of gender and sexuality have been created through a com-
plex confluence of social and historical events, discourses, and practices. Poststructuralist
frameworks do not attempt to map out universal laws concerning women’s or men’s experi-
ences but instead work to capture how gender is socially constructed as well as how gender
inequality is challenged and resisted. In contrast to traditional theories of gender, poststruc-
turalist theorists of gender have asserted that gender is not something that one has but is
something that one does or performs in particular sociocultural and sociohistorical contexts.
This approach can be seen in the work of scholars such as Baxter (2018), Butler (1990),
Cameron (2001), and Weedon (1997) among others. From this perspective, discourse is
seen as highly significant in the reproduction of and resistance to unequal relations of gen-
der and sexuality. This approach recommends a more thorough understanding of common
discourses of gender and sexuality in local contexts so that educators, policymakers, and
community members concerned with gender inequity can challenge the reproduction of
these ideas.
Weedon (1997), one of the most well-known scholars working in the feminist post-
structuralist tradition, and a contributor to this section of the Handbook, argues that it is
through language that the individual constructs her ‘subjectivity’, which she theorises as
‘the conscious and unconscious thoughts and emotions of the individual, her sense of her-
self, and her ways of understanding her relation to the world’ (1997: 28). Her use of the
term subjectivity reminds us that an individual can be simultaneously the subject of a set
of relationships (e.g. in a position of power) or subject to a set of relationships (e.g. in a
position of reduced power). In this view, the commonsense notion of ‘the real me’ remains
a fiction. In feminist poststructuralist theory, subjectivity and language are seen as mutually
constitutive, and are thus centrally important in how gendered subjects negotiate a sense of
self within and across a range of sites at different points in time. It is through language that
subjects gains access to, or are denied access to, powerful social networks that enable or
constrain human possibility.
Weedon uses the terms ‘subject’ and ‘subjectivity’ to signal a break with dominant
Western humanist views of the individual. While Western humanist philosophy stresses the
essential, unique, fixed, and coherent core of an individual, Weedon’s view, like that of other

401
Bonny Norton

poststructuralists, is that the individual (i.e. the subject) is diverse, contradictory, dynamic,
and changing over historical time and social space. Like Foucault (1980), Weedon argues
that subjectivity is discursively constructed, and is always socially and historically embed-
ded. Post-colonial theorists such as Hall (1997) and Bhabha (1994) used poststructuralist
identity theory to analyse how categories such as race and gender have been essentialised.
In theorising cultural identity, Hall focuses on identity as changing and in process, ‘becom-
ing’, and argues that identity is ‘not an essence, but a positioning’ (1997: 226) in particular
historical and cultural environments.
This means of theorising difference has not been entirely satisfactory to those who would
assert that their identities are homogenous and unitary, foregrounding a particular aspect of
their experience such as gender, race, or religious affiliation. Current worldwide expressions
of nationalism and religious fundamentalism testify to this. Such unitary assertions of iden-
tity are often explained as ‘strategic essentialism’ in service of political goals, as discussed
by Appleby (this volume). The terms ‘identity politics’ or the ‘politics of difference’ refer-
ence this particular coalescence of identity and power relations, and recent work by Moore
(2020) has explored the possibility that poststructuralist queer theory, for example, might
not be antithetical to such approaches, but rather can be seen as a critical extension of the
same movement.

Poststructuralism, language, and learning

The poststructuralist position that language is not only a linguistic system, but also a social
practice in which experiences are organised and identities negotiated, is highly appealing
to scholars interested in identity, language, and learning, particularly with respect to issues
of gender and sexuality (Appleby 2014; Gray 2016; Kamada 2010; Moffatt and Norton
2008; Moore 2016; Nelson 2009; Sunderland 2004; Takahashi 2013). Such scholars inves-
tigate the extent to which relations of power within classrooms and communities promote
or constrain the conditions under which learning takes place, and how gender and sexuality
is implicated in negotiations of power across time and place. Such scholars take the posi-
tion that when learners speak or remain silent; when they write, read, or resist, we need to
understand the extent to which the learner is valued in a particular classroom, institution,
or community. At the same time, however, it is imporant to understand the diverse ways
in which learners may challenge both subtle and overt forms of discrimination, and what
implications this has for learning.
I have argued that in order to claim more powerful identities from which to speak, learn-
ers can challenge unequal power relations by ‘reframing’ their relationship to others. This
reframing depends, to some extent, on what I have called the learner’s ‘investment’ in the
language practices of a given classroom or community (Darvin and Norton 2015; Norton
2013). The construct of investment, informed by Bourdieu’s (1977, 1991) theories of capi-
tal, language, and symbolic power, signals the socially and historically constructed relation-
ship of learners to a given language and their sometimes ambivalent desire to learn and
practice it. I have noted,

if learners ‘invest’ in the target language, they do so with the understanding that they
will acquire a wider range of symbolic resources (language, education, friendship) and
material resources (capital goods, real estate, money) which will increase the value of
their cultural capital and social power.
(Norton 2013: 6)

402
 Poststructuralist research

Unlike constructs of motivation, which frequently conceive of the learner as having a uni-
tary, fixed, and ahistorical ‘personality’, the construct of investment conceives of the learner
as having a complex identity, changing across time and space, and reproduced in social
interaction.
The poststructuralist theory of investment has much to offer research on language, gen-
der, and sexuality. In my early research with immigrant women in Canada (Norton 2013),
I found that the adult language learners in this qualitative research study were adept at
reframing their relations with native speakers of English in order to claim more power-
ful positions from which to speak. Martina, for example, an adult immigrant woman from
eastern Europe who worked in a fast food restaurant with young native English-speaking
Canadians, claimed the identity of ‘mother’ to resist the actions of co-workers who were
marginalising her. As Martina said (italics added),

In restaurant was working a lot of children but the children always thought that I am – I
don’t know – maybe some broom or something. They always said ‘Go and clean the
living room’, and I was washing the dishes and they didn’t do nothing. They talked to
each other and they thought that I had to do everything. And I said ‘No’. The girl is only
12 years old. She is younger than my son. I said ‘No, you are doing nothing. You can go
and clean the tables or something’.
(Norton 2013: 136)

In other contexts, such as Uganda, scholars such as Andema (2014) have found that advances
in digital technology have helped female teachers reframe their relationships with others in
order to claim more powerful identities in their communities. One female teacher named
Betty noted, for example, that she ‘felt like a man’ when using a digital camera:

I feel very powerful like a man because I had never held a camera in my life. I have
always seen only men carrying cameras and taking photos in big public functions like
may be independence celebration, political rallies and wedding ceremonies. But now as
I move in the community taking pictures with my camera, I feel I am also very power-
ful, like a man.
(Andema 2014: 91)

As Andema’s research indicates, technological innovations are transforming the twenty-

first century and shifting the communication landscape, both virtually and face-to-face,
with important implications for the gendered subject, as argued by Mortenson and Milani
(this volume) and Mackenzie (this volume). To address this changing digital landscape, I
have worked with Darvin to develop an expanded model of investment that responds to the
demands of a more mobile and digital world, in which language learners move in and out of
online and offline spaces (Darvin and Norton 2015). This model recognises how the skills,
knowledge, and resources that language learners have are valued differently across diverse
contexts, and how learners need to navigate a range of belief systems and worldviews. To
capture this complex learning terrain, the model locates investment at the intersection of
identity, capital, and ideology.
Darvin and I refer to ‘ideologies’ as ‘dominant ways of thinking that organise and stabilise
societies while simultaneously determining modes of inclusion and exclusion’ (Darvin and
Norton 2015: 72). As all contributors to this section of the Handbook indicate, ideological
assumptions about gender and sexuality, whether in the United Kingdom, Scandinavia, the

403
Bonny Norton

Middle East, or Australia, guide the choices people make until these assumptions become
‘common sense’. At the same time, ideologies are not entirely determined, and forms of
resistance, such as those of Fatimah in the UK (Weedon and Hallak, this volume) and Julia
Gillard in Australia (Appleby, this volume), are powerful examples of human agency in the
face of discrimination and marginalisation. From an identity perspective, it is intriguing to
ask, ‘What was Fatimah’s investment in discourses on the burkini? How was this investment
implicated in her acts of resistance?’ Similarly, ‘What investment did Julia Gillard have in
debates on her leadership style? How was this investment implicated in her 8th October
2012 address to parliament?’.

Poststructuralist research and methodologies

As indicated in the discussion above on poststructuralism and subjectivity, feminist scholars
advocate research on gender and sexuality in local contexts so that educators, policymak-
ers, and community members can challenge the reproduction of gender inequity in these
contexts. At the same time, such research seeks to make visible larger ideological practices
in the wider society that are not necessarily self-evident. It follows that poststructuralism’s
understanding of identity as a negotiated social practice, always embedded within networks
of power, has methodological implications. Researchers adopting a poststructuralist lens
to investigate the intersections of language, gender, and sexuality pay particular reflexive
attention to the entanglement of their own subjectivities in those local contexts and the
research process in general (Norton and Early 2011). The researchers’ own multiple and
shifting identities shape the particular research questions they ask, the methods used to
answer them, and the kinds of stories participants share with them. For example, Takahashi
(2013) reveals that, as a then-novice researcher, her implicit ideas about the sorts of issues
researchers and participants should discuss and what could constitute a serious research
study initially constrained her. It was only once she had relinquished this identity and its
imagined practices that she could embrace her landmark ethnographic study of gendered and
racialised language desire among heterosexual Japanese women sojourning in Australia. In
Moore’s (2016) study of a community English class for queer learners in Japan, he enhances
his thematic analysis by foregrounding his position as one of the volunteer teachers of the
class, critically exploring the tensions emerging from competing interpellations of queer
identities by the teachers and students. In subsequent work investigating the classroom iden-
tity management strategies of queer language learners, Moore (2019) concludes that some
of his participants used the discursive space offered by the research interview process to
carry out self-affirming identity work. Recent innovative work by Prior (2016) combines
conversation analysis with a discursive constructionist approach to investigate how he, as
the researcher, and his participants – many of whom identified outside of heterosexual-
ity – discursively co-constructed and co-managed emotions as social actions through their
talk-in-interaction.
Each of five chapters in this part of the Handbook describes a case study that raises
pressing questions about gender and sexuality in a given local context, with reference to
the particular investments of the researchers. Weedon and Hallak address the 2016 burkini
ban in France and its treatment in four UK newspapers, focusing on a personal account
of a disturbing UK incident in which a Muslim woman called Fatimah wore a burkini to
a public pool. Mortensen and Milani examine heterosexual desire through an analysis of
the face-to-face interaction of two young adult Danish women, Stine and Louise, as they

404
 Poststructuralist research

navigate a heterosexual online dating site. Of particular interest is the way in which the
features of the digital site make visible Stine and Louise’s perceptions of what constitutes
a desirable male partner. Mackenzie, also drawing on the affordances of digital technol-
ogy, examines contrasting views of what constitutes a ‘good mother’ in contemporary UK
society. Drawing on her larger study of Mumsnet Talk (Mackenzie 2018), she highlights the
respective discourses on ‘child-centred motherhood’ and ‘individuality’, which has traces in
the data on the working immigrant mother, Martina, discussed above. The last two chapters
in this section, by A’ali and Appleby, respectively, address the intriguing position of women
in positions of leadership in Bahrain and Australia.
Reading across these five chapters, there are a number of compelling findings that speak
to questions of gender and sexuality across global sites. Most studies found that women
are faced with contradictory pressures and expectations, in Bahrain, Britain, and beyond
– what Appleby calls the ‘double bind’. Such contradictions include the view that women
must be effective leaders, but not be ‘bossy’; women must be good mothers, but also pur-
sue independence; women must adapt to new societies, but not reject pre-existing cultural
practices. What is evident from the studies is that while patriarchy remains a powerful force
across global sites, other forms of discrimination, including those based on race, religion,
and social class, need to be incorporated in intersectional studies of gender and sexuality.
What is also evident is that women across each of these local sites did not acquiesce to the
dominant ideologies of marginalisation, but creatively and actively resisted those practices
that sought to silence and demean them.
The chapters also provide a window on the range of methodologies associated with post-
structuralist research. All of the authors would agree with Weedon and Hallak’s central
argument that feminist poststructuralism is not a theory or methodology in the conventional
sense. Rather, ‘It is a series of critical positions on language, subjectivity, the body, discourse,
and power that provide the grounds for mapping and analysing how relations of gender and
sexuality are socially constituted, lived, and reproduced’. In this view, poststructuralism
does not prescibe specific methods for studying language, but draws on a range of different
approaches to research, consistent with the questions asked and the resources available to
carry out the research. The methodologies used in the five studies discussed included, but
were not limited to, critical discourse analysis, interactional sociolinguistics, feminist post-
structuralist discourse analysis (FPDA), and document analysis. Methodologists who were
particularly influential included Baxter (2003), Bucholtz and Hall (2015), Gumperz (1999),
Jaspers (2012), and Wetherell (2012). A’ali provided a particularly comprehensive account
of the FPDA associated with the work of Baxter (2003), which enabled her to present a com-
pelling analysis of leadership language practices used by Bahraini senior women.

Conclusion: poststructuralism and the future

Authors in the Handbook were invited to comment on future directions in language, gen-
der, and sexuality research, particularly from a postructuralist perspective. It is appropriate
to conclude this introduction by reflecting on the hopes and predictions of the scholars in
Part VI. All the authors take the position that poststructuralism is committed to analyses
that contribute towards improving social relations, and would agree with A’ali that FPDA
would be useful in such research, since it transcends the polarity of micro- and macro-
analysis. For Weedon and Hallak, what remains a concern in British society is the way rac-
ism, Islamaphobia, and ethnocentricism are produced, naturalised, and reproduced, and they

405
Bonny Norton

advocate for a research agenda that addresses the often irreconcilable position of Muslim
women in Western society. Like Weedon and Hallak, Appleby notes that future research
should continue to interrogate the way language is used in social media to shape the politics
of gender and sexuality, and also recommends that poststructuralist researchers draw on
posthumanist ideas about the ‘liveliness of matter in the more-than-human world’. Such
advice would be welcomed by Mackenzie, as well as Mortensen and Milani, whose research
projects address the interaction between humans and the digital world. However, while
Mortensen and Milani take the position that research on desire, and affect more broadly,
would contribute to an enhanced understanding of the ‘body’ in this posthumanist world,
Mackenzie remains focused on people and parents, suggesting that fathers, same-sex par-
ents, and working-class parents on online parenting sites would be very fruitful areas of
research. There is no doubt that the future of poststructuralist research on language, gender,
and sexuality will continue to be bright and bold in the years to come.

References
Andema, S. (2014) Promoting Digital Literacy in African Education: ICT Innovations in a Ugandan
Primary Teachers’ College (Doctoral dissertation). Retrieved from https://elk.library.ubc.ca/han
dle/2429/48513.
Appleby, R. (2014) Men and Masculinities in Global English Language Teaching. Basingtoke:
Palgrave Macmillan.
Bakhtin, M. (1981) The Dialogic Imagination. Austin, TX: University of Texas Press.
Baxter, J. (2003) Positioning Gender in Discourse: A Feminist Methodology. Basingstoke: Palgrave
MacMillan.
Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press: A Poststructuralist
Approach. Basingstoke: Palgrave MacMillan.
Bhabha, H. K. (1994) The Location of Culture. London: Routledge.
Bourdieu, P. (1977) ‘The economics of linguistic exchanges’. Social Science Information, 16(6), pp.
645–668.
Bourdieu, P. (1991) Language and Symbolic Power, Thompson, J. B., ed.; Raymond, G. and Adamson,
M., trans. Cambridge, UK: Polity.
Bucholtz, M. and Hall, K. (2015) ‘Identity and interaction: A sociocultural linguistic approach’.
Discourse Studies, 4(4–5), pp. 585–614.
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Cameron, D. (2001) Working with Spoken Discourse. London, UK: SAGE.
Darvin, R. and Norton, B. (2015) ‘Identity and a model of investment in applied linguistics’. Annual
Review of Applied Linguistics, 35, pp. 35–56.
Derrida, J. (1978) Writing and Difference. Chicago, IL: University of Chicago Press.
Foucault, M. (1972) The Archaeology of Knowledge, Sheridan Smith, A., trans. London and New
York, NY: Routledge Classics.
Foucault, M. (1980) Power/Knowledge: Selected Interviews and Other Writings, 1972–1977, Gordon,
C., ed. New York, NY: Pantheon.
Gray, J. (2016) ‘Language and non-normative sexual identities’. In: Preece, S., ed. The Routledge
Handbook of Language and Identity. Abingdon: Routledge, pp. 225–240.
Gumperz, J. J. (1999) ‘On interactional sociolinguistic method’. In: Sarangi, S. and Roberts, C., eds.
Talk, Work and Institutional Order: Discourse in Medical, Mediation and Management Settings.
Berlin: De Gruyter, pp. 453–472.
Hall, S. (ed.) (1997) Representation: Cultural Representations and Signifying Practices. London,
England: SAGE.

406
 Poststructuralist research

Jaspers, J. (2012) ‘Interactional sociolinguistics and discourse analysis’. In: Gee, J. P. and Handford,
M., eds. The Routledge Handbook of Discourse Analysis. London: Routledge, pp. 135–146.
Kamada, L. (2010) Hybrid Identities and Adolescent Girls: Being Half in Japan. Bristol, UK:
Multilingual Matters.
Mackenzie, J. (2018) Language, Gender and Parenthood Online: Negotiating Motherhood in Mumsnet
Talk. London and New York, NY: Routledge.
Moore, A. (2016) ‘Inclusion and exclusion: A case study of an English class for LGBT learners’.
TESOL Quarterly, 50(1), pp. 86–108.
Moore, A. R. (2019) ‘Interpersonal factors affecting queer second or foreign language learners’
identity management in class’. Modern Language Journal, 103(2), pp. 428–442.
Moore, A. (2020) ‘Understanding heteronormativity in ELT textbooks: A practical taxonomy’. English
Language Teaching Journal, 74(2), pp. 116–125.
Moffatt, L. and Norton, B. (2008) ‘Reading gender relations and sexuality: Preteens speak out’.
Canadian Journal of Education, 31(1), pp. 102–123.
Nelson, C. D. (2009) Sexual identities in English Language Education. New York, NY: Routledge.
Norton, B. (2013) Identity and Language Learning: Extending the Conversation, 2nd edn. Bristol:
Multilingual Matters.
Norton, B. (2014) ‘Identity and poststructuralist theory in SLA’. In: Mercer, S. and Williams, M., eds.
Multiple Perspectives on the Self in SLA. Bristol, UK: Multilingual Matters, pp. 59–74.
Norton, B. and Early, M. (2011) ‘Researcher identity, narrative inquiry, and language teaching
research’. TESOL Quarterly, 45(3), pp. 415–439.
Norton, B. and Morgan, B. (2020) ‘Poststructuralism’. In: Chapelle, C., ed. The Concise Encyclopedia
of Applied Linguistics. Hoboken, NJ: John Wiley & Sons Inc., pp. 901–907.
Prior, M. T. (2016) Emotion and Discourse in L2 Narrative Research. Bristol, England: Multilingual
Matters.
Saussure, F. de (1966) Course in General Linguistics. New York, NY: McGraw-Hill.
Sunderland, J. (2004) Gendered Discourses. New York, NY: Palgrave Macmillan.
Takahashi, K. (2013) Language Learning, Gender, and Desire. Bristol: Multilingual Matters.
Weedon, C. (1997) Feminist Practice and Poststructuralist Theory, 2nd ed. Oxford, England:
Blackwell.
Wetherell, M. (2012) Affect and Emotion: A New Social Science Understanding. Thousand Oaks, CA:
SAGE.

407
 27
Analysing gendered discourses
online
Child-centric motherhood and individuality
in Mumsnet Talk

Jai Mackenzie

Introduction
In this chapter, I aim to show that the concept of gendered discourses continues to be highly
relevant for the study of language, gender, and sexuality and to illustrate some of the pro-
cesses by which these discourses can be identified and analysed. Drawing on my own study
of Mumsnet Talk, an online discussion forum that targets female parents, I will show how
I have brought feminist poststructuralist theory (Baxter 2003; Weedon 1997) together with
positioning theory (Davies and Harré 1990) in order to understand and analyse Mumsnet
users’ online interactions in relation to wider social forces, and to consider what options
are available to them: as individuals, as women, as parents, and as mothers. I suggest that
this approach is particularly relevant in busy, relatively unregulated digital contexts, which
provide spaces for multiple voices to be heard.
Discourses, in the Foucauldian sense of ‘practices that systematically form the objects of
which they speak’ (Foucault 1972: 42), have been a focal point for a great deal of research
in the field of language, gender, and sexuality, particularly in the first decade of the twenty-
first century. Much of this work has focused on ‘gendered discourses’ (Sunderland 2004),
also known as discourses of ‘gender differentiation’ (Baxter 2003) or ‘gender difference’
(Baker 2008). This collective name takes in a range of overlapping and interrelated dis-
courses that work to position ‘men’ and ‘women’ in distinct and binary subject positions,
or ‘ways of being an individual’ (Weedon 1997: 3). A focus on gendered discourses offers
a way of conceptualising and drawing attention to the profoundly gendered nature of our
social world, and naming specific ways in which individuals are positioned as gendered sub-
jects. For example, Sunderland (2000, 2004) has identified the complementary discourses
‘Father as mother’s bumbling assistant’ and ‘mother as manager of the father’s role in child-
care’ in parentcraft texts. Coupland and Williams’ (2002) exploration of a range of media
texts includes analysis of three discourses of the menopause: the ‘pharmaceutical’, ‘alter-
native therapy’, and ‘emancipatory feminist’ discourses. In Baker’s (2014) corpus study
of news articles from the Daily Mail and the Mail on Sunday, he identifies discourses of

408
 Analysing gendered discourses online

homosexuality such as ‘gay people are effeminate’ and ‘being gay is a secret shame’. The
work of these authors suggests that discriminatory practices, such as excluding fathers from
the role of main parent, valuing women exclusively for their youth, beauty, and fertility, and
restricting the rights of same-sex couples, are enabled through restrictive discourses of both
gender and sexuality.
Identifying and analysing discourses, however, is not a straightforward process. This is
perhaps because of a lack of clarity around the term ‘discourse(s)’ itself, which can some-
times be used in a rather vague and uncritical way, as I will show in this chapter. Identifying
discourses is also a subjective process (Reisigl and Wodak 2009; Sunderland 2004), leaving
attempts to name and analyse them vulnerable to criticism about the reliability of the find-
ings. Further, the interconnected, shifting, and unstable nature of discourses makes them
difficult to identify and delimit (Baxter 2003; Reisigl and Wodak 2009). Finally, language,
gender, and sexuality researchers may be reluctant to name specific discourses because to do
so may give the impression that certain forms of knowledge are fixed, and therefore make
them even more difficult to challenge. This point echoes some of the criticisms levelled at
much early language and gender research that focused on the differences between language
used by and about men and (especially) women, by scholars such as Bing and Bergvall
(1996), Cameron (1996) and Eckert and McConnell-Ginet (1992). However, in this chapter
I will show that the identification and analysis of discourses from a feminist poststructuralist
perspective remains extremely valuable because it can support an appreciation of multiple
power relations and competing forms of knowledge, as well as the emergence of new and
transformative meanings in relation to gender and sexuality. I will also show how such an
approach can be realised in practice, with reference to my own research exploring construc-
tions of motherhood in the Mumsnet Talk discussion forum.

Discourses and feminist poststructuralist theory

In this chapter, I approach the concept of discourses from a feminist poststructuralist per-
spective. Feminist poststructuralism, in brief, interrogates concepts of gender, sexuality, and
identity, facilitating the exploration of what it means to be, for example, a woman or a man,
both or neither, feminine or masculine, straight, gay, or bisexual (Mills and Mullany 2011;
Weedon 1997). The theory is well positioned to examine the ways in which individuals are
defined by gendered terms like these and the dominant forms of knowledge that constitute
their meanings. Feminist poststructuralist theory has an explicitly political agenda, con-
tributing to the disruption and continual redefinition of dominant norms, expectations, and
meanings, and the gradual erosion of grand narratives around gender and sexuality (Baxter
2003; Weedon 1997). Its focus on resistance, struggle, difference, and diversity supports the
emergence of new and transformative meanings that can contribute to a rich diversity of
‘ways of being an individual’ (Weedon 1997: 3). Feminist poststructuralism’s reluctance to
settle on fixed, unitary forms of knowledge and subjectivity make it markedly different from
‘modernist’ (Baxter 2003) or ‘second wave’ (Mills and Mullany 2011) feminism, which
tends to treat ‘men’ and ‘women’ as universal, stable groups (Baxter 2003).
The influence of Foucault (1972, 1978) brings to feminist poststructuralism an apprecia-
tion that, whilst social life is complex, heterogeneous, and replete with possibilities, and
whilst meaning is shifting and unstable, powerful forces still work to fix meaning; to con-
struct the social world in specific ways. These forces can be conceptualised as ‘discourses’;
regulated groups of statements that constitute knowledge, position subjects, and inscribe

409
Jai Mackenzie

power relations. In Foucault’s (1972, 1978) view, some discourses gain the status and cur-
rency of ‘truth’, coming to dominate how we define and organise both ourselves and our
social world. In other words, certain discourses can become synonymous with popular con-
ceptions of what is ‘everyday’ or ‘normal’, acquiring a ‘very special kind of obviousness’
(Althusser 1971: 139) that makes them difficult to escape. Discourses that have acquired
such ‘common sense’ legitimacy can be described as dominant, and they often work to mar-
ginalise other discourses that are not institutionally legitimised or widely recognised as the
‘norm’. Our sense of who we are and what we know is thus regulated through discourses,
especially dominant discourses that are entrenched in social structures and institutions such
as hospitals, schools, and prisons (Foucault 1967, 1972, 1977).
The relationship between knowledge, power, and subjectivity within Foucauldian and
poststructuralist theory raises questions about whether we have any control at all over our
conception of our world and ourselves. The way this relationship is understood will depend,
to a degree, upon the analyst’s interpretation of power. In this chapter, commensurate with
the work of poststructuralist theorists who avoid defining power in terms of stark contrast
or rule (Bakhtin 1981; Foucault 1978), I conceptualise the relationship between knowledge
and subjectivity in terms of power relations that are plural and competing (see Baxter 2003).
From this perspective, some discourses may be identified as ‘dominant’ or ‘marginalised’
in a particular culture or context, but this does not mean they are universally dominant
or marginalised, that the forms of knowledge and subjectivity they legitimise are fixed,
or that they cannot be challenged. Such a relational view of power can allow language,
gender, and sexuality researchers to move away from binary, ‘top-down’ perspectives that
position women as oppressed and constrained within a patriarchal system, and towards an
approach that can emphasise silenced, suppressed, and marginalised voices, allow for the
possibility that discourses can be negotiated, contested, or resisted, and thus give rise to
new and transformative meanings, behaviours, or ways of being an individual (Baxter 2003;
Mills 2003). Some recent studies of language and gender that achieve these goals include
Corwin’s (2017: 272) analysis of the speech of 15 genderqueer individuals, which focuses
on how gender emerges in interaction. In this study, Corwin (2017: 273) shows how one of
her participants makes flexible use of a range of embodied signs such as voice pitch and ges-
tures towards different parts of the body to both draw on indices of binary gender, but also
to resist them, and thus to create ‘new gender expressions’. In a different context, Baxter’s
(2018: 3) study of the way women leaders are portrayed in the UK press provides readers
with a toolkit for deconstructing and challenging stereotyped and sexualised portrayals of
these leaders, by reading news media ‘against the grain’.
Placing discourses at the heart of language, gender, and sexuality research can enhance
explorations of what might be called cultural ‘norms’ or ‘expectations’ around gender, sexu-
ality, and identity. By identifying and naming discourses, and deconstructing the ways in
which they operate through language, analysts can specify the forces that may both enable
and restrict different ways of understanding issues of gender and sexuality. As noted above,
however, discourses are complex, unstable, and shifting entities. This may explain why the
ways in which they are defined and identified are so rarely made explicit in research across the
social sciences. For example, sociologists exploring norms and expectations around gender
and parenting – the theme of the Mumsnet study that will be detailed below – have named
a range of discourses of parenthood such as ‘intensive mothering’ and ‘child-centredness’
(Wall 2013), ‘equality’, and ‘involved fatherhood’ (Miller 2011), without making the means
by which they come to name these discourses explicit. Readers are consequently relying on
the authors’, as well as their own, intuitions and assumptions in order to understand what

410
 Analysing gendered discourses online

it is that reveals the presence of these discourses, and indeed, why they are ‘discourses’ at
all: they might just as usefully be called ‘themes’ or ‘ideas’. Analyses like these will often
raise important issues, but they are unlikely to reveal very much about exactly how forms of
knowledge about gender and parenthood are recognised in the first place, how they operate,
compete, merge, and combine and, importantly, how they can be negotiated and challenged.
To offer (or follow) a prescribed, definitive method for identifying discourses, however,
would be counter to poststructuralist principles in many ways: the very nature of post-
structuralist thought encourages the analyst to embrace multiple perspectives; to resist pre-
scription and claims to ‘truth’ or ‘objectivity’ through ‘scientific’, ‘precise’ methodologies
(Graham 2005: 3). However, several discourse analysts, especially critical discourse ana-
lysts, have offered some guidance by describing their methods for discourse identification
and analysis in detail. For example, van Leeuwen (2009) focuses on a text’s representa-
tion of actors, actions, times, and places, through markers such as lexical choice and verb
type. Baxter (2010) examines lexical choices, turn-taking, and verb tense, whilst Reisigl
and Wodak (2009: 94) locate what they call ‘discursive strategies’ such as nomination and
predication, whereby people, objects, phenomena, and processes are named and charac-
terised, or intensification and mitigation, whereby the force of a statement is heightened
or reduced (Reisigl and Wodak 2009: 93–95). A range of factors inform the choices these
scholars make about which linguistic features to explore or emphasise, including their own
perspectives and the nature of both the context and the issues they investigate. For example,
Baxter’s (2010) attention to turn-taking practices is particularly appropriate for the analysis
of spoken interaction. Fairclough’s (1992) focus on transitivity, theme, and modality reveals
his commitment to the systemic functional linguistic approach and its applicability to writ-
ten discourse. What these analysts have in common, however, is their systematic evidenc-
ing of discourses through close scrutiny of language. This is based on the principle that,
although we may not be able to ‘see’ an entire discourse on the page, what we can see are the
linguistic practices through which discourses operate. It is through language, after all, that
discursive struggles are acted out (Mills 2004), and so it is through an analysis of language
that discourses can be reconstructed.
Sunderland’s (2000, 2004) inductive, ‘bottom-up’ approach to identifying and naming
discourses has been particularly influential in the field of language, gender, and sexuality. In
her analysis of gendered discourses in parentcraft texts (leaflets and books about pregnancy
and child care), Sunderland (2000, 2004) works to recognise and name discourses by iden-
tifying their linguistic ‘traces’. As she explains, discourses are not concrete entities, waiting
to be ‘spotted’, and are never truly present in a text in their entirety. But linguists can pin-
point linguistic features which hint at the existence of a particular discourse, and treat those
features as a starting point in the reconstruction of that discourse. For example, through her
scrutiny of the way the recurring linguistic items ‘play’, ‘fun’, ‘help’, and ‘share’ are attrib-
uted to male and female parents, Sunderland (2000, 2004) uncovers some of the gendered
discourses at work in parentcraft texts. So, she suggests, the ‘Part-time father/Mother as
main parent’ discourse ‘is realised through the recurrence of help’, which is largely attrib-
uted to fathers, and the ‘Father as baby entertainer’ discourse ‘is realised through recurrences
of play, fun and enjoy’, again attributed largely to male parents (Sunderland 2000: 265, her
emphasis). In keeping with Foucauldian poststructuralist theory, Sunderland pays attention
not only to what is present in the text, but also to what is absent. For example, the absence
of the linguistic items ‘share’ and ‘paternity leave’, as well as the backgrounding of fathers
through lack of specific reference to men as parental subjects, also points to the ‘Part-time
father/Mother as main parent’ discourse. Sunderland (2000: 255) develops her exploration

411
Jai Mackenzie

of the ways in which discourses operate and her identification of the discourses themselves
concurrently, so that discourses are ‘both the object and the result’ of her analysis.
Despite the examples that have been offered above, there still appears to be relatively
little methodological guidance on the identification and analysis of discourses in an oth-
erwise extensive field of language, gender, and sexuality studies. In this chapter, I aim to
address this problem by outlining my own approach to identifying and analysing discourses
of gender and parenthood in Mumsnet Talk interactions. This approach is aligned with
Sunderland’s (2000, 2004) in many respects, as well as sharing a similar research context
and aims. For example, I take an inductive approach, which involves identifying discourses
and analysing the mechanisms through which they operate concurrently. I also pay attention
to both recurrent and absent linguistic features, taking the view that it is not just what is said,
but what is not said, that can point to the presence of discourses. However, I take the discur-
sive analysis of language, gender, and parenthood to a new context, that of an online discus-
sion forum for parents, in which there is more space for a diverse range of voices to be heard
than in the relatively fixed content of parentcraft texts. The digital context of this forum also
means that it is not only linguistic but visual and other typographical features that are of
interest; accordingly, my approach is able to incorporate a range of semiotic forms. Finally,
my commitment to feminist poststructuralist theory, especially the Foucauldian perspective,
leads me to place more emphasis on the discursive nexus of knowledge, power, and subjec-
tivity than Sunderland (2000: 261), who tends to treat discourses more as groups of ideas
or values that recur in texts, and ‘ways of looking at the world’, than as powerful regulatory
practices that work to govern their subjects’ minds and bodies.

The Mumsnet study

This section turns to my study of multi-party interactions within the Mumsnet Talk discus-
sion forum, which aimed to explore how Mumsnet users negotiate discourses of gender and
parenthood in this context (Mackenzie 2017, 2018, 2019). This study was conducted in two
stages: ‘data construction’ and ‘identifying and analysing discourses’, which are detailed
in full in Mackenzie (2019). In this section, I focus on the main part of the second stage,
explicating the iterative process of exploratory linguistic analysis, discourse identification,
and discourse analysis that is shown in Figure 27.1.
The thematic coding and categorisation of 50 Mumsnet Talk threads in the first stage of
this study, and focused coding of two selected threads in the second stage (see Mackenzie
2019), led me to identify a number of potential discourses at play in these digital conver-
sations. As the second stage of analysis developed, I further investigated the presence of
these potential discourses in two Mumsnet Talk threads. This exploratory analysis focused,
following Sunderland (2000, 2004), on identifying ‘traces’ of these discourses. As well as
identifying linguistic traces, I also paid attention to digital aspects of Mumsnet Talk inter-
actions, paying equal attention to the meanings produced through the use of non-linguistic
features such as images, emoji, and strikethrough text. This analysis was not bound by a
predetermined framework; I did not set out to investigate any specific linguistic and digital
features, but to discover which features emerged as significant in relation to my aims.
In order to operationalise an analysis that was consistent with feminist poststructuralist
theory, I drew on Davies and Harré’s (1990) positioning theory, which considers how indi-
viduals are positioned as subjects through social interaction. This conceptual framework
facilitated a two-sided approach that focused not only on identifying traces of discourses,

412
 Analysing gendered discourses online

Figure 27.1 Research design for the Mumsnet study.

but also analysing the way these discourses operate through the discursive nexus of knowl-
edge, power, and subjectivity. From this perspective, I explored what meanings about
gender and parenthood were taken up and legitimised at different moments, through an
exploration of how contributors’ use of particular linguistic and digital resources worked
to legitimise or challenge particular forms of knowledge and/or position them in particular
ways. I also considered what power relations were inscribed by those forms of knowledge
and subjectivity, including considerations of whether they were dominant or marginalised,
empowering or restrictive. Some of the results of this analysis will be exemplified in the
section that follows.

Identifying and analysing discourses of ‘child-centric motherhood’

and ‘individuality’ in Mumsnet Talk
This section details part of my analysis of a single thread posted to Mumsnet Talk in the
summer of 2014: ‘Your identity as a mother’. In this thread, multiple relations between gen-
der and parenthood are explored and negotiated as participants openly discuss their sense of
self, particularly in relation to the category ‘mum’. The title and opening post of this thread
(Extract 1) sets out the agenda: to explore people’s experiences of motherhood, especially
how motherhood changes them, and their view of themselves. In this section, I pinpoint
some of the linguistic and digital traces that led me to identify the presence of two dis-
courses in this thread: ‘child-centric motherhood’ and ‘individuality’. I also consider how

413
Jai Mackenzie

these linguistic and digital forms work as discursive resources that contributors draw on to
position themselves in relation to these discourses.

Extract 1. Opening post to ‘Your identity as a mother’ (all extracts are reproduced
as they appear in the original posts)

pandarific Sun 01-Jun-14 14:43:17

1. I’ve been reading a lot of fiction that deals with motherhood and family relationships
2. and I’m curious as to how it changes people, and their view of themselves. Has your
3. perception of who you are changed since you had children? How much of your identity is
4. bound up with being a mum? Do you think the strength of your desire to be a
5. mum/what stage in your life you had them affected the degree of the changes?

6. For some reason this has come out reading like an exam question – it’s not meant to be!
7. Just curious about people’s experiences.

‘Child-centric motherhood’
The first linguistic trace of the ‘child-centric motherhood’ discourse in ‘Your identity as a
mother’ is a construction that includes the category ‘mum’ (an extremely common category
across the thread), qualified by an intensifier, as in the examples shown below.

Post 3. cakesonatrain. I think I am almost entirely Mum.

Post 11. EggNChips. As soon as I became a mum, I was 100% mum and loved it…
Post 16. cakesonatrain. I am almost wholly Mum,
Post 18. Kath6151. I have been so intensely mum for the last 10 months

In these examples, contributors’ use of the intensifiers ‘almost entirely’, ‘100%’, ‘almost
wholly’, and ‘so intensely’ conveys a sense that it is difficult for them to be anything other
than a ‘mum’; that their sense of self is intensely bound up with the subject position ‘mum’.
Traces of a discourse of ‘child-centric motherhood’ can also be found in the ‘Your iden-
tity as a mother’ thread where participants express positive feelings for and commitment
to their children. In the following examples, contributors employ a range of linguistic and
visual resources that suggest their love and care for their children is obvious and certain; that
their feelings are just ‘common sense’.

Post 18. Kath6151 My priority is now 100% my DS [darling son] though

Post 21. PoundingTheStreets. I love my DC [darling children]
Post 34. AssertiveDecorations. I feel I barely identify as a parent at all even though I love
the DC to death

Contributors again employ intensifiers in these posts; ‘100%’, the smiling emoji , and ‘to
death’. Rather than qualifying the nature of their position as ‘mums’ (or indeed ‘parents’),
however, in these instances they are qualifying the nature of the love for their children
(Posts 21 and 34), and the extent to which their children are a ‘priority’ (Post 18). Indeed,
AssertiveDecorations rejects any self-identification as a parent, but balances this assertion
with a hyperbolic statement of love for her children. These contributors also draw on a

414
 Analysing gendered discourses online

range of other resources to confirm their position as child-centric mothers. For example,
PoundingTheStreets’ monosyllabic sentence ‘I love my DC’ is positioned in a separate line
at the end of her post, making her unmitigated declaration of love even more emphatic. The
smiling emoji implies certainty that her statement of love will be well received by read-
ers. All of the above examples are taken from the final line of posts, adding to the force
and finality of these statements. These linguistic and digital resources suggest that mothers
loving their children and putting them first is a form of knowledge that has common sense
legitimacy in this context.
Close analysis of Post 23 (Extract 2) further evidences the presence and dominance of a
discourse of ‘child-centric motherhood’ in the ‘Your identity as a mother’ thread, and shows
how one contributor positions herself in relation to this discourse.

Extract 2. Post 23 to ‘Your identity as a mother’

Queen of Louisiana Sun 01-Jun-14 23:33:27

1. It was DS’s [darling son’s] birthday yesterday, so I was taking stock of life. I always do
2. when that date comes around!

3. I am still very bound up in being a mum although that role is now more about
4. promoting self reliance and supporting his independent skills. There is now a greater
5. balance of give and are [sic] between DS and I, I enjoy his company and we do
6. things together that we love. However, I am still astonished at times that I give up so
7. much time to encourage his interests (hours at the edge of a rugby pitch, early mornings
8. at a swimming pool) and I can only assume that this is pure maternal love!

In this post, QueenofLouisiana (henceforth ‘Queen’) draws on a range of linguistic resources

to position her son at the centre of her life, and thereby herself as a ‘child-centric mother’.
Between lines 6 and 8, she makes it explicit that she puts his needs before her own, for
example through her use of the verbal phrase ‘give up’ in line 6, which implies that she is not
only passing time, but that this is a selfless act that results in loss of time for herself. Queen
also suggests that she goes to extreme lengths in her commitment to her son, using the
intensifying qualifiers ‘so much’ and the potentially limitless descriptor ‘hours’ and ‘early
mornings’ (line 7) to emphasise the amount of time she has given to him and the extreme
nature of her commitment. The adverbs ‘always’ (line 1) and ‘still’ (line 6) suggest that
her commitment to her son is ongoing. Queen’s post also suggests that this child-centred
behaviour is involuntary. For example, from lines 6 to 8, she expresses surprise at, and lack
of understanding of, her own behaviour through the subordinate clause ‘I am still astonished
at times’ (line 6). The metapragmatic function of this clause, in which Queen reflects on her
feelings about her own actions, can be interpreted as unease with her position as a child-
centric mother, but at the same time implies that her actions are instinctive and ‘natural’,
thereby reinforcing the common sense status of this subject position. When Queen goes on
to attribute her behaviour to ‘pure maternal love’ (line 8, my emphasis), she seems to settle
on the latter interpretation; that her actions are part of the ‘obvious’ behaviour of a mother.
However, in line 8 she mitigates her explanation with another metapragmatic clause: ‘I can
only assume’. Here, the adverb ‘only’ suggests that this explanation still does not sit easy
with her; that ‘assumptions’ are all she can make, in the absence of any rational explanation.

415
Jai Mackenzie

My identification of a discourse of ‘child-centric motherhood’ in the ‘Your identity as a

mother’ thread echoes the findings of previous studies of motherhood in Western contexts.
Sociological studies by Hays (1996), Lawler (2000), Lowe (2016), and Wall (2013), for
example, all point to pervasive expectations that being a mother, and especially a ‘good’
mother, is often conflated with being completely child-centred and self-sacrificing. My
analysis of the ways in which this discourse is taken up and negotiated in ‘Your identity as
a mother’ demonstrates how forms of knowledge that emphasise the imperative positioning
of mothers in relation to children are mapped on to the subject position of the ‘child-centric
mother’. The positioning of self exclusively in relation to children, however, is often chal-
lenged in the ‘Your identity as a mother’ thread through an assertion of a self that is distinct
from the subject position ‘mother’, and the needs of children. The following section turns to
this site of opposition, in the form of a discourse of ‘individuality’.

‘Individuality’
Just as a discourse of ‘child-centric motherhood’ is realised partly through participants’ self-
identification as child-centric mums, a discourse of ‘individuality’ is repeatedly taken up in
‘Your identity as a mother’ through participants’ self-identification as individuals. This is
apparent in the following excerpts, where participants all employ a variant of the clause ‘I
am me’, in which a personal pronoun takes both grammatical subject and object position.

Post 12. IdealistAndProudOfIt. I am me as I have always been

Post 14. Casmama. I am me.
Post 44. catsrus. I am who I am.
Post 72. museumum. I am totally me.. the same me as before..

Through this double reference to self, participants make their claim to individuality. These
claims are made particularly emphatic in Posts 14, 44, and 72, where the clause stands as a
complete sentence, and in Post 72, where the participant uses the intensifier ‘totally’ to make
explicit her statement that she is ‘completely’ herself.
Further microlinguistic analysis of whole posts suggests that ‘individuality’ and ‘child-
centric motherhood’ often compete in the ‘Your identity as a mother’ thread. For example, in
Post 13 (Extract 3), these discourses seem to be at the centre of Crazym’s struggle to define
her own subjectivity.

Extract 3. Post 13 to ‘Your identity as a mother’

Crazym Sun 01-Jun-14 19:07:32

1. Hate being identified as “ mum”.
2. I was a person before I became a mum and that person still exists.being a mum is
3. just a part of who I am, not the whole.
4. Used to hate the silly bint at nursery who, when I went to collect the Dcs would say
5. “ and how are you today, mum?”
6. I have a name!!!! I am a person!!

By opening her post with the negative evaluation ‘hate’, Crazym resists being subject posi-
tioned exclusively as a ‘mum’ in favour of a more individualistic subject position; the ‘I’

416
 Analysing gendered discourses online

introduced in line 2. The opening and closing statements of her post capture her emphatic
resistance by presenting the subject positions ‘mum’ and ‘person’ as oppositional: ‘Hate
being identified as “mum”… I am a person!!’ Her use of six exclamation marks in two
declarative four-word sentences in line 6, furthermore, suggests that she is fighting to
express her individuality, and that by positioning herself as an individual, she resists being
positioned exclusively as a mum. Crazym’s rejection of being identified as a mum suggests
that, for her, this subject position restricts her access to other ways of being – especially as
an individual; a person in her own right. Despite Crazym’s emphatic resistance to ‘being
identified as “mum”’, however, she positions herself in this way through the relational pro-
cesses ‘became a mum…/ being a mum’ in line 2. She works to avoid positioning herself as
a ‘child-centric mother’, however, in the statement ‘a part of who I am, not the whole’ (line
3), which reiterates her partial identification with this subject position.
The oppositionality between ‘individuality’ and ‘child-centric motherhood’ is also evi-
dent in Post 59 (Extract 4).

Extract 4. Excerpt from Post 59 to ‘Your identity as a mother’

Viglioso Wed 04-Jun-14 08:26:39

1. Can you guess some have been a PITA [pain in the arse] already lecturing me (good
2. mums don’t, apparently, wear make-up: that money/time could be spent on PFB
3. [precious first born]).

4. Interestingly one of the most devoted mum in terms of practical things and
5. passionate adoration of PFB I know (of child with a disability requiring lots of care and
6. special input) is very much - and vocally - her “own woman” with her child by her side
7. IYSWIM

8. I’m actually a bit terrified of the “if you have any time for yourself you’re neglectful”
9. brigade. As I mentioned above, if anything I’ll end up accidentally attached or just
10. spoilPFB due to PFB being a bit of a miracle... but I would like to be allowed to be me.

In this post, Viglioso draws on a discourse of ‘individuality’ as part of her resistance

to a discourse of ‘child-centric motherhood’. The discourse of individuality is voiced
here through the ‘devoted mum’ she describes between lines 4 and 6. The self-possessive
‘own’ within the label ‘her “own woman”’ (line 6) works to suggest that this individual
is in control of her life; that she is able to determine her own subjectivity and is not con-
trolled by others. The way Viglioso positions children in relation to the ‘devoted mum’
also points to a discourse of ‘individuality’. Where, in lines 1–3, the child (‘PFB’ – ‘pre-
cious first born’) is positioned in a passive role, the child of line 6 is positioned as co-
existing alongside the ‘devoted mum’, through use of the prepositions ‘with’ and ‘by [her
side]’. This positioning of adult and child points to their co-existence as separate individu-
als, with neither being entirely reliant on the other. Viglioso shows that she approves the

417
Jai Mackenzie

‘individuality’ discourse by aligning with this ‘devoted mum’, through positive evalua-
tions such as ‘devoted’, ‘passionate adoration’, ‘care’, and ‘special input’. Viglioso also
positions herself more explicitly as an individual through her use of the personal pronoun
‘me’ in her closing statement ‘I would like to be allowed to be me ’ (line 10; her
emphasis). As with PoundingTheStreets’ post (see above), Viglioso’s use of a smiling
emoji at the end of her post seems to function as an emphatic full stop that gives a sense
of finality to her statement of individuality. As noted by Gibson et al. (2018), the function
of emoji is highly context-dependent and variable. Given Viglioso’s complex negotiation
of ‘child-centric motherhood’ and ‘individuality’, in which she both relies upon, but also
resists, the position of the ‘child-centric mother’, the placement of the smiling emoji at
the end of her post could also be read as a resource for mitigating this final statement and
encouraging others not to read it too seriously.
The above analysis shows how the discourses of ‘individuality’ and ‘child-centric moth-
erhood’ can compete in the context of the ‘Your identity as a mother’ thread, with con-
tributors such as Crazym and Viglioso working to position themselves within a discourse
of ‘individuality’ as part of their resistance of a discourse of ‘child-centric motherhood’. At
such moments, the subject position ‘me’ can become powerful and transformative, offering
Mumsnet users control over their sense of self and access to multiple possible ways of being
an individual, as well as contributing to the gradual erosion of what seems to be a dominant
discourse of ‘child-centric motherhood’. It is partly through the analysis of these struggles
to define forms of knowledge and subjectivity that ‘individuality’ and ‘child-centric mother-
hood’ are more confidently identified as discourses.

Discussion
This chapter has shown that the study of language, gender, and parenthood in a digital con-
text can be enriched through a focus on identifying and analysing discourses from a femi-
nist poststructuralist perspective. I have shown both how discourses such as ‘child-centric
motherhood’ and ‘individuality’ can be identified through examination of the linguistic and
digital traces of these discourses, and how the forms of knowledge, power, and subjectiv-
ity that operate through these discourses can be further analysed by drawing on Davies and
Harré’s (1990) positioning theory. For example, my analysis reveals that a discourse of
‘child-centric motherhood’ can work to position Mumsnet users entirely in relation to their
children, as devoted, loving parents, to the exclusion of other potential subject positions. On
the other hand, I also show that many Mumsnet users work to resist this dominant discourse
by drawing on a competing discourse of ‘individuality’ to position themselves as individu-
als, rather than members of a generic category of ‘mothers’.
The insights from this analysis reveal some of the demands, expectations, and restric-
tions that are placed on parents, especially mothers, in contemporary British society. For
example, the oppositional relations between ‘child-centric motherhood’ and ‘individuality’
point to the struggles and frustrations for some Mumsnet users at being unable to escape
gendered subject positions such as the ‘child-centric mother’, and show that notions of
child-centric love and devotion are intricately bound up with their sense of what it means
to be a mother. Such persistent expectations that mothers will be entirely child-centred will
make it difficult for male parents to adopt child-centred roles within the family, and difficult
for female parents to adopt valued and legitimised roles outside of their relation to children.
However, this analysis has also pointed to the emergence of new and transformative mean-
ings that can contribute to more richly diverse ways of being a parent.

418
 Analysing gendered discourses online

Future directions
Digital contexts such as the Mumsnet Talk forum can offer interesting and relevant sites for
exploring, challenging, and destabilising dominant discourses of gendered parenthood such as
‘child-centric motherhood’, providing unprecedented access to the perspectives of individu-
als in diverse family groups, as well as the discursive struggles and transformative practices
in which they may be involved. The Mumsnet study has revealed some important insights in
this area, but issues around language, gender, and parenthood are equally likely to be relevant
to groups who are not adequately represented within this forum, such as fathers, same-sex,
and/or working-class parents. Indeed, there is a pressing need to explore the experiences and
struggles of these groups, who have often been marginalised both in academic research and in
society more generally. I therefore suggest that constructions of parenthood in a wider range
of contexts, and by different groups of parents, are an important area for future research in
language, gender, and sexuality, as they may be able to raise awareness of the many ways in
which it is possible to be a parent, carer, or family in a contemporary context.
Research that explores the particular methodological challenges of conducting dis-
course-theoretical work in digital contexts is unfortunately still relatively scarce in the field
of language, gender, and sexuality. In the sample analysis offered above, the function of
digital resources (namely emoji) is considered as part of an analysis that focuses mainly
on linguistic features. This is appropriate for analysing digital interactions that primarily
rely on the written word. However, the analysis of language, gender, and parenthood online
may include the examination of media that are more reliant on other forms of expression,
such as images or videos. Zappavigna and Zhao’s (2017) analysis of ‘mommyblogging’, for
example, explores selfies posted to Instagram with the hashtags #motherhood and #mom-
life. Their article stops short of in-depth theorisation about what these selfies communicate
about discourses of motherhood, and Instagram users’ self-positioning as ‘mothers’, but
it does offer a new framework for categorising and analysing selfies as ‘metafunctional
resources’ for presenting an individual’s perspective on the world. Future explorations of
language, gender, and parenthood online will benefit from this kind of research. By paying
attention to these and other insights from the field of language and new media, analysts will
be well placed to conduct comprehensive analyses of the ways in which parents draw on
multimodal resources to navigate their own position in relation to wider discourses.

Further reading
Angermuller, J., Maingueneau, D., and Wodak, R. (eds.) (2014) The Discourse Studies Reader: Main
Currents in Theory and Analysis. Amsterdam and Philadelphia, PA: John Benjamins.
This edited collection offers a comprehensive overview of recent theoretical and methodological
developments in discourse studies. Its introduction provides clear and concise definitions of key terms,
concepts, and outlines the nature of the emerging discipline of discourse studies itself.
Leppänen, S., Westinen, E., and Kytölä, S. (eds.) (2017) Social Media Discourse, (Dis)identifications
and Diversities. New York, NY and London: Routledge.
Contributions to this collection are united by a linguistic focus on identity constructions in social
media contexts, including several contributions that explore the negotiation of gender roles online.
Poveda, D., Jociles, M. I., and Rivas, A. M. (2014) ‘Socialization into single-parent-by-choice family
life’. Journal of Sociolinguistic, 18(3), pp. 319–344.
This article contributes to a currently marginalised but emerging area of research that explores
alternative kinship structures from a sociolinguistic perspective.

419
Jai Mackenzie

Related topics
Digital ethnography in the study of language, gender, and sexuality; leadership language of Middle
Eastern women; doing gender and sexuality intersectionally in multimodal social media practices;
poststructuralist research on language, gender, and sexuality; gender and sexuality in discourse:
semiotic and multimodal approaches.

References
Althusser, L. (1971) ‘Ideology and ideological state apparatuses’. In: Althusser, L., ed. Lenin and
Philosophy and Other Essays. New York, NY and London: Monthly Review Press, pp. 127–186.
Baker, P. (2008) Sexed Texts: Language, Gender and Sexuality. London: Equinox.
Baker, P. (2014) Using Corpora to Analyze Gender. London, New Delhi, New York, NY, and Sydney:
Bloomsbury.
Bakhtin, M. M. (1981) The Dialogic Imagination: Four Essays, Holquist, M., trans. Austin, TX and
London: University of Texas Press.
Baxter, J. (2003) Positioning Gender in Discourse: A Feminist Methodology. Basingstoke: Palgrave
MacMillan.
Baxter, J. (2010) ‘Discourse-analytic approaches to text and talk’. In: Litosseliti, L., ed. Research
Methods in Linguistics. London and New York, NY: Continuum, pp. 117–137.
Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press: A Poststructuralist
Approach. Basingstoke: Palgrave MacMillan.
Bing, J. M. and Bergvall, V. L. (1996) ‘The question of binary questions: beyond binary thinking’.
In: Bing, J. M., Bergvall, V. L., and Freed, A. F., eds. Rethinking Language and Gender Research:
Theory Practice. London and New York, NY: Longman, pp. 1–30.
Cameron, D. (1996) ‘The language-gender interface: Challenging co-optation’. In: Bing, J. M.,
Bergvall, V. L., and Freed, A. F., eds. Rethinking Language and Gender Research: Theory Practice.
London and New York, NY: Longman, pp. 31–53.
Corwin, A. (2017) ‘Emerging genders: Semiotic agency and the performance of gender among
genderqueer individuals’. Gender and Language, 11(2), pp. 255–277.
Coupland, J. and Williams, A. (2002) ‘Conflicting discourses, shifting ideologies: Pharmaceutical,
“alternative” and feminist emancipatory texts on the menopause’. Discourse & Society, 13(4), pp.
419–445.
Davies, B. and Harré, R. (1990) ‘Positioning: The discursive production of selves’. Journal for the
Theory of Social Behaviour, 20(1), pp. 43–46.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Communities of practice: Where language, gender, and
power all live’. In: Locating Power: Proceedings of the 1992 Berkeley Women and Language
Conference. Berkeley, CA: Berkeley Women and Language Group, pp. 89–99.
Fairclough, N. (1992) Discourse and Social Change. Cambridge, MA: Polity Press.
Foucault, M. (1967) Madness and Civilization: A History of Insanity in the Age of Reason, Howard,
R., trans. London: Tavistock Publications.
Foucault, M. (1972) The Archaeology of Knowledge, Sheridan Smith, A., trans. London and New
York, NY: Routledge Classics.
Foucault, M. (1977) Discipline and Punish: The Birth of the Prison, Sheridan, A., trans. London:
Allen Lane.
Foucault, M. (1978) The History of Sexuality Vol. 1: An Introduction, Hurley, R., trans. London:
Penguin.
Gibson, W., Huang, P., and Yu, Q. (2018) ‘Emoji and communicative action: The semiotics, sequence
and gestural actions of ‘face covering hand’. Discourse, Context and Media. Available online at
https://www.sciencedirect.com/science/article/pii/S2211695818300801.

420
 Analysing gendered discourses online

Graham, L. J. (2005) ‘Discourse analysis and the critical use of Foucault’. In: Australian Association
for Research in Education, Sydney: Australian Association for Research in Education 2005 Annual
Conference, pp. 1–15.
Hays, S. (1996) The Cultural Contradictions of Motherhood. New Haven, CT and London: Yale
University Press.
Lawler, S. (2000) Mothering the Self: Mothers, Daughters, Subjects. London and New York, NY:
Routledge.
Lowe, P. (2016) Reproductive Health and Maternal Sacrifice: Women, Choice and Responsibility.
London: Palgrave Macmillan.
Mackenzie, J. (2017) ‘“Can we have a child exchange?” Constructing and subverting the “good
mother” through play in Mumsnet Talk’. Discourse & Society, 28(3), pp. 296–312.
Mackenzie, J. (2018) ‘“Good mums don’t, apparently, wear make-up”: Negotiating discourses of
gendered parenthood in Mumsnet Talk’. Gender and Language, 12(1), pp. 114–135.
Mackenzie, J. (2019) Language, Gender and Parenthood Online: Negotiating Motherhood in Mumsnet
Talk. London and New York, NY: Routledge.
Miller, T. (2011) ‘Falling back into gender? Men’s Narratives and practices around first-time
fatherhood’. Sociology, 45(6), pp. 1094–1109.
Mills, S. (2003) Michel Foucault. London and New York, NY: Routledge.
Mills, S. (2004) Discourse, 2nd ed. London and New York, NY: Routledge.
Mills, S. and Mullany, L. (2011) Language, Gender and Feminism: Theory, Methodology and Practice.
London and New York, NY: Routledge.
Reisigl, M. and Wodak, R. (2009) ‘The discourse-historical approach (DHA)’. In: Wodak, R. and
Meyer, M., eds. Methods of Critical Discourse Analysis. London, Thousand Oaks, CA, New Delhi,
and Singapore: SAGE, pp. 87–121.
Sunderland, J. (2000) ‘Baby entertainer, bumbling assistant and line manager: Discourses of fatherhood
in parentcraft texts’. Discourse & Society, 11(2), pp. 249–274.
Sunderland, J. (2004) Gendered Discourses. New York, NY: Palgrave Macmillan.
van Leeuwen, T. (2009) ‘Discourse as the recontextualization of social practice: A guide’. In Wodak,
R. and Meyer, M., eds. Methods of Critical Discourse Analysis. Los Angeles, CA, London, New
Delhi, Singapore, and Washington, DC: SAGE, pp. 144–161.
Wall, G. (2013) ‘“Putting family first”: Shifting discourses of motherhood and childhood in
representations of mothers’ employment and child care’. Women’s Studies International Forum,
40, pp. 162–171.
Weedon, C. (1997) Feminist Practice and Poststructuralist Theory, 2nd ed. Malden, Oxford, and
Victoria: Blackwell Publishing.
Zappavigna, M. and Zhao, S. (2017) ‘Selfies in “mommyblogging”: An emerging visual genre’.
Discourse, Context and Media, 20, pp. 239–247.

421
 28
Leadership language of Middle
Eastern women
Using feminist poststructuralist discourse
analysis to study women leaders in Bahrain

Haleema Al A’ali

Introduction
This chapter gives a comprehensive account of feminist poststructuralist discourse analy-
sis (FPDA) as a method of linguistic enquiry, and explains the reasons behind using FPDA
in my research of female leadership in Bahrain. It starts with an overview of the study
and the context, followed by a review of the basic principles of FPDA and how they were
applied to a selected case study of leadership language of a senior engineer in a Bahraini
company.
FPDA is a self-reflexive, multi-perspectival method of linguistic analysis with a decon-
structionist approach to discourse and gender, developed by Baxter (2003) in her study of
classroom discourse. It is based on the premise that speakers do not exist outside discourse,
and that ideas, concepts, identities, relationships, and so on are in constant flux and their
meaning is constantly changing; therefore, any FPDA analysis deconstructs a text through
offering in-depth multiple interpretations of it (Baxter 2003). With an eye on epistemologi-
cal enquiry and roots in the works of Bakhtin (1981), Derrida (1987), Foucault (1980), and
others (e.g. Walkerdine 1998; Weedon 1997), FPDA is concerned with studying intertextu-
alised discourses and the interplay between power, knowledge, subjectivity, and discourses
in any given context. Foucault defines discourses as ‘practices that systematically form the
objects of which they speak’ (1972: 49). Discourses are always interwoven and intertexu-
alised in the sense that every discourse can have traces of one or more different discourses.
In any given speech event, speakers are constantly shifting their subject positions between
powerfulness and powerlessness. Studying these critical moments in interactions and expos-
ing dominant discourses in the context, especially as they relate to gender, is the goal of any
FPDA research, and it is precisely the goal of my research on female leadership in Bahrain.
The case study in this chapter is of an exploratory nature and it examines the linguistic
practices senior women employ while ‘doing’ leadership in the context of corporate meet-
ings in one of the largest, most prominent corporations in Bahrain. In line with FPDA, I
adopt a social constructionist perspective which perceives workplace interactions as social

422
 Female leadership in the Middle East

practices in action (Holmes and Marra 2004). Within this framework, gender and leadership
are similarly regarded as social constructs. Butler’s (1990) view of gender as ‘performative’
is also central here where individuals are thought to ‘do’ or ‘perform’ being a ‘man’ or a
‘woman’ by displaying language and behaviour which conform to/resist the ideal or perfect
model perpetuated by the dominant ‘gendered discourses’ in a particular organisation or
community of practice. According to Sunderland (2004: 6), gendered discourses are ‘ways
of seeing the world’ through a gendered lens. Various dominant gendered discourses are
identified in the research literature. Drawing on Walsh (2001), Mullany (2007: 35) argues
that there are persistent ‘hegemonic discourses’ of masculinity and femininity that are
embedded in the discursive practices of any community. An example of such discourses is
the discourse of gender differentiation, a masculinist hegemonic discourse through which
differences between men and women in society are emphasised (Sunderland 2004).
Along the same line and based on the the social constructionist framework, leadership
is also perceived as ‘performed’ through language. When interacting with colleagues and
subordinates, leaders use language to perform, construct, and negotiate leadership. Holmes
(2006) posits that men and women leaders select from a repertoire of conventionally mas-
culine and conventionally feminine linguistic strategies to enact power and authority.
Conventionally masculine strategies (such as use of aggressive and competitive language,
controlling of the topic, interrupting, and issuing direct unmitigated orders, assertive state-
ments, bald on-record humour, and so on) correlate with a transactional style of leadership
which places more emphasis on solving problems and achieving work-related tasks. On the
other hand, conventionally feminine linguistic strategies (such as issuing indirect mitigated
orders, sharing power and authority, listening, giving compliments, using collaborative lan-
guage, and so on) correlates with a relational style of leadership which prioritises fostering
workplace relationships. Marra, Schnurr, and Holmes (2006) further note that effective lead-
ers, regardless of their biological sex, are linguistic experts who are skilled at deploying a
range of strategies to achieve the various goals of leadership. In this research I use the term
‘practices’ rather than ‘strategies’ or ‘styles’ because it is more compatible with the post-
structuralist view of language as a social practice.
Feminist research in the West has shown wide evidence of the ‘double bind’ (see also
Appleby, this volume) where women leaders are viewed as incompetent and are condemned
for their choice of either type of linguistic practices: they are deemed ‘unfeminine’ if they
use traditionally masculine language, or ‘unprofessional’ if they use traditionally feminine
language (Alvesson and Billing 1997; Brewis 2001). I believe that the validity of these find-
ings should be questioned when applied to the Arab Middle Eastern context. Middle Eastern
workplaces are significantly different from Western models. Weir (2003: 10) argues that
‘the very texture and processes of management in this region remained different from their
Western models’. Therefore, the case study in this chapter will contribute to the growing
body of research that focuses on the regional context.
What’s more, Middle Eastern women’s experience in the public sphere is relatively new
compared to their Western counterparts, hence there is a noticeable lack in role models for
women aspiring to reach higher management and leadership positions. The case study offers
models of good practice in Middle Eastern women’s leadership, by acquiring an insider’s
knowledge of context and a deeper understanding of the women leaders’ backgrounds,
intentions, agendas, and the linguistic practices they use to achieve their goals. For this,
FPDA, alongside other qualitative methods, is utilised to give a multidimensional and multi-
perspectival analysis of the data.

423
Haleema Al A’ali

Backround and methodology of the study

Bahrain is an Arab Middle Eastern country and a leading example of women’s empowerment in
the region (Gharaibeh 2011; World Economic Forum 2017). The country has been experienc-
ing growth and diversification in the economy and some shifts in the political scenery. These
changes, accompanied with the government’s grand plan towards the inclusion of women in
the public sphere of politics and the workplace, have resulted in a shift in society in relation
to gender ideologies and gender roles (Metcalfe 2007). The gender system in the region is pri-
marily shaped by Islamic principles and long-standing Arab cultural values. Traditional Arab
societies are patriarchal; gender and age are determinant factors for the amount of power an
individual enjoys in the community; the most powerful members are men and elders (Sabbagh
2005). The traditional gender system in Islam is grounded on the principle of biological dif-
ference between men and women. According to this principle, men and women should be
assigned different but complementary roles, rights, and responsibilities in society (Metcalfe
2011; Muhdina 2017). In traditional Arab-Islamic societies, Islamic and Arabic values perme-
ate all aspects of the private and public spheres; women are typically associated with the pri-
vate sphere and men with the public (El-Rahmony 2002). This is reflected in the segregation
between men and women’s jobs in the workplace; until recently, female employment in the
region had been clustered in traditionally ‘feminine’ jobs such as teaching, nursing, and other
care-taking professions, which were perceived to be in alignment with their ‘maternal nature’
(Metcalfe 2007). Metcalfe (2011: 133) further argues that while governments are making tre-
mendous efforts to promote women in the public spheres of politics and the workplace, it is
still in the context of the ‘Islamic gender regime’. Equally, management and gender studies
in the Middle East (e.g. Ozbilgin and Healy 2003) reveal that even when women ‘shatter the
glass ceiling’ and become leaders and managers, they are often ‘constituted along patriarchal
lines with women’s role as Mother emphasised’ (Metcalfe 2007: 58).
Nowadays Bahraini women’s work is no longer restricted to certain fields nor is it viewed
as marked or unnecessary; on the contrary, women are encouraged to enter the workforce
and are given the opportunities to excel and reach leadership positions. To empower Bahraini
women in the workplace, the government (represented by the Supreme Council of Women
and led by Princess Sabeeka bint Ibrahim Al Khalifa, the King’s wife) has been support-
ing and monitoring women’s progress in all public and private companies. Most recently,
numerous Arab and Bahraini business women have been featured in Forbes magazines as
top influencers in the region and in the world. According to Forbes Middle East (2017),
women in the Arab world are finally ‘breaking the glass ceiling’, especially in the banking
sector. Forbes’ list for the most powerful Arab women includes 15 Bahraini women who
made significant contributions locally and internationally in different fields, mainly poli-
tics, business, and management; they are ministers, business leaders, and influencers of all
types. With the recent dramatic changes in the region, I consider that there is a dire need for
research that explores in depth the repertoire of leadership language practices available for
senior women in the region, and also to present examples of effective leadership that may
differ from the western models.

Feminist poststructuralist discourse analysis (FPDA)

FPDA is primarily interested in studying the interplay of discourses and exposing power
relations in any given context. It doesn’t pursue a political agenda or a theoretical mission.

424
 Female leadership in the Middle East

Rather, it supports social transformations in small-scale, localised contexts (Baxter and Al

A’ali 2016) Moreover, FPDA views gender as a ‘site of struggle’, therefore, FPDA research
always has a specific feminist focus. It supposes that women can be both momentarily pow-
erful and controlling or powerless and marginalised in the same interaction or event. Herein
lies the goal of my research. My study is not seeking emancipation for women in the work-
place nor is it interested in solely studying interactional patterns in conversations. Rather
it seeks to reveal the diversity and complexity of women’s positioning in Middle Eastern
workplaces. Typically, Middle Eastern women are viewed and portrayed as victims, power-
less and marginalised in Western research and media (Abu-Lughod 2002; Qutub 2013). As
a linguist and as a Middle Eastern woman, I am aware of this narrative, yet I also realise that
this is a rather naive and simplified view of a widely diversified and constantly changing
demographic. I opted for FPDA in order to contest this view and shed light on the complex
and competing aspects unique to the Middle Eastern context. Along the same lines, rather
than assuming that women are positioned powerlessly within patriarchal discourses, I intend
to show that women in my study can indeed be more powerful than their male counter-
parts and that there are many ways in which positions of power can be achieved, however
momentary.
FPDA analysis is executed on micro and macro levels, and it has many other dimensions.
These are presented next.

FPDA as a method of analysis

Any textual analysis using FPDA should take into account the following elements, dimen-
sions, and processes (Baxter 2003):

• The synchronic-diachronic dimensions:

FPDA analysis requires a synchronic ‘detailed micro-analysis of stretches of text’
(Baxter 2008: 251), which involves identifying critical moments in conversations where
a power shift may occur. This is quite different from the diachronic dimension which
usually requires the use of ethnographic methods to observe and study the change in the
linguistic practices of members of a certain community over a longer period of time.
Owing to the limited access I was granted to the company, I was unfortunately unable
to achieve the diachronic dimension in this study.
• Micro–macro levels of analysis (denotative–connotative analysis):
Analysis using FPDA is executed on two levels, micro and macro. Starting at the
micro-linguistic level, the researcher identifies the significant moments in the interac-
tion where speakers shift subject positions. This is based on the premise that in any
speech event, speakers constantly shift their positioning between powerfulness and
powerlessness. These positions are made available by the competing discourses in the
context. Therefore, at the macro-discoursal level, the researcher looks for evidence of
the prevailing, competing discourses and explains the shifts in power through examin-
ing the interplay between these discourses.
Therefore, during the process of data analysis, the researcher utilises tools from dis-
course analysis methods (for instance, conversational analysis or interactional linguis-
tics) necessary to conduct a descriptive non-evaluative micro-analysis on the data; this
is referred to as the denotative analysis, which also forms the basis for the connotative
macro-discursive reading of the text. A number of important processes take place in the

425
Haleema Al A’ali

connotative analysis such as identifying key discourses, analysing speakers’ position-

ing, and examining the process of negotiating power and the interplay between the
competing and intertextualised discourses in the context.
• Intertextuality and inter-discursivity:
FPDA considers that discourses are always interwoven, and analyses the ways in
which every discourse contains traces of other discourses. For instance, a discourse of
gender differentiation can also be interwoven with and work alongside other discourses
to enhance the positioning of a certain individual in an interaction, or it can be contested
or undermined by another discourse in the context (see analysis).
• Self-reflexivity:
One of the most crucial aspects of the use of FPDA is researchers’ commitment to
being self-reflexive, constantly revaluating their position and questioning their own
values and assumptions and how these may affect the process of analysis.

In the following section I present the case study along with example extracts from the meet-
ing and interview data and their FPDA analysis.

Hanan’s case study

The study is set in a large company in Bahrain (Bahrainco, a pseudonym), known for its tra-
ditionally masculine environment due to its history and the nature of its business. The par-
ticipant of the study is a woman who had recently been promoted to a managerial position
after climbing the professional ladder for over 20 years. Hanan is a senior support engineer
in the engineering division who has spent her entire career experience in Bahrainco. Despite
the traditionally masculinised nature of the department where men are exceedingly higher
in number and career opportunities, Hanan has managed to receive recognition for her hard
work and has been entrusted to lead important and critical national projects.
The study primarily sought to examine the leadership language practices that Bahraini
senior women use with colleagues and subordinates within the context of corporate meet-
ings in addition to the significant interacting discourses at play in the context and how they
shape the leadership and language practices of the senior women.
For the purpose of data collection, multiple qualitative methods were utilised: attending
and recording corporate meetings, interviewing Hanan and some of her colleagues and sub-
ordinates who attended the meetings, and shadowing her outside the meetings, observing
her, taking notes, and listening to the small talk and casual conversations in and outside the
office (for instance in the cafeteria or in the parking lot).
Around the time of the data collection, Hanan was leading a team of engineers (all males)
in a nationwide project of a critical nature. The circumstances of the meeting were peculiar;
the language in the meeting was full of jargon which presented a challenge to my analysis.
Besides this, I accompanied her to the meeting which was scheduled at noon in a cabin
situated in a gas field in the middle of the desert. Upon entering the cabin, I felt awkward
and self-conscious because we were the only women in the cabin. Hanan’s body language
showed confidence. She greeted everyone with a smile and even a little bit of banter; and
she kept the small-talk short. Everyone seemed to be engaged despite her characteristically
low-pitched voice. In a later interview with Hanan, she informed me that it took her a while
and a great deal of courage to overcome this feeling of intimidation to always be the only
woman in the meetings and field sites. Even more, it took a great amount of effort for her to
be taken seriously.

426
 Female leadership in the Middle East

This is one of a series of meetings Hanan has had with her team members for the pur-
pose of following up implementation and work progress, and solving any arising problems
along the way. She heads a team of three engineers, none of whom is her direct subordinate:
Ameer is a senior engineer in Bahrainco; he is around ten years younger than Hanan, hence
less experienced. Raj and Vivek are junior engineers who work at an Indian contracting
company; they are in Bahrain for a short period of time to work on this particular project for
Bahrainco. During the course of the meeting, Hanan and her team members engage in the
joint construction of talk, where Hanan goes through a check-list of action plan and Amir,
Raj, and Vivek cooperate and alternate in answering her inquiries and updating her with the
work progress.

Denotative analysis
Hanan starts the meeting unofficially during the process of seating by explaining the agenda
of the meeting. In the first extract, the team are discussing the implementation of a new
system (ATG1), most specifically, the message that should be displayed to operators. Hanan
thinks that providing the term (ATG1) only is likely to stir some confusion among the opera-
tors because they are not familiar with the system. Transcription conventions are provided
at the end of the chapter.

Extract 1: ‘Even I will forget’

(H=Hanan; Chair; A=Amir; engineer, male; R = Raj; contractor engineer, male; V=

Vivek; contractor engineer, male; ATG=operation system; K40=radar detector; FAC=
field advanced controllers)

1
2 R: yeah basically ATG1 (.) yeah the only controllers come in
3 particular ATG only (.) so (.) what we are describing here is (.)
4 it is ATG1 and the corresponding loop controller (.) that is say
5 (.) ATG1 K40(.) ATG1[(-)
6 H: [(-) you mean this message will be displayed to
7 operators↑
8 R: yes
9 H: you think operators will understand ATG1↑
10 A: it is confusing
11 V: that’s ok [we will
12 H: [but it is er in a way (.) it’s it’s good as a
13 maintenance er (.) when the maintenance guy come (.) he will interpret
14 it he will say yeah this is coming from ATG1 (.) so yes maybe the
15 message is not (.) cannot be fully interpreted by the operator (.)
16 eventually I think (.) we need the word ATG (.) I guess in the er
17 message (.) we need it=
18 A: = we need it
19 R: we get it↑
20 A: yes in case that you say that all the the controllers are off then we
21 will display a message [say that ATG

427
Haleema Al A’ali

22 V: [no we will (--) we will educate the operator

23 that (.) er it is in the model (.) already it has been (-) under the
24 FAC FAC the controller (.) you have to see that alarm (.) based on that
25 controllers (.) yeah otherwise we will educate the er operators
26 H: emm
27 (After 55 lines of a discussion between Amir, Raj and Vivek about the best
28 way to implement the new system without confusing the operators)
29 H: well er (0.3) (looking at some papers) ATG1 because I am a system
30 person (.) I understand [ATG means something to me=
31 R: [yeah =meaningful
32 H: meaningful (.) but as long as we will get an alarm (3)
33 R: but we can educate them (.) it’s very
34 H: you forget (.) even me after a while I will forget (.) ATG1 (.) it is
35 connected to controllers or to BMS (.) I will forget (.) I [will forget
36 A: [you have to
37 go back to the drawings=
38 R: =yeah exactly (.) so it’s better to have ATG1
39 and AC00

The extract begins with Raj explaining an earlier suggestion he made about providing a
simple description of the new ATG system to the operators. Hanan seems surprised and
asks for further clarification using a string of statements with a rising intonation: ‘you think
operator will understand ATG1 ↑’ (line 9), which indicates her initial disapproval of the idea.
Amir takes Hanan’s side and issues a criticism using a negative evaluative adjective in line
10: ‘It’s confusing’. Just when Vivek takes the floor to express his compliance to act upon
Amir’s criticism – ‘that’s ok we will’ (line 11) – Hanan interrupts him, this time with a rather
mitigated monologue in which she weighs the pros and cons of Raj’s suggestion and builds
up her argument, stating at first that ‘it’s it’s good as a maintenance’ (lines 12–13) followed
by more explanation of why it is a favourable idea, and that despite her earlier concerns, she
is considering this suggestion. To wrap up her argument, she uses multiple hedging devices:
’maybe’, ‘I think’, ‘I guess’ to show that she understands Raj’s point of view and partially
agrees with him. Finally, she uses the inclusive pronoun ‘we’ and deontic modal verb ‘need’
to draw on a conclusive decision – ‘we need the word ATG’ (line 16); her intention is per-
haps to strengthen the sense of collective work and shared responsibility. Amir immediately
seconds Hanan’s decision by echoing her statement ‘we need it’. This time, Raj shows his
direct compliance and readiness to follow Hanan’s orders by issuing a question: ‘we get it↑’
(line 19). Here, Amir takes the floor and responds to Raj with an affirmative ‘yes’ followed
by an order to carry on the implementation process. Vivek contributes with more explana-
tion of the particulars of the operation from his point of view (lines 22–25). During this
exchange, Hanan stays silent except for the minimal response ‘hmm’ to indicate that she has
been attentively following the conversation.
The discussion of the implementation of the process between Amir, Raj, and Vivek takes
a few more minutes, and Hanan, who has been attentive the whole time, finally takes the
floor using the discourse marker ‘well’ to establish her turn (line 29). She examines the
papers in her hand for few seconds, perhaps to gather her thoughts or to prepare the team
for her counter argument. She revisits her earlier concern over the ATG message being
incomprehensible to the operators, this time by using a more personal approach and refer-
ring to her experience and expertise in ‘systems’; in fact, she defines herself as ‘a system

428
 Female leadership in the Middle East

person’ (lines 29–30), identifying with a community of engineers and disassociating herself
from operators who have less expert knowledge: ‘I am a system person I understand ATG
means something to me’ (line 30). Raj uses the minimal response ‘yeah’ and the paraphrase
‘meaningful’ to show his total agreement with Hanan, who then echoes him with the word
‘meaningful’, and continues her turn (line 32). When she stops, Raj allows a few seconds
to pass by before he makes further suggestions, also using the inclusive ‘we’ to stress the
joint endeavour: ‘we can educate them’ (line 33). Before he carries on with his proposition,
Hanan interrupts and disagrees with the words: ‘you forget’, and ‘even me after a while I
will forget’. Here, Hanan seems to be generalising about people’s abilities to remember dif-
ficult acronyms. She is also revealing her own weakness here in order to soften the force of
her disagreement with Raj. She clarifies her point further and finally repeats ‘I will forget’
twice for the purpose of emphasis (line 35). Amir uses a cooperative overlap to build on her
argument, adding more reasons why he is not in favour of Raj’s suggestion. Faced with such
opposition, Raj immediately shows his total compliance – ‘yeah exactly’ – and modifies his
proposition.
The next extract takes place towards the end of the meeting; before wrapping up, Hanan
stresses the importance of meeting the deadlines and working within the specific time frame;
she criticises Raj and Vivek for lacking time management strategies.

Extract 2: ‘Your visa is valid until 29’

(H=Hanan; Chair; A=Amir; engineer, male; R = Raj; contractor engineer, male; V=

Vivek; contractor engineer, male; HMI= Human machine interface)

40 A: you have to do the description (.) for the controller and pop ups
41 R: description (.) it’s both er finished up already
42 H: everything will be↑
43 R: no no the er (.) ok (.) this er (.) no this I know I will complete it
44 H: yeah but (.)£ tell me when (I mean) £
45 (everybody is laughing)
46 H: £what’s the time now£↑
47 (more laughter)
48 H: I don’t want you to die [hehehehe
49 V: [£(---)£
50 R: [£ if this guy er this guy (says) today means
51 till tomorrow 12 till tomorrow morning£
52 H: tomorrow (-) till tomorrow morning 6 am (.) and you will come tomorrow
53 morning (.)
54 V: HMI I can er work on (.) job design I can work on
55 R: no actually (.) once we complete this er dryer testing and the fixing
56 of the small er that (.) HMI things (---) and myself and Amir (.) we’re
57 concentrating on the 39 and those communication and testing so [I thi-
58 A: [we
59 don’t need the N44 for testing IL (.) two days↑
60 R: in fact (.) he was asking me if I will do the er 39 communication (-)
61 but I told him you have (.) any work here [so
62 H: [£you ha- there are other

429
Haleema Al A’ali

63 work [I haven’t spotted (.) your visa is valid until 29

64 A: [hehehe
65 V: yeah 29
66 A: today is 22
67 V: some er
68 H: you have how many hours until 29↑
69 (Laughter from all)

The extract commences with Amir issuing a directive to Raj and Vivek about some implemen-
tation detail; for this purpose he uses a deontic modal expression ‘you have to do the descrip-
tion’ (line 40). Raj echoes Amir ‘description’ and states, with some hesitation (‘er’), that it
was finished. Hanan senses his reluctance and issues a checking statement – ‘everything will
be↑’ (line 42) – to which Raj, taken aback by her intuitive remark, responds with a number of
hedges (‘no no the er (.) ok (.) this er’) and a promise to complete the task. Hanan doesn’t seem
to be fully convinced of the sincerity of Raj’s promise and continues to put him ‘on the spot’ by
issuing a series of exaggerated questions, mitigated by a wide smile on her face: ‘yeah but (.)
tell me when’ (line 44), and ‘what’s the time now’ (line 46). With everyone laughing (either out
of embarrassment or because they are amused), the banter carries on as Hanan ironically states
that she doesn’t want Raj to die out of hard work, apparently to imply that he is in fact a lazy
procrastinator. Raj instantly tries to deflect the blame by referring to Vivek’s lack of time man-
agement strategies (lines 50–51), yet Hanan ignores his defence and keeps the exaggerated
questioning going (‘tomorrow (-) till tomorrow morning 6 am (.) and you will come tomor-
row morning’ (lines 52–53)). While Vivek responds to Hanan’s questioning with a number of
promises about finishing some tasks (line 54), Raj’s strategy is less direct as he positions Amir
as a partner or an accomplice when he mentions other tasks that they are working on together:
‘and myself and Amir (.) we’re concentrating on the 39 and those communication and testing
so’ (line 57). Amir interrupts him to show his instant resistance to this positioning and chal-
lenges him in lines 58-59 ‘we don’t need the N44 for testing IL (.) two days↑’.
As Raj continues his attempts to deflect the blame from himself, he accidently reveals
other unfinished tasks that were not known to Hanan and Amir. Hanan’s first response is to
smile and interrupt his turn to show her surprise at this new revelation ‘£you ha- there are
other work I haven’t spotted’ (lines 62–63). Hanan starts a banter sequence about Raj and
Vivek’s visa expiration date – ‘your visa is valid until 29’, ‘you have how many hours until
29↑’ (lines 63 and 68 respectively). Amir cooperates with Hanan to co-construct the banter
by laughing throughout the sequence and issuing a supportive comment, ‘today is 22’ (line
66). All participants in the meeting laugh at Hanan’s humorous remarks even though they
clearly realise that it is embedded with criticism.
The next extract is taken from the interview data. Hanan talks about gender equality in
Bahrainco.

Extract 3: ‘This is your culture here’

70 H I mean up to (.) last year for example (.) they wouldn’t really recruit
71 er females for engineers they don’t trust them as er engineers (.)
72 especially in engineering er (.) so many graduates (.) from (.) Bahrain
73 University from outside univ- sure they apply (.) I mean and they are
74 very good and distinguished (.) but you hardly see I mean them recruit
75 (.) anyone (---) we have I mean myself I (.) initially I didn’t think

430
 Female leadership in the Middle East

76 (.) of that (0.1) and then er (.) believe it or not I mean one of the
77 Westerners he brought that up to me (.) he said I mean err (.) he spoke
78 to all those Bahrainis who are higher than him (0.1) and he said er
79 (0.1) they don’t er because you you are a female (.) they don’t (0.1) I
80 mean they put some (0.1) (cap) on your er (0.1) advancement (.) this is
81 they say this is your culture here (.) °and errr° I cannot change it

Hanan explains that she believes the reason behind the lack of women in the engineering
department lies within the discriminatory recruitment process. She starts by giving examples
of actual events which are indicative of such discriminatory practices: ‘last year for example
(.) they wouldn’t…’ (line 70). As she narrates a recent incident where professional women
engineers were denied the chance to compete for jobs in Bahrainco, she refers to the man-
agement and decision-makers in the company vaguely as ‘they’. This is perhaps to hedge
her further accusations of Bahrainco’s decision-makers as sexist: ‘they don’t trust them as er
engineers’ (line 71). Hanan emphasises that such discrimination is especially practiced in the
engineering department and gives more examples to confirm her claim that the engineering
profession is being gendered and as a result, women are being excluded (lines 72–75).
In line 75, Hanan shifts to the inclusive pronoun ‘we’ then the singular pronoun ’I’ to
refer to her personal experience with the ‘glass ceiling’ as a woman engineer in the male-
dominated company of Bahrainco. She claims that she was oblivious to the actual reasons
behind deferring her promotions despite her expertise and her long years of experience, and
that her superior, a Western expat is the one who brought it to her attention (lines 77–81).
According to him, despite his efforts, it is the wider Bahraini patriarchal culture that is
behind the gendered inequality and ‘glass ceiling’ in the company.
Despite the existence of the glass ceiling, Hanan is still considered an expert and is
trusted to lead national wide projects. In an interview with Amir, he repeatedly refers to
Hanan’s expertise and length of experience as the reason behind her seniority. In the next
extract, he is asked about her leadership style.

Extract 4: ‘She has experience more than me’

82 A and that’s natural (.) she has experience more than me (.) for example
83 I do it some- in a way (.) she says ok (.) you have done it in a right
84 way but it’s better to do it like this (0.2) she’s she’s a senior
85 engineer I am an engineer so (.) I I was working on a differen- not
86 different department I was looking after a different system (.) I
87 recently joined this err (.) supporting this system for the past two
88 three months so (.) I am little bit new in this field (.) err she was
89 looking after this for for I donno (.) few years

Amir refers to any disagreement between himself and Hanan as ‘natural’, given the fact that
she is more experienced than he is (line 82). The specific example he gives does not indicate
any type of argument though, just a reference to the effect of his lack of experience in some
aspects and the indirect, face-saving way she deals with such mistakes. Then he immedi-
ately shifts to focus on Hanan’s and his different areas of expertise: ‘I was working on a
different department I was looking after a different system’ (lines 85–86). He emphasises
his recent involvement in the project by using the adverbials of time ‘recently’ and ‘three
months ago’ (lines 87 and 88 respectively).

431
Haleema Al A’ali

According to the principles of FPDA, the denotative analysis of the data is the basis for
the macro-discoursal analysis; in the next section this will be reviewed.

Connotative analysis
There are many discourses at play in the context. The first dominant discourse that I have
identified is the discourse of masculinisation, which according to Baxter (2003), normal-
ises conventionally masculine leadership practices such as aggressiveness, competitiveness,
assertiveness, boldness, dominance, and so on. This discourse is manifested in a number of
ways. As analysed above, Hanan often resorts to such linguistic strategies to enact power
and authority and to get the job done even if it is at the expense of her team members’ ‘face
needs’ (Brown and Levinson 1987). Hanan and Amir’s co-construction of the on-record
banter in Extract 2 is an exemplar of the normalcy of this behaviour. The extracts provided
in this review and many other parts of the meeting and interviews indicate the critical and
task-oriented nature of work in the engineering department, which, according to Hanan,
requires her and her team members to be fairly formal and direct with each other, and most
of all, prioritise task accomplishment over other relational goals. Baxter (2010) notes that in
such male-dominated workplaces, display of traditionally feminine language and practices
may be viewed as a weakness in character and even a waste of valuable company time.
Accordingly, Hanan, may feel pressurised to acquire the mainstream traditionally masculine
language in order to be recognised and appreciated.
While Hanan is potentially disempowered by the gendered discourse of ‘masculinisation’
which, according to the interview data, may have worked against her work progression,
there are a number of competing discourses which can be detected in the context that work
to empower her such as the discourses of ‘historical legacy’ and ‘expertise’.
According to Baxter (2003), historical legacy favours employees with seniority who
have worked in the company for a considerable amount of time and have built long-standing
relationships and connections. Hanan is one of the employees who have spent their entire
career in the same department; over 20 years in Bahrainco has awarded her with a sense
of familiarity with the company and its practices and has granted her a certain amount of
power and privilege. This is evident in the interview with Amir, her male subordinate, where
he referred multiple times to her years of experience (Extract 4). It can further be indexed
through the length of time Hanan spent in the company and also the amount of confidence
she carries as she walks in the all-male cabin and how she greets everyone with a sense
of familiarity. It is also reflected linguistically through her use of strategies with her team
members which index confidence and self-assurance such as direct unmitigated language
(e.g. ‘you have to do the description’ (line 40)) and contested humour (e.g. ‘yeah but (.) tell
me when’ (line 44), and ‘what’s the time now’ (line 46)).
Another discourse which is manifested throughout the data is the discourse of expertise.
Working and progressing through the ranks in the same department for over 20 years has
given Hanan a great amount of experience and expertise, therefore, she is entrusted to lead
nationwide projects despite her middle-management status. In the meeting, Hanan’s exper-
tise is evident; when she speaks about processes, her team members listen attentively (lines
11–16); also in Extract 2, she refers to herself as ‘a system person’ who knows all about
specialised technical concepts and processes such as ATG1. In Extract 4, Amir admits that
Hanan has more expertise than he does – ‘I am little bit new in this field (.) err she was look-
ing after this for for I donno (.) few years’ (lines 88–89).
The dominant corporate discourses detected in the context compete at times and inter-
act at others to variably position Hanan between powerfulness and powerlessness. The

432
 Female leadership in the Middle East

discourse of masculinisation dominates the engineering department, overpowering the other

discourses most of the time. This is evident in the all-male environment and other dis-
criminatory practices against Hanan and other female employees as shown in Extract 3 (for
example, despite her expertise and her long working experience, she is yet to be promoted
to a manager). However, in the meeting itself, Hanan is shown to use conventionally mas-
culine language (e.g. direct unmitigated orders, banter, etc.) especially in critical moments
such as the possibility of missing deadlines or delay in accomplishing tasks, perhaps to
lessen the limitation imposed by the discourse of masculinisation on her as a female.
On the other hand, while the discourse of masculinisation seems to have hindered
Hanan’s career progression, it is her long experience and expertise that distinguish her from
her male counterparts and position her powerfully. When discourses of historical legacy and
expertise interact, the discourse of masculinisation is undermined and Hanan is entrusted to
handle critical projects, and she feels confident enough to do so using linguistic strategies
that index power and authority. There are critical moments in the meeting where Hanan is
positioned powerfully by the discourse of expertise; she refers to herself as a ‘system per-
son’, uses technical jargon throughout the meeting, and shows her apparent expertise as she
evaluates and assesses her subordinates’ work.

Conclusion
Through this research study I hope to present examples of good practice in a context where
women are new to power and authority. With the recent trend in the Middle East of women
entering the workforce and occupying jobs that are traditionally associated with men,
women in the region, and particularly in Bahrain, are in need of role models. Therefore, it
is necessary to give voice to senior women who have achieved a certain degree of success
and progressed to senior roles in workplaces that have a prevailing traditionally masculine
culture. FPDA seemed an obvious choice because it aims to bring to the surface the silenced
voices, expose the hidden aspects of the context, and bring about social transformations.
Women in the Middle East are still fighting a battle their Western counterparts fought dec-
ades ago. They are still struggling to prove that they are eligible to work alongside men in
certain professions. In this case study, women are being overshadowed in the engineering
department, as Hanan puts it ‘because you you are a female (.) they don’t (0.1) I mean they
put some (0.1) (cap) on your er (0.1) advancement’. Hers is the minority voice which needs
to be heard so that change can take place and the situation can be corrected in a way that is
congruent with the social and cultural peculiarities of the society.

Future directions
FPDA’s reflexive approach to data enhances the quality of today’s research and should be
considered as a main method when researching language and gender. It can be used along-
side other methods at the micro and macro levels; in fact, it transcends the polarity of micro
and macro analysis and focuses on providing a thick description of the data that would
ultimately enhances the understanding of the context. Hence, I believe FPDA should be
utilised in a wider range of geographical contexts in order to produce fine-grained analysis
of unique localised settings where generalisations are not applicable and change is essential.
While this study tackled leadership and women in Bahrain, research is needed to explore
Middle Eastern women of other countries in the region, given the vast economic, social, and
ideological differences between the countries under the umbrella of the Middle East.

433
Haleema Al A’ali

Transcription conventions
This transcription key is mainly based on the “Jefferson system”1

1, 2 Line numbering
A, B, C Name of speaker (anonymised and abbreviated)
Word Translated talk
((word)) Transcriber’s comment on what happened
(word) Transcriber’s guess at what have been said
() Unclear talk
(-) Omitted talk
(.) Noticeable pause
(0.2), (2.5) Example of timed pauses
↑word, word↓ Rising and falling of intonation
WORD High volume, loud
ºwordº Low volume, attenuated speech
[word
[word Overlapping talk
=word
=word Latching, simultaneous talk
wor- Sharp cut-off
Wo:rd Prolonged sound
£word£ Smiley voice, humorous tone

Note
1 Jefferson, G. (2004) ‘Glossary of transcript symbols with an introduction’, in Lerner, G. (ed)
Conversation analysis: studies from the first generation. Amsterdam and Philadelphia: John
Benjamins, pp. 13–31.

Further reading
Baxter, J. and Al A’ali, H. (2016) Speaking as Women Leaders: Meetings in Middle Eastern and
Western Contexts. Basingstoke, UK: Palgrave McMillan.
This book examines the leadership language of six senior business women from the UK and
Bahrain; it challenges the preconceived notions about the discrepancies in the challenges faced by
women in the West and the Middle East. The use of feminist poststructuralist discourse analysis
(FPDA) has revealed that senior women in both cultural contexts are constrained, positioned, and
influenced by similar discourses such as ‘masculinisation’ and ‘hierarchy and status’.
Mullany, L. (2007) Gendered Discourse in the Professional Workplace. Basingstoke: Palgrave
McMillan.
This book uses sociolinguistic data to analyse workplace interactions taking place in two companies,
and explores the underlying gendered discourses at play in the context and their role in perpetuating
discriminatory practices.
Metcalfe, B. (2011) ‘Women, empowerment and development in Arab Gulf States: A critical appraisal
of governance, culture and national human resource development (HRD) frameworks’. Human
Resource Development International, 14(2), pp. 131–148.
The article addresses gender issues and human resource development (HRD) in three Arab Gulf
States (Bahrain, UAE, and Saudia Arabia), particularly the social and cultural factors that shape
women in the region and their livelihood. It assesses the current HRD frameworks and calls for

434
 Female leadership in the Middle East

tailored and localised strategies based on a better understanding of the social and cultural context in
order to support and empower women and help them develop their own version of Islamic feminism.

Related topics
Poststructuralist research on language, gender, and sexuality; feminist poststructuralism – discourse,
subjectivity, the body, and power; identity construction in gendered workplaces; leadership and
humour at work: using interactional sociolinguistics to explore the role of gender; language, gender,
and the discursive production of women as leaders.

References
Abu-Lughod, L. (2002) ‘Do Muslim women really need saving? Anthropological reflections on
cultural relativism and its others’. American Anthropologist, 104, pp. 783–790.
Alvesson, M. and Billing, Y. (1997) Understanding Gender and Organisations. London: SAGE.
Bakhtin, M. M. (1981) The Dialogic Imagination: Four Essays, edited by Holquist, M. and translated
by Emerson, C. and Holquist, M. Austin and London: University of Texas Press.
Baxter, J. (2003) Positioning Gender in Discourse: A Feminist Methodology. Basingstoke: Palgrave.
Baxter, J. (2008) ‘Feminist Post-structuralist discourse analysis: a new theoretical and methodological
approach?’. In: Harrington, K., Litosseliti, L., Sauntson, H., and Sunderland, J., eds. Gender and
Language Research Methodologies. Basingstoke: Palgrave Macmillan, pp. 243–255.
Baxter, J. (2010) The Language of Female Leadership. Basingstoke: Palgrave Macmillan.
Baxter, J. and Al A’ali, H. (2016) Speaking as Women Leaders: Meetings in Middle Eastern and
Western Contexts. Basingstoke, UK: Palgrave McMillan.
Brewis, J. (2001) ‘Telling it like it is? Gender, language and organizational theory’. In: Linstead, R.
W., ed. The Language of Organization. London: SAGE, pp. 283–309.
Brown, P. and Levinson, S. (1987) Politeness: Some Universals in Language Use. Cambridge:
Cambridge University Press.
Butler, J. (1990) Gender Trouble: Feminism and Subversion of Identity. New York, NY: Routledge.
Derrida, J. (1987) A Derrida Reader: Between the Blinds. Brighton: Harvester Wheatsheaf.
El-Rahmony, S. (2002) ‘Women in the Arab world: From role conflict to effective participation’.
Al-Mustaqbal Al-Arabi (Arab Future), pp. 93–107.
Foucault, M. (1972) The Archaeology of Knowledge and the Discourse on Language. New York, NY:
Pantheon.
Foucault, M. (1980) Power/Knowledge. Brighton: Harvester Press.
Forbes Middle East (2017) The Top 100 Most Powerful Arab Businesswomen 2017. Available at: http://
www.forbesmiddleeast.com/en/list/top-100-powerful-arab-businesswomen-2017/ (Accessed: 23rd
July 2017).
Gharaibeh, F. A. (2011) ‘Women’s empowerment in Bahrain’. Journal of International Women’s
Studies, 12(3), pp. 96–113.
Holmes, J. (2006) Gendered Talk at Work. Oxford: Blackwell.
Holmes, J. and Marra, M. (2004) ‘Relational practice in the workplace: Women’s talk or gendered
discourse?’. Language in Society, 33, pp. 377–398.
Marra, M., Schnurr, S., and Holmes, J. (2006) ‘Effective leadership in New Zealand: Balancing gender
and role’. In: Baxter, J., ed. Speaking Out: The Female Voice in Public Contexts. Basingstoke:
Palgrave, pp. 240–260.
Metcalfe, B. (2007) ‘Gender and human resource management in the Middle East’. International
Journal of Human Resource Management, 18(1), pp. 54–74.
Metcalfe, B. (2011) ‘Women, empowerment and development in Arab Gulf States: A critical appraisal
of governance, culture and national human resource development (HRD) frameworks’. Human
Resource Development International, 14(2), pp. 131–148.

435
Haleema Al A’ali

Muhdina, D. (2017) ‘Gender equality perspective in Islam based on the Holy Quran’. Social Sciences,
12(12), pp. 2314–2320.
Mullany, L. (2007) Gendered Discourse in the Professional Workplace. Basingstoke: Palgrave
McMillan.
Ozbilgin, M. and Healy, G. (2003) ‘Don’t mention the war. Middle Eastern careers in context’. Career
Development International, 8(7), pp. 325–327.
Qutub, A. (2013) ‘Harem girls and terrorist men: Media misrepresentations of Middle Eastern
cultures’. Colloquy, 9, pp. 139–155.
Sabbagh, A. (2005) ‘The Arab States: enhancing women’s political participation’. In: Ballington,
J. and Karam, A., eds. Women in Parliament: Beyond Numbers Handbook, 2nd ed. Stockholm:
International Institute for Democracy and Electoral Assistance, pp. 52–72.
Sunderland, J. (2004) Gendered Discourses. Basingstoke: Palgrave.
Walkerdine, V. (1998) Counting Girls Out: Girls and Mathematics. London: Falmer Press.
Walsh, C. (2001) Gender and Discourse: Language and Power in Politics, the Church and
Organisations. London: Longman.
Weedon, C. (1997) Feminist Practice and Post-Structuralist Theory, 2nd ed. Oxford: Blackwell.
Weir, D. (2003) ‘Human resource management in the Arab Middle East’. In: Lee, M., ed. HRD in a
Complex World. London: Routledge.
World Economic Forum (2017) Global Gender Gap Report 2017. Available at: http://www3.wef
orum.org/docs/WEF_GGGR_2017.pdf (Accessed: 27th June 2018).

436
 29
Feminist poststructuralism:
discourse, subjectivity, the body,
and power
The case of the burkini

Chris Weedon and Amal Hallak

Introduction
This chapter takes the example of the 2016 burkini ban in France and its treatment in four
UK newspapers to explore how feminist poststructuralism can be used to better understand
questions of power and representation. The key focus of feminist poststructuralism is the
relationships between discourse, subjectivity, the body, and power. It is concerned with
identifying the assumptions and power relations that underpin different forms of gender
and sexuality. It analyses language as social, political, and ideological discourses, show-
ing how these reproduce and sometimes challenge specific social practices. It is concerned
with where and how individuals and groups are produced as embodied subjects within the
discursive fields that constitute and reproduce social relations of power. Understanding of
this is crucial to social change.
Feminist poststructuralism is not a theory or methodology in the conventional sense.
It is a series of critical positions on language, subjectivity, the body, discourse, and power
that provide the grounds for mapping and analysing how relations of gender and sexual-
ity are socially constituted, lived reproduced and challenged. Language is central to these
processes. Feminist poststructuralism assumes that all theories are partial, constituting their
objects of study in particular ways that necessarily exclude other possible ways of inter-
preting gender and sexuality. In practice, they are also partial in a second sense of serving
historically specific, social interests.
Poststructuralism assumes that language constructs rather than reflects reality and that the
meanings of experience are constituted within language. Thus language works by temporar-
ily fixing meaning, but this meaning is constantly open to resignification and contestation.
Meanings are produced within discourses, which are more than spoken or written language.
They have a material dimension that helps shape subjectivities through relations of power.
For example, discourses of gender, sexuality, race, and religion are realised in individual bod-
ies, identities, material culture, and institutional practices. Discourses shape bodies, mind,
and emotions and produce and reproduce forms of agency, subjectification, and relations of

437
Chris Weedon & Amal Hallak

power (Butler 1990; Foucault 1981). Deeply ingrained cultural assumptions about the mean-
ings of women’s ‘nature’ or, in the case of this chapter, what it means to be a Muslim woman,
belong to wider discursive fields that have specific histories. The multiple discourses, institu-
tions, and social practices that constitute discursive fields produce meanings, subject positions,
and forms of subjectivity. They consist of competing and contradictory meanings structured
through relations of power according to which some forms of knowledge have more power
and status than others. Understanding ideas of women’s ‘nature’ as part of much wider discur-
sive fields, grounded in a range of social institutions and practices such as science, medicine,
religion, philosophy, the law, education, marketing, literature, media, and the arts, opens up
new understandings and possibilities of resistance and change.
In the 1970s and 1980s, feminist discussions and activism gave rise to new ways of
thinking, speaking, and performing gender and sexuality that highlighted not only similari-
ties in women’s experience but also differences. The powerful feminist language of sister-
hood, and the theories of patriarchy on which it was based, had immense emotional power,
but tended to mask real differences between women, rooted in class, sexuality, race, ethnic-
ity, religion, location, and disability (Weedon 1999). The many forms of theory and identity
politics that developed gave a voice to women who were often subject to multiple forms of
oppression. Thus feminist poststructuralism broadened its focus to encompass an intersec-
tional approach in which power relations of class, race, disability, ethnicity, and religion are
recognised as crucial components of feminist analysis.
Feminist poststructuralism seeks to understand how existing and new meanings, and the
power relations that they support, emerge and are internalised, embodied, performed, repro-
duced, and challenged. Drawing on the work of Foucault (1981) feminist poststructuralism
sees discourses of gender and sexuality as produced within social structures, institutions,
and practices, and in opposition to them. Discourses are structured by power relations and
produce subject positions of both subjectification and agency. The meanings that they pro-
duce and reproduce need to be understood in relation to their specific historical and cultural
location and the interests that they serve. In the twenty-first century the meanings of Muslim
women’s bodies, subject positions, and identities in the West have become a site of struggle
between multiple interests, as can be seen, for example, in the arguments used to justify the
War on Terror, constructions of the British or French ‘way of life’, or debates on Jihadism,
Islamism, fundamentalism, and social cohesion.

Language and subjectivity

Subjectivity is a crucial concern of feminist poststructuralism (Weedon 1996). According to
this theory, language works through the illusion of the sovereign speaking subject. Speakers
and writers see themselves as the origins of the meanings they produce, but these meanings
are always social. Put simply, language speaks us. Subjectivity is assumed to be sovereign,
but is defined by lack, an effect in part of the structures of language that produce subjects
who function as if they themselves are the source of the meanings that they speak. Language
constructs identities and subject positions. These have relative degrees of power. Gender
and sexual identities are assumed by taking up positions within social discourses and living
the identities that they offer (Butler 1993). Unlike many other forms of feminism, feminist
poststructuralism does not signal a particular politics, beyond the general principle that
power inheres in language and in all discursive practices, which are also material practices.
It serves as a way of analysing and identifying social networks of power and how they work
discursively to produce and reproduce forms of agency and subjectification.

438
 Feminist poststructuralism

Whose freedom? The case of the burkini

Muslim women in Western societies often face irreconcilable discourses of gender in which
their identities and agency are put into question. These discourses not only produce the
meanings by which we live but also the forms of embodied subjectivity that we inhabit.
They are not all socially accepted or valued. This can be clearly seen in the case of reli-
giously observant Muslim women who choose to cover up and whose agency is questioned
by many feminists as well as by mainstream society. In this section we turn to recent debates
in Britain and France about the meaning and acceptability of the burkini – a type of sports-
wear not unlike a wetsuit with a hood – that was originally designed in 2003 for Muslim
women. It aims to show how feminist poststructuralist insights might produce politically
useful analyses. This case study was chosen because it questions what is often understood
as freedom within feminism in different contexts and cultures and shows how Islamophobia
often takes a gendered form ‘which targets and affects women uniquely, adding to their
misogynistic oppression and religious victimisation’ (Jawad 2016). It suggests that, in any
feminist analysis, all founding assumptions need to be made explicit in order to under-
stand what they exclude and how they reaffirm or challenge existing power relations, which
encompass disability, race, ethnicity, and religion, as well as gender and sexuality. The
importance of recognising where one is speaking from is central to feminist analysis and
as authors we were positioned very differently. One of us is a Syrian Muslim who wears
hijab. The other is white British with no religious affiliation who has worked on multi-ethnic
Britain, including discourses about Muslims, since the 1990s.

Questions of method
Feminist poststructuralism is primarily a tool kit of theoretical principles with which to
approach how signifying practices function in society. It does not prescribe specific meth-
ods of studying language but draws on different approaches including deconstruction, criti-
cal discourse analysis, close reading, semiology, and symptomatic reading. It is composed
of analytic principles drawn from a range of theories, including Saussurean linguistics,
Lacanian psychoanalysis, continental philosophy (Foucault, Deleuze, Guattari, Lyotard),
and Marxism (in particular Althusser). Poststructuralist insights have been taken up and
developed in different ways by feminist theorists including, for example, Julia Kristeva,
Hélène Cixous, Luce Irigaray, Gayatri Spivak, Judith Butler, and Liz Grosz (see Weedon
1999). They are widely used by scholars working on gender and sexuality in colonial and
postcolonial contexts.
Whatever critical analytic techniques they use, poststructuralist readings question appar-
ently natural meanings and show them to be effects of power that privilege certain interests.
They may also offer oppressed subjects ways of resisting victim status. The analytical map-
ping of press coverage of the burkini in this chapter examines the structure and functioning
of language and its manifestation in discursive practices that are part of wider discursive
fields and suggests some of the ways in which power functions and where and how resist-
ance and change might be possible.
This case study of newspaper coverage of the burkini looked at over 60 articles published
in online editions of the Guardian, the Telegraph, the Daily Express, and the Independent
in August 2016 when the banning of the burkini from Cannes beaches hit the international
headlines. Analysis of the articles was conducted following Baxter (2018) via a process that
involved micro-linguistic, textual, representational, and discourse analysis. The analysis of

439
Chris Weedon & Amal Hallak

the language of the newspapers looked for linguistic evidence of stereotyping, for under-
lying assumptions about Muslim women, for whose voices were represented, and for the
textual strategies employed to normalise particular meanings and encourage the reader to
accept specific versions of reality as true. The analysis also looked for gaps, ambiguities,
and contradictions in the texts that might offer readers ways of contesting the positioning
of subjects. In poststructuralist terms, the readings were deconstructionist and attempted
‘to make the not-seen accessible to sight’ (Derrida 1967: 163). The focus on four different
newspapers aided the identification of the dominant messages in circulation and the binary
oppositions and stereotypes on which they rely. The articles were coded for stereotypes and
repeated linguistic tropes. The lexis used was analysed along with strategies of normali-
sation, (making things appear normal) found for example in the use of verbs, evaluative
adjectives, and adverbs and leading to ‘linguistic stance taking’ (Baxter 2018: 35). The use
of direct or indirect reported speech threw light on questions of whose voices are heard
and how voices are variously used to affirm, neutralise, or contest particular stereotypical
representations. Newspaper images were looked at in terms of their denotative and conno-
tative meanings (Barthes 1977, 2009) with attention to the ways in which the newspapers
anchored specific readings of images and ruled out others.
A key issue in this project was the question of voice, in particular the absence of Muslim
women’s voices from press coverage. In order to sharpen our analysis, we listened to
accounts by 15 Muslim women who have worn burkinis in public swimming pools of their
experience. We selected one story representative in terms of issues it raises to illustrate the
key concerns of a poststructuralist mode of reading and bring to light aspects of the wider
discursive field within which the newspaper case study is located.

Fatimah’s story
I’m 33. I’m a Syrian Muslim. I have a BA in sociology. I came to the UK in 2007. I have
a fair complexion and green eyes and I used to think that I’m lucky; I could pass for a
Brit if I kept silent and smiled. Unfortunately, I found it difficult to navigate through
my daily life wearing a headscarf. I had no idea how much prejudice it would attract.
Occasionally I would be harassed or insulted in the streets by boys and young men.
Although my spoken English was good, I would sometimes be neglected or met with
impatience by shop assistants when I asked for help. I started having social anxiety.
After I fell and broke my leg, I put on weight, which made my social anxiety worse.
I used to be active and sporty, but the bigger I got, the more anxious I became about
going to the gym. My physiotherapist encouraged me to go swimming but the last time
I went to a women’s session, I had a panic attack because I felt fat and exposed. Then
a friend gave me an M&S gift card. Although I have come to hate shopping because
it provokes my anxiety, I gathered my strength and went. I was more than surprised to
come across what looked like a wetsuit for women with a head covering attached to it
called a ‘burkini.’ For a moment, I forgot about my anxiety and the prejudice my head-
scarf attracts and decided to buy the burkini and go swimming.
Days later I went to my local swimming pool. As I was about to dip into the water,
an instructor yelled at me and told me that I couldn’t swim in that section of the pool
adding aggressively I might not be allowed to swim at all while wearing ‘that’ pointing
to my burkini. I was really upset, but decided I wasn’t going to let him discourage me;
I had come this far and was determined to have my swim. I went into a different section
of the swimming pool and started swimming.

440
 Feminist poststructuralism

By the time I finished, I’d recovered some confidence and decided I wasn’t going
to let the instructor’s behaviour go unreported. Before I got a chance to make my com-
plaint, the manager told me that a parent had come to him to complain about me pulling
their kid’s hair, and that the parents felt unsafe with me around. I was mortified and
outraged. First, I was nowhere near any children (the rude instructor made sure of that),
and second, and more importantly, I would never ever do such a thing. I tried to tell the
manager that that was simply not true, but he wouldn’t listen to me and instructed me
to wait for the police to arrive because he couldn’t leave the incident unreported upon
the insistence of the parent.
To say that I was shocked by the accusation is a gross understatement; I was shaking
with disbelief, anxiety, panic and rage. ‘They felt unsafe around me!’ What about me? I
did nothing to deserve such false accusation or unjustified treatment. All I wanted to do
was to get back to swimming and exercise. I should’ve known that something of the sort
would happen. How could I fool myself into believing that my headscarf would ever
cease to provoke unwarranted hostility? To think I was willing to believe that seeing
the burkini being sold in a respected store like M&S meant that the public is ready to
accept people like me! Even worse, to think I was going to complain about the instruc-
tor’s behaviour! I should’ve kept my head down. I should’ve adhered to my ‘remain silent
and smile’ policy. If I’d done so, I could’ve probably escaped unnoticed and unscathed.
I should’ve known that wearing a burkini was going to cause trouble. All these thoughts
and more were raging in my head while I had to wait for the police to arrive.
After a couple of hours – my husband had long since arrived to pick me up – the
manager came to let me go. He wouldn’t let me talk to the police. He told me that
the parent had agreed to drop the matter if I were banned from the swimming pool.
When we got home, my husband and I were shaking. I had what was the worst anxiety
attack of my life. I had reluctantly gotten used to being ignored, neglected, or harassed
because of my headscarf, but I had never ever in my life been criminalised for wearing
it. What else would’ve made the manager quick to believe such false accusation?
Once I calmed down, I thought something simply didn’t add up. There must be secu-
rity cameras everywhere in the swimming pool and it would’ve been easy to make sure
I wasn’t near any kids. If only the manager listened to me. Also, how come the police
didn’t want to talk to me? I told my husband I wasn’t going to be bullied into silence. I
wanted to report the matter to the police, and he called them. The police told him that
they would conduct a blind investigation to avoid any prejudice.
A few days later we received a phone call from the swimming pool’s manager to
apologise about the whole thing, claiming that there must have been ‘some sort of mis-
understanding’. He told us that they had revoked the ban on me going to the swimming
pool. We weren’t satisfied with a phone apology and my husband demanded a formal
written one. I would be lying if I said that the written apology magically made things
better. I was traumatised, but the one good thing that came out of the whole matter, I
suppose, was my decision to never remain silent again when it came to how I was con-
stantly mistreated because of wearing my headscarf.

In a close reading of this narrative, we focused on voice and subjectivity and how this is
established by choice of content and tone. The first person narrator begins by establishing her
class and education and her awareness of racism expressed in her references to her silence and
fair complexion. She describes her confrontation with everyday Islamophobia for which she
is ill-prepared and which produces social anxiety augmented by Fatima’s internalisation of

441
Chris Weedon & Amal Hallak

discourses of the slim female body. The narrative describes emotional responses and stresses
the subject’s positive sense of agency and capacity for resistance that is repeatedly undermined
both by the response of others to her body image and her own internalisation of repressive
norms of beauty. The chance nature of her acquisition of a burkini shows both how prejudice
can be temporarily forgotten and niche markets misread as acceptance or normalisation. The
narrative also points to the mixed effects of Islamophobic and sexist behaviour by the swim-
ming pool staff and their assumptions about what they can get away with.
At the level of subjectivity, the text signals how the narrator moves through a range of
emotional states and related subject positions. Everyday manifestations of Islamophobia
are shown to have reduced her to a strategy of trying to be as invisible as possible. With
the burkini incident, she moves from outrage to self-blame and then via rational reflection
from victimisation to anger and resistance. She is far from the oppressed Muslim woman
of Western media constructs and indeed it is aspects of Western society not Islam that are
shown to be oppressive.
Fatimah’s story offers insights into interrelated discourses of Islamophobia, subjectivity,
body image, self-blame, and white male power and their effects; yet it is also a story of resist-
ance and individual change. It raises a number of important issues about being a Muslim
woman in Britain today. The narrative details some of the effects of Islamophobic discourses
and practices on individual subjectivity with specific attention to its gender dimensions. It
shows how the practice of hijab co-exists with discourses of the desirable female body and
how both produce social anxiety in a patriarchal, Islamophobic society. Fatima is forced to
recognise that inclusion is not just a matter of pale complexion and class in an Islamophobic
society. If she continues to wear Muslim dress, she will meet rejection. At the same time, she
also comes to realise that she has agency, can contest Islamophobia, and can use the law to do
so. The tone of the narrative is descriptive and reflective rather than emotive. It suggests how
just being herself in any uncomplicated way is impossible unless she relinquishes one of the
mainstays of her identity, that is, wearing ‘hijab’ or modest Muslim dress.

Contested meanings and the importance of historical specifcity

Fatimah interprets seeing the burkini on sale in a major UK store as signifying that her
desire to integrate into British society is not only welcome but also encouraged. The fact
that her individual need to cover up while swimming is met by the market seems to her to
signify acceptance of difference. Yet for those with Islamophobic views and feelings, and
the power to act upon them, the values according to which Fatimah lives, expressed in her
mode of dress, are confronted as in some way threatening and incompatible with a ‘British
way of life’ that is based on ethnocentric, often romanticised, partial reconstructions of a
largely white and Christian Britain before major postwar immigration and globalisation. Yet
as cultural and social history suggest, the British ‘way of life’ has never been singular or
static and has changed continuously due to many complex factors that include race, ethnic-
ity, and religion.

Resistance
Resistance is a key component of Foucault’s theory of discourse, subjectivity, and power
and it is widely used in feminist poststructuralism (Foucault 1981). For Foucault, power
inheres in discourses and the subject positions that they produce. It has multiple forms
and effects, yet wherever it comes from, it has the capacity to produce resistance. Thus,

442
 Feminist poststructuralism

Fatimah’s negative experience leads to a transformation in her subjectivity. In an act of

agency, as she resists her subjection to Islamophobia, she determines never to be silenced
again.
Muslim subjectivity is an issue that goes far beyond dress codes and has entered the main-
stream political agenda in the UK in response to social unrest, the effects of Islamophobia
on individuals, and the turn by a small number of young British Muslims to extremism and
terrorism (Weedon 2016). Social research and Muslim writers have voiced the effects of
Islamophobia and racism on individuals and communities (Omaar 2006). These may be
far-reaching, especially for the young and vulnerable, as can be clearly seen in the case
of French Muslims and the ban on wearing hijabs in public places (Gemie 2010). At the
swimming pool, Fatimah’s dress seems to give non-Muslim men the licence to vent their
Islamophobic attitudes and even to ban her from the facility. Why is this the case? Here we
need to look to the wider discursive field and in the next section of this essay we draw on
examples of press coverage of the French burkini ban in the summer of 2016 as a pointer to
this discursive field and the power relations and interests structuring it.

Contested terrains: the burkini

Since its launch in 2003 the sportswear item, designed and marketed as the ‘Burkini’ by
Australian fashion designer, Aheda Zanetti, has attracted much media attention. In her anal-
ysis of press coverage of the burkini up to 2007, Fitzpatrick (2009: 2) describes how:

Around January 2007, a bathing suit designed for Muslim women became a media
sensation. The multi-piece, water-repellent suits, designed especially for women who
practice sartorial hijab, cover all of a woman’s body except the face, hands, and feet.
Although multiple companies sell these swimsuits, which retail for $100-$200, the one
that has received the most media attention is the Australia-based brand Ahiida, which
trademarked the name ‘Burqini.’ The press around the world eagerly adopted this port-
manteau of ‘burqa’ and ‘bikini,’ and generally refers to all full-body swimsuits mar-
keted to modestly-dressing women as ‘burkinis’.

The burkini hit the international headlines again in August 2016 and Debating Europe, a
website supported by the European Parliament, which initiated a debate on ‘Should the
burkini be banned?’ on 5th December 2016, summarised the affair as follows:

In summer 2016, the mayor of Cannes banned the burkini. Over two dozen other French
towns soon followed suit, many of them along the French Riviera. The mayor of Nice
supported the ban, arguing that its wearing was a ‘provocation’ in the aftermath of the
terror attack on his city in July 2016.
Photographs soon emerged of women being fined by police for wearing the burkini,
including one controversial set of pictures seemingly showing armed police forcing a
Muslim woman to remove her long-sleeved clothing (not a burkini) on a beach in Nice.
Witnesses allege that bystanders were shouting for the women to ‘go home’.
The ban was soon overturned by France’s highest administrative court.
(Debating Europe 2016)

Events in the South of France prompted social, cultural, and political debate and contro-
versy in the press internationally, on the internet and in social media where films of the

443
Chris Weedon & Amal Hallak

forced removal of both items of clothing and of Muslim women themselves from beaches
are available on YouTube. Many of the arguments in play had been widely discussed in
France and internationally when France banned the hijab in public places in 2004. A reading
of British newspaper coverage of the burkini ban suggests that while the focus of this debate
was the meaning and connotations of the burkini, as with Fatimah’s story above, the debate
was infused by deeper forms of hostility and power rooted in a discursive field in which the
long history of colonialism and Islamophobia has portrayed veiled women as other and a
threat to Western societies and values (Gemie 2010). In the burkini affair, multiple, often
contradictory, discourses can be identified in which the burkini is defined by French politi-
cians as:

• A challenge to the French secular state;

• he free expression of religious belief;
• A threat to security;
• A statement of support for terrorism;
• A symbol of the silencing and oppression of Muslim women;
• An example of women’s choice and agency.

As Dearden (2016) argued in the Independent online edition on 24th August 2016, inter-
pretations of the ban include arguments that it ‘preserves “security and secularism”’, along-
side those of more liberal critics who condemn it as a ‘sexist attack on human rights and a
valuable recruiting tool for Isis and other jihadist groups propagating the idea of a war on
Muslims in the West’.
Detailed attention to the issues outlined above in the section on questions of method sug-
gests that much of the British press coverage was concerned not just with factual reporting
but with interpretations that signal the wider discursive field within which the burkini debate
is located. For example, Dearden (2016) describes how the ban had become:

a lightning rod for a multitude of divisive issues. The imposition of local by-laws on
swimwear may seem minor but the ‘burkini bans’ have tapped into division over immi-
gration, sexism, religion and extremism as the country continues to reel from a series of
deadly terror attacks by Isis supporters.

Both terrorism and secularism are central to the French sources quoted in the British press.
Here the widespread binary opposition in Western discourse between religion and secular-
ism slips into an opposition between secularism and terrorism in which the positions of the
majority of Muslims who do not support radical Islamism become invisible. This points to
the power relations in play in the debate. Ironically, religious beliefs and practices become
politicised precisely in a way that most Muslims reject. In the process, mainstream Muslim
religious beliefs about female modesty are largely silenced and subsumed by a discourse
that insists on hijab as a political statement. Thus at one end of the mainstream political
spectrum, the Daily Express highlights statements by local politicians describing burkinis
as ‘beachwear ostentatiously showing a religious affiliation’ and ‘clothing that refers to
“terrorist movements”’ (Moore 2016). Analysis reveals how this article fails to offer any
alternative positions on the burkini or question the statements by French politicians but
implicitly reaffirms the implied connections between all Muslims and terrorism. In contrast,
Dearden in the Independent reports how French Prime Minister Manuel Valls told the La
Provence newspaper that he was not in favour of a national law but condemned burkinis:

444
 Feminist poststructuralism

‘“The burkini is not a new range of swimwear, a fashion,” he said. “It is the expression of a
political project, a counter-society, based notably on the enslavement of women”’ (quoted
in Dearden 2016). Like other less right-wing papers, the Independent reports the events and
French sources but places them within the broader debate on the meanings and implications
of the ban. A close reading of these newspapers also points to the spectre of specific political
agendas, foremost among which in the UK is the need for inclusion, social cohesion, and the
fight against both Jihadism and Islamophobia.

Women’s bodies, modesty, sexuality, and the male gaze

In their article ‘Reflections on the participation of Muslim women in disability sport: hijab,
Burkini, modesty and changing strategies,’ Limoochi and Le Clair (2011: 1300) point out
how: ‘The inclusion of women and girls in sport has been contested for centuries, tied to dif-
ferences in the gendered view of what is seen as “suitable” or appropriately “modest”’. Indeed
the history of women’s participation in sport in Western Europe since the end of the nineteenth
century shows how much women’s sportswear has changed with challenges to norms of femi-
ninity. While today Muslim men participate widely in sport at all levels, the participation of
Muslim women remains contested in many contexts. At issue are codes of behaviour and
dress. As Limoochi and Le Clair (2011: 1310) explain, for many Muslim women:

there are special circumstances related to the importance of tradition, and religious and
cultural values in the context of Islam that influence specific rules about the presenta-
tion of the female self and Islamic dress (hijab). These values impact on the participa-
tion of Muslim women in sport, and the different requirements for appropriate ‘modest’
dress which vary by country and by region, as well as by varying religious and cultural
values, and class/socio-economic factors.

The differentiation made here between different groups of Muslims on the basis of location,
religious and cultural values, and class/socio-economic factors is distinctly lacking in the
media where ideas of ‘modest dress’ have been widely satirised and attacked as an extreme
form of the oppression of Muslim women. As Fitzpatrick (2009: 77) points out:

Media stories persistently ignore the fact that for many Muslim women outside of the
Middle East, hijab is not a tradition that they are forced to follow, but a conscious deci-
sion that they made as young women or adults meant to be a cultural, political, and even
feminist statement. The possibility that some women wear burkinis as a bold rejection
of the dominant culture’s conception of what constitutes ‘progress’ is left conspicu-
ously unexplored by the mass-media.

The absence of such readings in the press points to questions of power as they affect whose
voices are included in public debates. They also signal questions of readership and point to
discourses about the ‘British of way of life’ that are widely used to contest multiculturalism
and religious diversity and are variously rooted in ethnocentrism, racism, and Islamophobia.
Yet, this absence of Muslim women’s voices can also be seen as part of the broad-based
ongoing struggle over control of women’s body images. As Fitzpatrick argues ‘the burkini’s
media popularity is rooted in its constructed identity as a visual symbol that refracts mul-
tiple overlapping contemporary debates about immigration, feminism, “culture-clashes,”
and national identity’. She adds that her ‘analysis contributes to a body of scholarship

445
Chris Weedon & Amal Hallak

that examines the ways in which women’s fashion and female bodies persistently serve as
discursive sites of meaning and arenas of contestation in the construction of national and
international imaginaries’ (Fitzpatrick 2009: 3). The meanings and uses of the burkini are
multiple: designed to meet the needs of Muslim women to dress modestly and counteract
the male gaze, it also sells widely to non-Muslim women who wish to cover up for a variety
of reasons. The burkini is also under challenge from Muslim fundamentalists who prefer
women to be absent from public life.
Writing in the Guardian in August 2016, Zanetti describes how it was her niece’s desire
to play netball that motivated her to design and market the burkini, along with her wish to
diversify the ethnic makeup of lifeguards on Australian beaches by creating a uniform that
Muslim women could wear.

After September 11, the Cronulla riots, the banning of the veil in France, and the inter-
national backlash that came with it – about us being the bad people all because of a few
criminals who do not speak on behalf of Muslims – I really didn’t want anyone to judge
girls wearing these. It’s only a girl being modest.
(Zanetti 2016)

Zanetti argues that the burkini was:

about integration and acceptance and being equal and about not being judged. It was
difficult for us at the time, the Muslim community, they had a fear of stepping out. They
had fear of going to public pools and beaches and so forth, and I wanted girls to have
the confidence to continue a good life. Sport is so important, and we are Australian! I
wanted to do something positive – and anyone can wear this, Christian, Jewish, Hindus.
It’s just a garment to suit a modest person, or someone who has skin cancer, or a new
mother who doesn’t want to wear a bikini, it’s not symbolising Islam.
(Zanetti 2016)

In our view, these meanings have very little institutional power and while they have been
recognised as signalling a marketing opportunity by retailers, they remain largely invisible
in mainstream public debate.

Agency and voice

Newspaper coverage of the burkini ban raises questions of agency and voice. With a few
exceptions, Muslim women’s voices are absent from articles and they become victims with
little agency or individual power who need rescuing (see, for example, Pearson 2016 writing
in the Telegraph). News articles concentrate on the competing meanings attributed to the
burkini rather than their implications for women or how the women in question understand
or experience their involvement. A notable exception is an article by Fishwick (2016) in
the Guardian in which five Muslim women speak confidently about why they wear the
burkini. For the most part, Burkini-clad women become the target of a range of attitudes,
fears, and anxieties that draw on Islamophobia and a range of other contingent national and
international issues linked to Muslim minorities in the West or to Muslim majority societies.
As Fitzpatrick (2009: 3) points out, press representations require location ‘within the his-
toric context of recent increased Muslim immigration to Christian-majority nations where
ideals of multiculturalism have existed alongside realities of cultural conflict, segregation,

446
 Feminist poststructuralism

suspicion, and racist nativism, especially in the wake of the terrorists’ attacks of September
11, 2001’. By 2016, repeated terror attacks by al Qaeda and Isis supporters had become
another significant contingent factor.
The analysis of press coverage of the burkini points to the multiple interests at stake in the
debate. In the context of the French Riviera, for example, unspoken economic interests tied
to Islamophobic views feature strongly, if often masked by appeals to preserving the ‘French
way of life’. Thus the insistence – addressed by French politicians and publicised by the media
– on excluding burkini-clad women from ‘Western’ beaches can be read as in part attempting
to maintain a specific image of glamour, wealth, and prosperity, underpinned by Islamophobia.
As Gemie (2010) argues of the French context, women in Islamic dress are trying to integrate
into public life but their efforts are rejected if they do not integrate in a form considered accept-
able. Writing in the Guardian on 25th August 2016, Farrer and Chrisafis (2016) suggest that
if Muslim women in France insist on hijab, their presence is viewed as a contamination and a
warning, which former President Sarkozy described as ‘provocation’.
A key issue raised by this brief examination of press coverage of the burkini affair is the
need for a full analysis of the genealogy of the interpretations of the burkini, the power rela-
tions structuring them, the contingent factors provoking them, and their effects on Muslim
and non-Muslim subjectivities and behaviour. A more extensive feminist poststructuralist
analysis of the burkini ban would require detailed empirical research that takes us beyond the
immediate language of the case study into the structure and genealogy of the wider discursive
field within which Muslim women in the West are located. It would raise further questions of
Eurocentrism and the assumptions underpinning the common sense idea that the discourses
produced from Western human rights organisations and Western feminism are best.
This chapter has briefly examined the example of the burkini to argue that a feminist
poststructuralist focus on the discursive production of meanings and subjectivity offers a
useful approach to understanding the constitution of meanings, subject positions, and the
power relations in play. As suggested above, reading newspaper articles from this perspec-
tive involves identifying stereotypes, inconsistencies, and contradictions, and examining
the language used to encourage the reader to accept specific truth claims. These are inherent
in the range of meanings and the broader cultural assumptions and binary opposition used
to establish a particular version of reality as true. Detailed textual analysis points the way
to identifying the social discourses reproduced and occasionally contested within the text.
It also enables the analyst to identify the subject positions in play and how they might be
resisted and transformed. As the case of the burkini shows, language constitutes conflict-
ing and contested meanings that produce and reproduce forms of subjectivity, subjectifica-
tion, and relations of power, empowering some groups and individuals while silencing and
excluding others. These meanings have real effects in the world. The interests and motiva-
tions behind the ways in which discourses function are multiple and complex and draw on
both contingent and historical meanings. Yet these meanings are constantly open to resig-
nification, and resistance to power can produce new and empowering forms of subjectivity
and voice. Institutions such as, in this case, the media and local and central government
play crucial roles in articulating and affirming discourses that victimise minorities and
legitimate sexism, racism, and Islamophobia, affecting individual bodies, identities, mate-
rial culture, and institutional practices as well as individual and group life chances. Often
women’s bodies become metaphors in discursive conflict over issues that go way beyond
questions of gender or sexuality yet this discursive conflict still shapes bodies, minds, and
emotions within social networks of power and works to produce and reproduce forms of
subjectification.

447
Chris Weedon & Amal Hallak

Future directions
Feminist poststructuralism is committed to analyses that might contribute towards
improving social relations. The future agenda for this approach to language, gender, and
sexuality studies is necessarily given by the social issues of the day. In the case of repre-
sentations of Muslim women, the analysis and contestation of the ways in which minor-
ity ethnic and religious communities are discursively constructed via representations in
the media, including the press and social media remains a pressing concern. This work
needs to locate the discourses in play in the broader history of representations of others
in British society and to identify how racism, Islamophobia, and ethnocentrism are pro-
duced, naturalised, and reproduced. In particular, such work needs to look at the power
relations in play in determining which voices and points of view are heard and which are
excluded. Writing about the burkini ban in the Independent, Jawad (2016) commented:
‘What hurts the most is the silence of fellow mainstream and “western” feminists whose
voices would have a significant impact on how these issues are framed and articulated’.
This is a challenge that needs to be taken seriously. The critical positions on language,
subjectivity, the body, discourse, and power that feminist poststructuralism opens up raise
a challenging agenda of important issues that have material social effects. While draw-
ing on individual acts of speech and writing and seeing discourse as performative, this
approach allows us to begin to access the power relations that structure the discursive
field and gain useful insight into how power functions in Western societies in relation to
Muslim women. This is an important step towards change.

Further reading
Baker, P., Gabrielatos, C., and McEnery, A. (2013) Discourse Analysis and Media Bias: The
Representation of Islam in the British Press. Cambridge: Cambridge University Press.
This book combines corpus linguistics with critical discourse analysis to look at how the lives of
Muslims and Islam are represented in the British press between 1998 and 2009.
Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press. Cham: Palgrave
Macmillan.
This book offers an excellent guide to press analysis.
Poole, E. (2002) Reporting Islam. London: I. B. Tauris.
This book analyses media representations of British Muslims.
Said, E. (1997) Covering Islam. Harmondsworth: Penguin Vintage Classics.
A classic analysis of press representations of Muslims.
Weedon, C. (1996) Feminist Practice and Poststructuralist Theory, 2nd ed. Oxford: Blackwell.
An introduction to postructuralist theory.

Related topics
Language, gender, and sexuality: reflections on the field’s ongoing critical engagement with the
sociopolitical landscape; poststructuralist research on language, gender, and sexuality; performance in
action; multimodal constructions of feminism; leadership language of Middle Eastern women.

References
Barthes, R. (1977) Image, Music, Text, Heath, S., trans. London: Fontana.
Barthes, R. (2009) Mythologies. Harmondsworth: Penguin Vintage Classics.

448
 Feminist poststructuralism

Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press. Cham: Palgrave
Macmillan.
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Butler, J. (1993) Bodies That Matter: On the Discursive Limits of Sex. London and New York, NY:
Routledge.
Dearden, L. (2016) ‘Burkini ban: Why is France arresting Muslim women for wearing full-body
swimwear and why are people so angry’. Independent [Online], 24th August 2016. Available
at: http://www.independent.co.uk/news/world/europe/burkini-ban-why-is-france-arresting-mu
slim-women-for-wearing-full-body-swimwear-and-why-are-people-a7207971.html (Accessed:
11th May 2017).
Debating Europe (2016) ‘Should the burkini be banned?’ [Online]. Started 12 May 2016. Available
at: http://www.debatingeurope.eu/2016/12/05/should-the-burkini-be-banned/#.Wb_qEIqQwn0
(Accessed: 18th September 2017).
Derrida, J. (1967) Of Grammatology. Baltimore, MD: John Hopkins Press.
Farrer, M. and Chrisafis, A. (2016) ‘Sarkozy calls burkinis a “provocation” that supports radical
Islam’. Guardian [Online], 25th August 2016. Available at: https://www.theguardian.com/wo
rld/2016/aug/25/burkini-ban-sarkozy-calls-swimsuits-a-provocation-that-support-radical-islam
(Accessed: 14th May 2017).
Fishwick, C. (2016) ‘Why we wear the burkini: Five women on dressing modestly at the beach’.
Guardian [Online], 31st August 2016. Available at: https://www.theguardian.com/world/2016/a
ug/31/why-we-wear-the-burkini-five-women-on-dressing-modestly-at-the-beach (Accessed: 15th
May 2017).
Fitzpatrick, S. (2009) ‘Covering Muslim women at the beach: Media representations of the Burkini’.
Thinking Gender Papers, UCLA Center for the Study of Women [Online]. Available at: http://
escholarship.org/uc/item/9d0860x7 (Accessed: 19th September 2017).
Foucault, M. (1981) The History of Sexuality, Vol. 1. London: Penguin.
Gemie, S. (2010) French Muslims: New Voices in Contemporary France. Cardiff: University of Wales
Press.
Jawad, H. (2016) ‘The burkini ban is misogynistic – And western feminists are turning a blind eye’.
Independent [Online], 13th August 2016. Available at: http://www.independent.co.uk/voices/burk
ini-cannes-islamophobia-banning-the-burkini-is-misogynistic-and-western-feminists-are-turning
-a-a7188806.html [Accessed: 6th May 2017].
Limoochi, S. and Le Clair, J. M. (2011) ‘Reflections on the participation of Muslim women in disability
sport: Hijab, burkini®, modesty and changing strategies’. Sport in Society, 14(9), pp. 1300–1309
[Online]. Available at: http://www.tandfonline.com/doi/abs/10.1080/17430437.2011.614788?src
=recsys&journalCode=fcss20 (Accessed: 20th August 2017).
Moore, F. (2016) ‘Now Cannes mayor bans burkini on French Riviera resort following waterpark
backlash’. Daily Mail [Online]. http://www.express.co.uk/news/uk/699216/David-Lisnard-Cannes
-mayor-ban-burkini-French-Riviera-Marseille-waterpark (Accessed: 25th September 2017).
Omaar, R. (2006) Only Half of Me: Being a Muslim in Britain. London: Viking Penguin.
Pearson, A. (2016) ‘Burkinis? here’s why we should fight them on the beaches’. Telegraph [Online],
30 August 2016. Available at: http://www.telegraph.co.uk/women/life/burkinis-heres-why-we-
should-fight-them-on-the-beaches/ (Accessed: 3rd May 2017).
Weedon, C. (1996) Feminist Practice and Poststructuralist Theory, 2nd ed. Oxford: Blackwell.
Weedon, C. (1999) Feminism and the Politics of Difference. Oxford: Blackwell.
Weedon, C. (2016) ‘Stuart Hall, the British multicultural question and the case of western jihadi
brides’. International Journal of Cultural Studies, 19(1), pp. 101–117.
Zanetti, A. (2016) ‘I created the burkini to give women freedom, not to take it away’. Guardian [Online],
24th August 2016. Available at: https://www.theguardian.com/commentisfree/2016/aug/24/i-created
-the-burkini-to-give-women-freedom-not-to-take-it-away (Accessed: 11th May 2017).

449
 30
Affect in language, gender, and
sexuality research
Studying heterosexual desire

Kristine Køhler Mortensen and Tommaso M. Milani

Introduction
Over the last 30 years or so, poststructuralism has become a key paradigm of thought in
critical feminist scholarship on language, gender, and sexuality (see Bucholtz 2014 for an
overview). Within this tradition, a large body of work has offered convincing evidence that
gender and sexual identities are not natural givens, but are ‘social constructions’ that are
produced, negotiated, and contested through discursive means, and are bound up with power
and control (see Ehrlich, Meyerhoff, and Holmes 2014). In this chapter, we take such a post-
structuralist approach to identities, but we also add another dimension, one that accounts for
the realm of the affective in relation to gender and sexuality. Bringing affect into the analyti-
cal repertoire means paying attention to the ‘role of fear, desire, anger and other powerful
feelings in shaping forms of action’ (Lemke 2007: 23). Of these, heterosexual desire is put
under the analytical spotlight in this chapter.
With regard to the remit of this section of the Handbook, it is important to state upfront
that our interest in affect is neither at odds with the role of language, gender, and sexuality
scholarship as a feminist practice, nor is it at variance with a poststructuralist interest in the
role played by discourse(s) in the production of identities and subjectivities (Norton 2000;
Weedon 1987). Quite the contrary, we argue that an affective lens allows us to achieve a
more nuanced understanding of meaning-making practices, and, gain deeper insights into
the ways in which felt subject positions are constructed in discourse and are linked to power
negotiations.
While there are of course many different emotions that may be relevant in relation to
gender and sexuality (see Milani 2015 and Milani et al. 2018 for analyses of shame in
sexual activism), this chapter focuses on the discursive construction of heterosexual desire
in a particular context: online dating. Empirically, we study heterosexual desire through an
investigation of two female friends browsing, reading, and commenting on male online dat-
ing profiles. By analysing these meaning-making practices, we demonstrate how desire is
what lies ‘beside’ (Sedgwick 2003) and enables the discursive construction of identities in a
homosocial context. At this juncture it is important to clarify that homosociality should not
be confused with homosexuality: it refers to the construction of ‘affiliation’ and ‘bonding’

450
 Affect in LGS research

in peer groups consisting of individuals of the same sex. While the notion of homosociality
has been particularly popular in the study of heterosexual male friendships and heterosexual
male bonding (see e.g. Hammarén and Johansson 2014; Kiesling 2005), we concentrate
here on heterosexual women (see also Coates 1996). Moreover, while homosociality is
often employed to indicate non-romantic connections, we highlight in our analysis how the
non-sexual desire to connect with persons of the same sex goes hand in hand with the pro-
duction of erotic desire for a person of the opposite sex, and vice versa (see Kiesling 2005,
2011; Milani and Jonsson 2011; Mortensen 2015a; Sedgwick 1985).
As we explain in more detail in the following section, the analysis in this chapter is
grounded on a key principle of poststructuralist thought, namely that both identity and desire
take shape and gain meaning through discourse; they are something we do with the help of
meaning-making resources: language understood as spoken and written code as well as
visual and embodied means, such as the arrow of a mouse on a computer screen. However,
poststructuralist theory does not translate into a distinct method or a single technique that
can be applied to any context (see the breadth of methodologies in the chapters in this part
of the Handbook). In other words, there is no recipe for how one goes about analysing iden-
tity and desire. In the case of the data presented below, we operationalise poststructuralist
thinking by paying attention to practices, that is, what people do with language and other
meaning-making resources in relation to online dating. In such an analytical enterprise, we
are guided by the following research questions:

How is desire brought into being in a same-sex peer group?

What kind of identity work is enabled by desire in such a group?

Postsructuralism and affect

Any research project and concomitant publications are based on specific assumptions about
the world. Because of various constraints, authors often do not have the space to spell out
such presuppositions. In contrast, we want to present the two main assumptions that under-
pin this chapter.
Assumption one: we take a poststructuralist approach to language. This means that we
do not view language as a mirror of a pre-existing world. Instead we see language as ‘per-
formative’, that is, constituting the world. To take gender as an example, Butler cogently
argues that gender is ‘a performative – that is, constituting the identity it is purported to
be. In this sense, gender is always a doing, though not a doing by a subject who might be
said to pre-exist the deed’ (Butler 1999: 33). This means that gender is not a ‘thing’ we
have that causes a particular behaviour. Rather, it is the discursive effect of socioculturally
situated discourses and meaning-making practices. In other words, gender is something
human beings do with the help of meaning-making resources, which comprise language,
body, dress code, make-up choices, and so on. If one takes a poststructuralist approach to
language, it is impossible to distinguish between what counts as ‘speech’ and what counts
as ‘action’, because speech and other meaning-making practices are themselves acts. In this
way, poststructuralist thinking destabilises the very foundations upon which many common-
sense assumptions rest, such as the distinction between reality and representation.
Assumption two: similarly to how we view language, we believe that emotions are also
performative: they are not states lodged somewhere in people’s minds or bodies, and there-
fore invisible, but are social forces that are produced, circulated, and materialised through
discourse. Analytically, a performative view of affect requires us to pay attention to the

451
K. Køhler Mortensen & T. M. Milani

practices through which emotions materialise. These practices can be linguistic – what peo-
ple say to each other e.g. the utterance ‘I love you’, visual – which pictures people exchange
with each other, for example on online platforms in order to generate desire, or embodied
– what people do with, say, the computer mouse when scrolling through online personal
profiles in order to indicate their attraction (or lack thereof) to the person in the profile.
Moreover, the analysis below illustrates how heterosexual desire does not simply romanti-
cally connect people of different sex, but also ‘sticks together’ (Ahmed 2004) people of the
same sex, in this specific case, two heterosexual women, creating a sense of momentary
coherence among them.
Crucially, our study of affect is part of a larger feminist project of relevance for scholar-
ship on language, gender, and sexuality. Emotions have problematically been viewed in the
Western tradition as ‘less valuable’ elements of the private sphere – women’s realm – as
opposed to rationality, which characterises the male space of the public sphere (Cvetkovich
2007; Gal 2002; Johnson and Ensslin 2007). In light of this, bringing affect into the scien-
tific spotlight can be seen as part of a feminist endeavour of valuing areas of human experi-
ences that have historically been devalued or even dismissed by male rationalist discourses.
A focus on affect seeks to debunk another widespread common-sense distinction – that
between ‘reason’ and ‘emotion’ – and forces us to resist the temptation to treat emotions as
less important ‘debris’ to be filtered out of social interaction.

Desire and discourse

Desire is a form of affect that has been discussed at length within language, gender, and sexu-
ality research. Interestingly, however, such debates have been rather disconnected from the
theoretical developments within the so-called ‘affective turn’ in the humanities and social
sciences. Against this backdrop, this chapter aims to partly redress this disjuncture. It does so
by showing how a performative approach to affect demystifies the view of desire as belonging
to an unreachable psychic inner life, a view that has caused discomfort among some sociolin-
guists and linguistic anthropologists (see e.g. Bucholtz and Hall 2004; Eckert 2002).
For example, in their ground-breaking collection entitled Language and Desire Harvey
and Shalom’s (1997) seem to presuppose that desire is a pre-discursive phenomenon, an
inner force that comes from the inside:

While uncompromising in its demand for attention, desire is also elusive and destined to
fade. To attempt to encode it – to write it, to speak it – is a way of capturing it, of attempt-
ing to delay the onset of its decline, and in providing us with a trace of the vividness of
our experience once it is past. But to give linguistic form to our desires for another human
being is also, importantly, to try to understand an experience that overwhelms us and
thereby threatens constantly to outmaneuver and outclass our verbal resources, the princi-
pal means at our disposal for ordering and making sense of our lives.
(Harvey and Shalom 1997: 1)

This quote touches upon one of the key challenges that desire has brought to the linguistic
study of gender and sexuality. If desire points to an internal psychic dynamic that is prior
to language, it is contradictory to poststructuralist thinking about the linguistic construction
of sexual and gendered subjects. By describing desire as a phenomenon with the capacity
to ‘outmaneuver’ and ‘outclass’ language, Harvey and Shalom implicitly position linguistic
analysis as not fully adequate to grasp the workings of desire. If desire is something that

452
 Affect in LGS research

exists in some prior form and can only tentatively be captured by language, how can a
linguist or discourse analyst study this phenomenon? Harvey and Shalom (1997) seem to
get around this challenge by categorising language as the individual’s primary resource for
‘ordering’ and ‘making sense’ of the world, desire included.
Although Harvey and Shalom (1997) introduced the notion of desire into sociolinguis-
tics, it was not until the publication of Kulick’s (2000) provocative review of research on
gay and lesbian language that an emotionally laden academic debate broke out. According
to Kulick (2000), language and sexuality scholarship had paid too much attention to issues
of sexual identity, and as a result failed to account for a key aspect of sexuality, namely the
realm of the erotic. As Cameron and Kulick explain,

[…] sexuality is a social and psychological phenomenon that often exceeds, and some-
times contradicts, the sexual identities people consciously claim or disclaim. What peo-
ple desire often clashes with, undermines or disrupts who they consider they are or
ought to be.
(Cameron and Kulick 2003: 113)

Methodologically, they suggest creating a ‘map’ of desire, showing not only how desire is
produced, circulated, but also on how it is blocked. A challenge to this mapping exercise
lies in the lack of more precise guidelines about how to go about operationalising desire
in linguistic analysis. Even though Cameron and Kulick’s (2003) proposal appears to be
practice-centred, issues of method are not resolved; they are rather formulated as questions
to the reader and the research community more broadly: ‘How do we do the mapping?’
(Cameron and Kulick 2003: 113).
Several studies have tried to answer this question by employing a variety of approaches
such as corpus linguistics (Bogetić 2013; Milani 2013), textual analysis (Barrett 2003), con-
versation analysis (Korobov 2011; Speer 2017; Stokoe 2010), as well as ethnography and
interactional analysis (Kiesling 2013). In what follows, we zoom in on the case of online
dating in order to demonstrate step-by-step how an analysis of desire in interaction can be
pursued.

Unpacking the language of desire: the case of online dating

We already mentioned in the introduction that we operationalise a poststructuralist approach
to language and desire by paying attention to practices, what people do with linguistic and
other meaning-making resources in the context of online dating. In order to understand
these practices, we first want to present online dating platforms; we then move on to discuss
the issue of ‘browsing together’ profiles on a Danish dating site, before offering a detailed
analysis of the ways in which heterosexual desire and homosociality are produced in the
interaction between two female friends.

Collecting data on online dating

Online dating sites are discursive platforms that foster initial communication between
potential romantic/erotic partners, thus providing internet users with various tools for con-
veying and negotiating their identities and desires. In contrast to offline dating such plat-
forms expand the romantic market by offering a large network of potential partners literally
with the tap of a finger. Additionally, they can improve partner-seeking through the use of

453
K. Køhler Mortensen & T. M. Milani

romantic compatibility algorithms. Online dating websites typically require members (1)
to construct a profile containing texts and/or photographs that convey personal information
(e.g. height, body type, age, occupation, etc.), and (2) identify the qualities they desire in a
potential partner. Moreover, most online dating services provide communicative tools such
as email and messaging, as well as lists of ‘favourites’. Once initial contact has been estab-
lished, members of the dating platform need to determine whether to pursue other forms of
communication outside the online context.

Browsing together on dating.dk

The dating website in the examples in this chapter is dating.dk, which is the first Danish
online dating site, founded in 1998. In 2011 when the data was collected, dating.dk was the
largest online dating platform in Denmark. The examples presented below are a very small
portion taken from a larger qualitative research project on heterosexual online dating among
young adults in Denmark conducted by the first author of this chapter (Mortensen 2015b).
Methodologically, how was the data collected?
The first step consisted in ethnographic style interviews and participant observation (for
more detail and ethical considerations see Mortensen 2015c). Participants in the study were
recruited through personal networks, as well as through interactions on the online dating site
dating.dk. Through ethnographic observations, it became clear how important it was for the
participants to share online dating activities with friends. A typical way of engaging friends
in one’s romantic life included showing them the user profiles that one found attractive.
Thus, online dating was not a solo activity that participants carried out while on their own
in front of their screens.
The second step of data collection emerged out of the ethnographic insights, and was
geared to better understanding the shared aspect of online dating. Recordings were made of
joint browsing practices among three pairs of friends with varying levels of online dating
experience. This chapter focuses on one of these conversations with two participants who
were the most experienced online daters. Experienced online daters in the project typically
spent less time talking about the online platform compared to their more inexperienced
peers; instead they dedicated more time negotiating their stances towards the profiles they
were browsing through. Thus, the conversation below was selected because it contains a
high degree of assessment talk – an interactional phenomenon, which we argue is a rich
‘epistemological site’ (Sunderland 2004) for the discursive construction of desire. The par-
ticipants were asked to sit together and browse while being recorded. The interactions were
arranged and cannot, therefore, be considered spontaneous. However, they simulate a type
of activity that had been naturally occurring among these friends. The recordings lasted
between one and two hours, and took place in the participants’ homes, a setting that was
chosen in order to make them feel as comfortable as possible. The screen-tracking software
‘Hypercam’ was used in order to capture footage of the screen movements as well as record
the sound in and around the computer.

Building desire through collaborative assessment

Stine and Louise – the names are pseudonyms – have been close friends since their teenage
years. At the time of the recording they were in their late twenties/early thirties and often met
to discuss all sorts of private matters. Ever since their teenage years, romantic relationships
had been a common topic of their mutual conversations. In the extract below, the women are

454
 Affect in LGS research

sitting together in Louise’s living room and are logged onto Stine’s online dating profile on
dating.dk. Contrary to what one might expect it is not Stine who controls the mouse when
browsing through her profile but Louise. This is because the device they are using belongs to
Louise, and they are in Louise’s home. Thus, Louise embodies an authoritative position with
regard to the hardware. The women check the list of ‘interested male users’ who have visited
Stine’s profile and have clicked on the ‘interested’ icon. They go through these men one by
one in order to decide whether to reciprocate interest. Figure 30.1 illustrates the configura-
tion of a male dating profile as seen from Stine and Louise’s viewpoint.
From the profile layout it is evident that the photo is highly salient. It is therefore no
surprise that when reading the profiles, the female friends pay most attention to the pictures.
In order to read the profile text, the reader has to make the effort of scrolling down the page.
The profile additionally includes a photo album that requires additional movements of the
mouse in order to access it. Throughout their browsing the women choose mainly to view
the photographic archive (the main profile photo and the photo album) and only engage
with the profile texts if they find the visual appearance attractive. Thus, the women to some
extent follow the reading path laid out by the design architecture of the dating site.
Going through the men’s profiles the women assess the profile owners. In conversational
interactions, assessment can be defined as ‘evaluating in some fashion persons and events
being described within talk’ (Goodwin and Goodwin 1992: 154). By analysing how assess-
ment is carried out and which elements are evaluated as attractive and unattractive we show
how desire comes into existence. Linguistically, desire, pleasure, and affect more broadly,
are not exclusively articulated through explicit statements such as ‘I love …’, ‘I hate’,
or ‘response cries, i.e. exclamatory interjections’ (Goffman 1978: 800) such as ‘mmh’ or
‘eeuw’. Such overt formulations are obvious discursive formulations through which affect is
produced. However, affect is also materialised discursively through more implicit meaning-
making processes (Besnier 1990; Irvine 1982) such as prosodic means (voice quality) and
laughter, as will become evident in the analysis below.

Figure 30.1 Configuration of a male dating profile.

455
K. Køhler Mortensen & T. M. Milani

Extract 11,2

Louise 01 °excuse me° but isn’t he a little bit pop-eyed

Stine 02 yes then it’s just- he’s really got a bit-
03 a very round head
Louise 04 @[@@]
Stine 05 [@] @ isn’t he [lacking-]
Louise 06 [without being] fat
Stine 07 lack[ing]
Louise 08 [he’s] actually quite-
Stine 09 isn’t he lacking a bit of cheekbone
10 and a bit of [u:h ]
Louise 11 [JAW] he’s lacking jaw
Stine 12 and jaw
Louise 13 but he’s also lacking cheekbones
14 and then a bit argh I do:n’t-
15 °and then a bit retardo eyes°
Stine 16 yes
Louise 17 cuckoo ((animated))
Stine 18 @ @[@@ @@@]
Louise 19 [@@ @@@]@

In this excerpt Louise starts by giving a negative assessment of the man’s looks (line 1), which
sets off an evaluation of a series of attributes. Stine introduces a specific assessment object, the
man’s round head (3), which generates simultaneous laughter. In the following turns (6–17),
the women further construe the man as laughable and deviant by pinpointing several deficien-
cies (jaw, cheekbones). Interactionally, assessment is achieved through overlappings (4–5,
5–6, 7–8, 10–11) and co-produced turns (10–12). According to Goodwin overlapping turns are
a ‘[…] way the participants can display that their minds are together’ (2008: 202). A similar
effect is produced in this excerpt by the mutual usage of ‘a bit’ (2, 9, 10, 14, 15). Overall, the
women construct shared agreement about every detail in the man’s appearance, and finally
Stine clicks the ‘not-interested’ icon thus ending the assessment activity.
Laughter serves two functions in the extract above. It builds up an affective affiliation in
the homosocial relation between the two friends. Put differently, laughter is a way of showing
enjoyment in the browsing activity at the same time as building rapport. Moreover, it serves
the function of devaluing the Other, that is, the male profile owner. In this sense, laughter
has the function of ridicule in this context. Louise further strengthens the ridicule by adding
the problematic expression ‘retardo eyes’ to her initial observation about the man’s eyes. By
being compared to a person with disabilities, the man in the profile is positioned by Louise
as an unsuitable object of desire. Research within disability studies has demonstrated that
disabled men and women are most often viewed as asexual (Gartner and Joe 1987; Kulick
and Rydström 2015). Since gender and sexuality are bound together, the expression ‘retardo
eye’ in this interaction not only contributes to desexualising the man in the profile, but also to
demasculinising him (Robertson 2004). In such a way, he is discursively relegated to a posi-
tion of marginalisation in the heterosexual marketplace (Eckert 2011). The women’s negative
assessments can be interpreted as an act of ableism; it is an alignment to ideologies accord-
ing to which only abled bodies are legitimate objects of desire. In dismissing the man as
unattractive, the women position themselves as legitimate subjects with the authority to pass

456
 Affect in LGS research

judgement and laugh at the deviant ‘Other’, at the same time as they reinscribe the ‘Self’ as
normal and attractive subjects within the same heteronormative ideology. In their collabora-
tively produced assessment, they affirm each other’s positions as abled and hence appealing
heterosexual women and acknowledge their co-performed non-desire as normative. The pin-
pointing of unattractiveness plays an important function in producing heterosexual desire in
the homosocial context. Display of non-desire is part of the coordinated development of intel-
ligible erotic taste, which ultimately solidifies the women’s homosociality.
In the second excerpt we shall see how, over a longer stretch of talk, Stine and Louise
negotiate desire and non-desire through their collaborative assessments. In this extract we
have included illustrations of the screen as well as descriptions of the mouse movements.
By studying the movements of the cursor on the screen it becomes evident how desire is
not simply brought into being through spoken discourse in this particular setting but also
through the dynamic interplay between embodied acts and verbal assessments.

Extract 2

Louise 01 @@[@@@@@@@ oh just take moves the

02 a look at his nose mouse towards
03 there] the man’s nose
Stine 04 [@@@@ it’s just about
05 switching something
06 around @@@]
Louise 07 [@@@@@@@@@@]
Stine 08 [@@@@@@@@] moves the
09 [@] mouse to the
photo album
index and
clicks on the
next photo
Louise 10 [@]
11 @@@ @[@]
Stine 12 [oh] I don’t know

13 (1.4)
Stine 14 [oh it’s a freaking conk
15 he’s got]
Louise 16 [oh it’s freaking big]
17 (2.0) clicks on the
next photo

Louise 18 from the front[he’s clicks on the

really-] next photo

457
K. Køhler Mortensen & T. M. Milani

Stine 19 [what what] was his clicks on the

20 height and so on by the next photo
21 way
22 he looks quite u:h clicks on the
next photo

moves the
mouse to
the man’s
username
and clicks on
it - leads back
to the profile
overview
Louise 23 he looks quite big
Stine 24 quite big yeah
Louise 25 no he’s not really it’s
26 really pretty normal
27 (1.39)
Stine 28 oh well still almost one
29 ninety
Louise 30 yeah yeah but but and moves the
31 then u:h in terms of mouse across
32 weight the physical
information
and marks
height and
weight
Stine 33 yes
Louise 34 it’s well quite nor-
35 really
36 (1.55)
37 *normal*

Aligning assessments are produced in several ways in this excerpt. The women employ
similar prosodic and lexical features in order to create affiliation. In response to a photo,
they say in overlap: ‘wow it’s a freaking conk he’s got’ and ’wow it’s freaking big’ (14–16).
The turns are synchronised with the explicitly affective exclamation ‘wow’. Wordings are
repeated in overlap and with identical prosody. Deployment of identical intensifiers further
displays reciprocity. Thus, the two women create affective affiliation in turn taking, lexis
and prosody. Moreover, they incorporate the mouse into their activity and make use of it
for interactional purposes, underscoring verbally uttered assessments and information. In
lines 1–3 the cursor is moved directly to the man’s nose and in lines 30–32 it is used to point
to the information about height and weight. In a sense, the mouse becomes a prosthetic
device with which the women can touch the male faces and point out specific attractive or

458
 Affect in LGS research

unattractive features. Additionally, the mouse is the tool through which they can act upon
their desire, since pressing the ‘interested’ icon is itself an act of desire in that it communi-
cates romantic/erotic interest to the receiver.
After agreeing on a man’s attractiveness, the women have to decide whether to click
interested/not interested and thereby initiate or close down further contact. In this case,
Louise takes the lead on making a decision.

Extract 3

Louise 01 there’s a perfect match Moves the mouse

((soft voice)) to the ‘perfect
match’ icon

02 (1.25)
Louise 03 and he’s online draws the mouse
((soft voice)) back to the profile
photo
04 sure you’re not interested draws the mouse
towards the centre of
the page and further
up to the right corner
05 no I’d better not control
06 the mouse
Stine 07 no uh uh if just press
08 interested but [if you-]
Louise 09 [°it’s
10 yours°]
Stine 11 he writes then you’ll
12 answer
13 (2.19)
Louise 14 that’s ok moves mouse to
the ‘interested’
icon and clicks
15 is that allowed
Stine 16 well I guess we are
17 allowed to make our own
18 rules
Louise 19 well okay returns to visitor
overview

Through simultaneous linguistic utterances and cursor movements, Louise draws Stine’s
attention to the fact that the man is a ‘perfect match’, and that he is online, thereby implying
that communication in real time is possible (1–3). At this point, the prosodic softening of
her voice works as a discursive device that marks a shift in terms of both identity and affect
in the interaction: from laughter, ridicule, and the production of the undesirable Other to
complicity and the discursive framing of a potentially desirable partner.

459
K. Køhler Mortensen & T. M. Milani

After a pause with no response from Stine, Louise withdraws from her attempts to con-
vince Stine to approach the man: ‘no I’d better not control the mouse’ (5–6). By manoeuvring
the mouse, Louise is able to act upon her own desire for the man as if she were Stine. She
then offers Stine the opportunity to take over the control of the mouse and choose whether to
approach the man or move on to the next one on the list. Stine turns down Louise’s offer but
does not fully dismiss the man as undesirable. Rather Stine involves Louise in the co-construc-
tion of desire by instructing her to answer if he returns the interest with an email (6–7, 10–11).
At first Louise accepts, but after a pause she questions the proposition they have agreed upon
(15). By requesting an account from Stine on this matter, Louise seems to imply some set
of norms about how ‘proper dating’ should be conducted. However, Stine opposes such a
normative stance by suggesting that they should make up their own rules. The women eventu-
ally engage in collaborative email correspondence with a male user and compose responses
together. Unbeknownst to the man in question, Stine’s romantic interest is not an individual
emotional status but is the result of the homosocial activity between Stine and Louise. On the
basis of this, it could be suggested that heterosexual desire and homosociality are ‘sticky feel-
ings’ (Ahmed 2004): they are glued together and stick different people together.

Discussion
In this chapter, we have argued for an approach to desire that is not concerned with its psy-
chic origin or locus, but focuses on meaning-making practices in social interaction. By pay-
ing close attention to situated contexts, we have stepped back from scholarly work taking
desire as naturally given; we have instead teased apart the ways in which desire is articulated
through meaning-making practices. The main point we made is that heterosexual desire
is a social phenomenon accomplished in interaction. Establishing a romantically intimate
heterosexual relationship does not exclusively involve interaction between two potential
partners of different sex, but, in the case of our example above, it also involves interactions
in same-sex peer environments. Hence, in this case, heterosexual desire is a collaborative
undertaking generated through homosocial bonding.
In considering desire as a social phenomenon, the analysis has also touched upon issues
of gender and heterosexuality. Female friends produce a body-centred desire in which they
split men into body parts, to which they attach different values of desirability. This can
be interpreted as a contestation of a hegemonic system of heterosexual desire in which
men typically value female partners according to physical attributes and women assess men
on the basis of socio-economic status (Coupland 1996). While the women’s emphasis on
physical features may be read as an agentive act against gendered expectations of ‘proper’
female desire, it may at the same time be seen as an internalisation of what film scholar
Mulvey (1975) calls the ‘male gaze’, a male-centred way of desiring through looking. It
could be argued that the physically focused desire that Stine and Louise are practising in
their reading of the male profiles is forced upon them by the very design of this online dating
site, which gives great prominence to personal pictures (see Figure 30.1). In other words,
the visual salience on the website may be viewed as a technological materialisation of the
‘male gaze’ with its requirement of public display of female attractiveness. One could go as
far as to suggest that the visual architecture of dating.dk structured the desiring paths – or
at least some of them – of the women in this study. However, as can be seen in Extract 3,
Stine and Louise creatively negotiate expectations about heterosexual desire practices; they
exploit the technological affordances of the website in ways that allowed them to engage

460
 Affect in LGS research

homosocially in their male-oriented desire. In a sense, Stine and Louise made up ‘their
own rules’ (Extract 3). Thus, the analysis gives visibility to the complexity of women’s
discursive positions. While deeply involved in compulsory heterosexuality (Rich 1980), the
women in this study challenge it as an exclusively two-party (female–male) relationship of
desire. By allowing the intimacy of the homosocial relation to play a managing role in act-
ing their desire online, the women’s practices illustrate heterosexual desire is a homosocial
collaborative endeavour.

Future directions
With the help of an approach informed by poststructuralism, this chapter has aimed to
demystify the ‘naturalness’ of desire, or any other affect for that matter. This has been
done through an analysis of the situatedness of heterosexual desire in a homosocial con-
text. The search for distinctive linguistic features and patterns denoting desire touches
upon the question of accessibility: do all people have access to the same meaning-making
resources? Not necessarily, we would argue (see Bucholtz and Hall 2004). Moreover, his-
torically entrenched social structures have a bearing on who can say what to whom; they
also work as normative frameworks about what people are expected to desire (or not) (cf.
Milani 2013; Piller and Takahashi 2006). These semiotic, ideological, and institutional
constraints are exciting topics of investigation in future research because they help us to
unveil the often subtle ways in which discipline and control operate not so much through
the mobilisation of individuals’ ‘rational capacities to evaluate truth claims but through
affects’ (Isin 2004: 225; see also Wetherell 2012). That being said, ‘where there is power,
there is resistance’ (Foucault 1978: 95), even when affect is concerned. Existing studies
about the political implications of affect-laden stances illustrate how different emotions
can be harnessed as powerful weapons in the subversion of hegemonic discourse (e.g.
Frederiksen 2012; Milani 2015).
Suggesting that desire, and any other affect, are socially constituted through linguistic
practices should not lead us to lose sight of bodily matters. Our analysis illustrated how
female users overwhelmingly focus on attaching desire and non-desire to the face and body
features of male users. Thus, this study demonstrates that, no matter how problematic this
might be, the body is a key locus of desire, being continuously inscribed with cultural val-
ues. However, the body that we meet in the domain of online interaction is not a fixed entity,
subject to the rules of biological science, but an unbounded object that can be fragmented
in the stimulation of desire. The body is made meaningful through talk as it is dissected and
carved up in particular ways. Outside the realm of the virtual, the body similarly persists as
an important locus for affective articulation. A burgeoning body of scholarship in linguistic/
semiotic landscape has demonstrated how the strategic placement of the body in space,
such as in the case of political demonstrations, can be powerful ways of creating affective
responses and consequently bring about political change (Milani 2015). Such research has
also shown how the skin is a fruitful site for the study of affective trajectories (Peck and
Stroud 2015).
In conclusion, by studying affect it is possible to bring to the fore what has previously
been viewed as ‘hidden’, belonging to the ‘private sphere’ or the realm of an ‘inner psychic
life’, and therefore dismissed as too slippery a terrain for scholars of language. It is our hope
that this chapter has demonstrated that what is ‘hidden’ is actually in plain sight. One needs
to read the data with an affective eye!

461
K. Køhler Mortensen & T. M. Milani

Transcription conventions
° lowered volume
CAPITAL LETTERS raised volume
[] overlapping speech
@ laughter, each token marks one pulse
(( )) transcriber comment
: length
- self-interruption

Notes
1 The extracts throughout the chapter have been translated into English from the original Danish.
2 Transcription conventions are provided at the end of the chapter.

Further reading
Ahmed, S. (2004) The Cultural Politics of Emotions. Edinburgh: Edinburgh University Press.
This book offers an in-depth discussion of how affects such as love, disgust, fear, hate, and shame
circulate and create problematic groupings of people in contemporary societies.
Cameron, D. and Kulick, D. (2003) Language and Sexuality. Cambridge: Cambridge University Press.
This book mounts a fully-fledged argument about the importance of desire in language and sexuality
research. It gives detailed summaries of different theories of desire as well as a comprehensive
overview of discourse-oriented studies of sexuality.
Leap, W. L. (ed.) (2018) Language/Sexuality/Affect. Special issue of Journal of Language and
Sexuality, 7(1), p. 300.
An important special issue that showcases the relationship between affect and sexuality through a
variety of empirical examples, from New Orleans bounce music to Black queer spaces, from the novel
Fifty Shades of Grey to university classrooms and Argentinian cinema.
Wetherell, M. (2012) Affect and Emotion: A New Social Science Understanding. Thousand Oaks, CA:
SAGE.
This book gives a comprehensive presentation of the social study of affect and offers examples of
the different approaches to emotions in social psychology and discourse studies.

Related topics
An ethnographic approach to compulsory heterosexuality; analysing gendered discourses online;
gender in interaction: ethnomethodological and CA approaches to gender; poststructuralist research
on language, gender, and sexuality; analysing gendered discourses online.

References
Ahmed, S. (2004) ‘Affective economies’. Social Text, 22(2), pp. 117–139.
Barrett, R. (2003) ‘Models of gay male identity and the marketing of “gay language” in foreign
language phrasebooks for gay men’. Estudios de Sociolingüistica, 4(2), pp. 533–562.
Besnier, N. (1990) ‘Language and affect’. Annual Review of Anthropology, 19, pp. 419–425.
Bogetić, K. (2013) ‘Normal straight gays: Lexical collocations and ideologies of masculinity in
personal ads of Serbian gay teenagers’. Gender and Language, 7(3), pp. 333–367.
Bucholtz, M. (2014) ‘The feminist foundations of language, gender and sexuality research’. In:
Ehrlich, S., Meyerhoff, M., and Holmes. J., eds. The Handbook of Language, Gender and Sexuality.
Malden: Wiley Blackwell, pp. 23–47.

462
 Affect in LGS research

Bucholtz, M. and Hall, K. (2004) ‘Theorizing identity in language and sexuality research’. Language
in Society, 33(4), pp. 469–515.
Butler, J. (1999) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Cameron, D. and Kulick, D. (2003) Language and Sexuality. New York, NY: Cambridge University
Press.
Coates, J. (1996) Women Talk: Conversation between Women Friends. Oxford: Blackwell.
Coupland, J. (1996) ‘Dating advertisements: Discourses of the commodified self’. Discourse and
Society, 7(2), pp. 187–207.
Cvetkovitch, A. (2007) ‘Public feelings’. South Atlantic Quarterly, 106(3), pp. 459–468.
Eckert, P. (2002) ‘Demystifying sexuality and desire’. In: Podesva, R. J., Roberts, S. J., Wong, A., and
Campbell-Kibler, K., eds. Language and Sexuality: Contesting Meaning in Theory and Practice.
Stanford, CA: CSLI Publications, pp. 99–110.
Eckert, P. (2011) ‘Language and power in the preadolescent heterosexual market’. American Speech,
86(1), pp. 85–97.
Ehrlich, S., Meyerhoff, M., and Holmes, J. (eds.) (2014) The Handbook of Language, Gender and
Sexuality. Malden: Blackwell.
Foucault, M. (1978) The History of Sexuality. Vol. 1. London: Penguin.
Frederiksen, M. (2012) ’Dimensions of trust: An empirical revisit to Simmel’s formal sociology of
intersubjective trust’, Current Sociology, 60(6), pp. 733–750.
Gal, S. (2002) ‘A semiotics of the public/private distinction’. Differences: A Journal of Feminist
Cultural Studies, 13(1), pp. 77–95.
Gartner, A. and Joe T. (1987) ‘Introduction to images of the disabled, disabling images’. In Gartner, A. and
Joe, T., eds. Images of the Disabled, Disabling Images. New York, NY: Praeger Publishers, pp. 1–7.
Goffman, E. (1978) ‘Response cries’. Language, 54(4), pp. 787–815.
Goodwin, M. H. (2008) The Hidden Life of Girls. Malden: Blackwell.
Goodwin, C. and Goodwin, M. H. (1992) ‘Assessment and the construction of context’. In: Duranti, A.
and Goodwin, C., eds. Rethinking Context: Language as an Interactive Phenomenon. Cambridge:
Cambridge University Press, pp. 147–190.
Hammarén, N. and Johansson, T. (2014) ‘Homosociality: In between power and intimacy’. SAGE
Open, pp. 1–11.
Harvey, K. and Shalom, C. (1997) ‘Introduction’. In: Harvey, K. and Shalom, C., eds. Language and
Desire: Encoding Sex, Romance and Intimacy. London: Routledge, pp. 21–42.
Irvine, J. (1982) ‘Language and affect: Some cross-cultural issues’. In: Byrnes, H., ed. Contemporary
Conceptions of Language: Interdisciplinary Dimensions. Washington, DC: Georgetown University
Press, pp. 31–47.
Isin, E. (2004) ‘The neurotic citizen’. Citizenship Studies, 8(3), pp. 217–235.
Johnson, S. and Ensslin, A. (2007) ‘“But her language skills shifted the family dynamics dramatically”:
Language, gender and the construction of publics in two British newspapers’. Gender and
Language, 1(2): 229–254.
Kiesling, S. F. (2005) ‘Homosocial desire in men’s talk: Balancing and re-creating cultural discourses
of masculinity’. Language in Society, 34(5), pp. 695–726.
Kiesling, S. F. (2011) ‘The interactional construction of desire as gender’. Gender and Language,
5(2), pp. 213–239.
Kiesling, S. F. (2013) ‘Flirting and “normative” sexualities’. Journal of Language and Sexuality, 2(1),
pp. 101–121.
Kulick, D. (2000) ‘Gay and lesbian language’. Annual Review of Anthropology, 29, pp. 243–285.
Kulick, D. and Rydström, J. (2015) Loneliness and Its Opposite: Sex, Disability, and the Ethics of
Engagement. Durham, NC: Duke University Press.
Korobov, N. (2011) ‘Gendering desire in speed-dating interactions’. Discourse Studies, 13(4), pp.
461–485.
Lemke, J. (2007) ‘Identity, development and desire’. In: Caldas-Coulthard, C. and Iedema, R. eds.
Identity Trouble: Critical Discourse and Contested Identities. Basingstoke: Palgrave, pp. 17–42.

463
K. Køhler Mortensen & T. M. Milani

Milani, T. M. (2013) ‘Are “queers” really “queer”? Language, identity and same-sex desire in a South
African online community’. Discourse and Society, 24(5), pp. 615–633.
Milani, T. M. (2015) ‘Sexual cityzenship: Discourses, spaces and bodies at Joburg Pride 2012’.
Journal of Language and Politics, 14(3), pp. 431–454.
Milani, T. M. and Jonsson, R. (2011) ‘Incomprehensible language? Language, ethnicity and
heterosexual masculinity in a Swedish school’. Gender and Language, 5(2), pp. 241–269.
Milani, T. M., Levon, E., Or, I., and Gafter, R. (2018) ‘Tel Aviv as a site of affirmation versus
transformation: Language, citizenship and the politics of sexuality in Israel’. Linguistic Landscape,
4(3), pp. 278–297.
Mortensen, K. K. (2015a) ‘A bit too skinny for me: Women’s homosocial constructions of heterosexual
desire in online dating’. Gender and Language, 9(3), pp. 461–488.
Mortensen, K. K. (2015b) Language and Sexuality in an Online-Mediated World: Interactional
Workings of Desire in Heterosexual Online Dating. Published PhD Thesis. Copenhagen: University
of Copenhagen.
Mortensen, K. K. (2015c) ‘Informed consent in the field of language and sexuality: The case of online
dating research’. Journal of Language and Sexuality, 4(1), pp. 1–29.
Mulvey, L. (1975) ‘Visual pleasure and narrative cinema’. Screen, 16(3), pp. 6–18.
Norton, B. (2000) Identity and Language Learning: Gender, Ethnicity and Language Learning.
London: Longman.
Peck, A. and Stroud, C. (2015) ‘Skinscapes’. Linguistic Landscape, 1(1–2), pp. 133–151.
Piller, I. and Takahashi, K. (2006) ‘A passion for English: Desire and the language market’. In:
Pavlenko, A., ed. Bilingual Minds: Emotional Experience, Expression and Representation.
Clevedon: Multilingual Matters, pp. 59–83.
Rich, A. (1980) ‘Compulsory heterosexuality and lesbian existence’. Signs, 5(4), pp. 631–660.
Robertson, S. (2004) ‘Men and disability’. In: French, S., Barnes, C., Thomas, C., and Swain, J., eds.
Disabling Barriers – Enabling Environments, 2nd ed. London: SAGE, pp. 75–80.
Sedgwick, E. K. (1985) Between Men: English Literature and Male Homosocial Desire. New York,
NY: Columbia University Press.
Sedgwick, E. K. (2003) Touching Feeling: Affect, Pedagogy, Performativity. Durham, NC: Duke
University Press.
Speer, S. (2017) ‘Flirting: A designedly ambiguous action?’. Research on Language and Social
Interaction, 50(2), pp. 128–150.
Stokoe, E. (2010) ‘“Have you been married, or …?”: Eliciting and accounting for relationship histories
in speed-dating interaction’. Research on Language and Social Interaction, 43(3), pp. 260–282.
Sunderland, J. (2004) Gendered Discourses. Basingstoke: Palgrave.
Weedon, C. (1987) Feminist Practice and Poststructuralist Theory. Oxford: Blackwell.
Wetherell, M. (2012) Affect and Emotion: A New Social Science Understanding. Thousand Oaks, CA:
SAGE.

464
 31
Language, gender, and the
discursive production of women as
leaders
Roslyn Appleby

Introduction
In this chapter I adopt a poststructuralist approach to understanding the ways in which
language, gender, and sexuality come together in the discursive construction of women
in political leadership. As an illustrative example, I focus on the media coverage of Julia
Gillard, Australia’s first and only female Prime Minister (2010–2013). Poststructuralism,
in the sense that I use the term in this chapter, views gendered identities as constructed or
enacted through discourses, which in turn are produced through language, text, and social
practices. Discourse is used here in the Foucauldian sense to refer to ‘practices which sys-
tematically form the objects of which they speak’ (Foucault 1972: 49); that is, discourses
give rise to particular and situated constructions of reality, thereby reproducing forms of
knowledge that entail particular relations of power (Litosseliti 2006). When used in this
critical theory tradition, and with a focus on language in use, ‘discourses’ refer to con-
ventional patterns of speaking and writing about any given topic in ways that uphold and
give expression to a powerful set of social and institutional values, beliefs, attitudes, and
practices (Cameron 2001).
When individuals – including politicians and media commentators – talk about a topic
or a person, they draw from shared resources, and through such talk a particular version of
reality is constructed (Cameron 2001). Following from this understanding, discourse analy-
sis is concerned with how people and all aspects of social life are labelled and categorised
(Lee and Petersen 2011). Gendered discourses thus position women and men in certain
ways, as particular types of people. In political leadership, a woman may be constituted by
her colleagues and by the media as a particular type of person through the deployment of
normative discourses of gender and sexuality that circulate in society and serve to maintain
broader, structural relations of power.
My poststructural analysis also draws on related understandings of gender as a fabri-
cation, rather than a fixed, pre-given entity, and a ‘performative accomplishment’ that is
enforced through social sanctions (Butler 1988: 520). Within this rigid regulatory frame-
work, individuals identifying as women are subject to discourses of femininity that promote
particular ideals and norms of physical appearance, speech, and behaviour (Baxter 2018).

465
Roslyn Appleby

These discursive frameworks define the ways in which we can be seen as intelligible gen-
dered beings (Mills 1997). For individuals, a recognisable, credible, conventional perfor-
mance of gender is therefore ‘a strategy of survival’, and ‘those who fail to do their gender
right’ through the performance of interrelated socially accepted discourses ‘are regularly
punished’ (Butler 1988: 522).
On the basis of this theoretical understanding, the particular methodology I use to inves-
tigate the construction of gender and sexuality is informed by the framework proposed by
Bucholtz and Hall (2005) for analysing linguistically produced identity. This framework
explains how identity emerges in linguistic interaction through the roles and orientations
assumed by participants, and through the evaluative social category labels used to describe,
position, and differentiate self and others. The particular identities – or rather subjectivities
– that emerge through these linguistic means may be temporary, shifting, and deeply embed-
ded in specific localised contexts and domains, and yet they also function to link individuals
(self and others) to broader macro-categories including gender, age, sexual status, profes-
sion, and social class.

Women leaders and the news media

One important public domain in which language, gender, and sexuality have been studied
from a poststructuralist perspective is that of political leadership. As Baxter (2018: 5) has
pointed out, despite women’s significant progress in politics over recent decades, political
leadership is still primarily defined as a masculine domain in which women are positioned
as outsiders. Working with a normative model of leadership, news media often represent
women politicians in essentialised or sexualised ways that foreground physical appearance,
and women who do not meet stereotyped modes of sexualised attractiveness are often sub-
jected to vicious attacks in the media.

Making a Prime Minister

To illustrate the theoretical framework discussed so far, I turn to an exploration of the ways
in which normative discourses of gender and (hetero)sexuality were deployed in the politi-
cal arena and in the media to depict Julia Gillard, Australia’s first female Prime Minister
(2010–2013), as a particular type of gendered subject. On the one hand, Gillard’s appoint-
ment was initially celebrated as a sign of progress in gender equality; on the other hand,
she was repeatedly ‘punished’ as one who had failed to ‘do [her] gender right’ (Butler
1988: 522). Her lack of conformity with normative gender regimes, primarily to do with
her unmarried and child-free status, had a significant deleterious effect on perceptions and
evaluations of her performance as Prime Minister.
Gillard was initially elected to the House of Representatives in the Australian parliament
in 1998. As a member of the (social democratic) Labour Party she was committed to social
justice and had a special interest in education and workplace reform. With the Labour Party
then in opposition, Gillard was recognised as a talented debater on the floor of the parlia-
ment and was seen by some as a potential future party leader. In December 2007, Gillard
became Australia’s first female Deputy Prime Minister when Labour won the general elec-
tion and in this role was again recognised as a popular and consummate performer.
By mid-2010, a decline in Prime Minister Kevin Rudd’s popularity in the opinion polls
and dissatisfaction with his leadership style prompted a change in leadership. Gillard was
elected unopposed as party leader and thus became Australia’s first female Prime Minister

466
 Women as leaders

in June 2010. In order to legitimise her position, Gillard called a general election that was
held in August 2010. Despite her initial popularity, particularly amongst women voters, the
internal political and external media campaigns against her were effective in diminishing
her overall support base. In the wake of the 2010 election Gillard became the Prime Minister
of a minority government and, over the succeeding years, introduced a raft of significant
social reforms. However, she continued to face significant opposition from various quarters:
from aggressive opposition parties, from a hostile, scandal-mongering media, and from a
clique of embittered Rudd-supporters within her own party. As a consequence, her ratings in
the polls declined, and defeat at the 2013 general election looked certain. Fearing electoral
disaster, the Labour Party once again moved to change its leadership; in an internal ballot,
Gillard was defeated and Kevin Rudd was once again elected party leader in June 2013. The
Labour Party under the leadership of Kevin Rudd lost the general election in September
2013, and Tony Abbott, as leader of the Liberal Party, became Australia’s twenty-eighth
Prime Minister.
The reasons behind Julia Gillard’s rise and fall have been widely debated, with some
critics pointing to failures of leadership, and others claiming that she had been subjected to
a media crusade of sexism and misogyny previously unseen in Australian political history.

Analytical approach
The data I discuss in this chapter come from my own random selection of texts that appeared
in the Australian media during the years of Gillard’s term in office as a Member of Parliament
and as Prime Minister. Most of these texts can be accessed electronically. They comprise
newspaper and television reports (including reports of debates and speeches in parliament
and press conferences), media commentaries, magazine stories, images (photographs and
cartoons), and radio broadcasts and interviews. The texts have been analysed to identify
recurring patterns and emergent discourses, the latter term used in the sense described below.
In the following analysis, the first section focuses on the discursive dilemma that consti-
tutes the ‘double bind’ (Hall and Donaghue 2012) whereby women who are leaders are not
only expected to demonstrate the toughness and authority stereotypically associated with
the masculine domain of political power but also, as women, expected to behave in ways
that are stereotypically associated with femininity, by demonstrating practices of caring,
collaboration, and consensus. The second section focuses on the question of gender ‘intel-
ligibility’. As Butler (1990: 17) observes, normative acts of gender serve to render a person
‘intelligible’ as an individual in contemporary culture, and if one does not conform with
the normative practices that define, govern, and regulate gender (and sexuality), one risks
becoming ‘unintelligible’ as a person, and thereby subject to the ‘punitive consequences’
discussed above.
The third section focuses on the ways in which Gillard eventually adopted a practice of
‘strategic essentialism’, demonstrated most forcefully in a key speech through which she
performed as an intelligible gendered subject. The term ‘strategic essentialism’ was first
coined by Gayatri Spivak in the context of postcolonialism and refers to the adoption of an
essential identity (defined, for example by race, class, or gender) by a subordinated group
in order to counter dominant colonial narratives. In feminist philosophy, essentialism has
been acknowledged as descriptively false, in that it denies the real diversity of women’s
lives and experiences by assuming that all women share a set of invariant biological and/or
social characteristics. In this light, essentialism can be oppressive because it regulates what
is and is not considered appropriate for women by privileging certain normative forms of

467
Roslyn Appleby

femininity (Stone 2004: 137). However, acts of strategic essentialism, whereby a speaker
deliberately aligns with the discourses associated with a particular identity category, have
been recognised as a practical political tool that can articulate agency, encourage shared
identification, and mobilise forces for transformation.
These three sections also reference three phases in the media representation of Gillard
as Prime Minister. In the first phase are press reports of her ascension to the position of
Prime Minister (24th June 2010) and her subsequent attempt to secure electoral endorse-
ment just two months later (21st August 2010). In the second phase are the increasingly
virulent attacks generated by or reported in the media throughout her term in office and
directed at her gender ‘failings’ (June 2010–June 2013). The third phase marks a turning
point in Gillard’s own performance as a recognisable yet resistant gendered being where,
increasingly over the last year of her term in office, she voiced her objection to sexist and
misogynist attacks in the media and in parliament. Throughout this analysis I have selected
media reports that most clearly demonstrate the gendered discourses that circulated during
Gillard’s term as a Member of Parliament and as Prime Minister.

The double bind

When Julia Gillard assumed the role of Australian Prime Minister in December 2010 it
was indeed an historic moment: not only was Gillard the first woman in this role, she was
also unmarried (and in a de facto relationship), childless, and an atheist, all conditions that
were almost unheard of amongst Australian political leaders. Her appointment ran counter
to the pattern of gender inequality in Australian politics, business, and the workforce more
broadly, where the gender pay gap was 17 per cent and widening, and the representation
of women in politics and on corporate boards in Australia remained at dismally low levels.
The difficulty of combining family and work responsibilities has been a significant problem
for Australian women in a way that it never has been for Australian men, and yet women
who have forged a career and – like Gillard – ‘neglected’ to have children are ‘alternately
castigated and pitied’ (Summers 2013a: 4).
Barriers to women’s participation in Australian politics include the persistence of long
irregular working hours and a highly combative debating style. When women do appear
as participants or leaders in these male-dominated communities, they find themselves in a
double bind: on the one hand, in order to garner support and respect, they must exhibit lead-
ership qualities of strength, authority, and decisiveness, traits that are traditionally perceived
as masculine; on the other hand, they must contend with the prescriptive gender stereotypes
which demand that women should demonstrate characteristics of warmth, nurturance, sen-
sitivity, and self-effacement. If a woman in a leadership position demonstrates behavioural
characteristics of strength, agency, and authority – which Gillard clearly did – she risks
being seen as insufficiently feminine and cast as an aggressive, calculating ‘Iron Maiden’,
one of four gendered stereotypes that render women ‘unsuitable for leadership’ (Baxter
2018: 23). At the same time, if she demonstrates characteristics associated with a softer
femininity, she risks being perceived as lacking the qualities of toughness required to be a
good political leader.
The efforts to manage discourses of gender were evident on both sides of parliament in
the political campaign of 2010, during which Gillard sought an electoral endorsement for
her position as Prime Minister. From the outset, Gillard’s popularity amongst women vot-
ers contrasted sharply with women’s lack of enthusiasm for the Leader of the Opposition,

468
 Women as leaders

Tony Abbott, whose patriarchal stance on issues affecting women posed problems for his
appeal to women voters (Denmark, Ward, and Bean 2012). Known as a ‘hyper-masculine
opposition leader and ironman triathlete’, Abbott had habitually projected the persona of an
‘action man’ with a predilection for involvement in strenuous sports (Sawer 2010, para 2).
Abbott’s hypermasculinity inevitably cast Gillard in a contrasting role and emphasised her
performance as a female politician, potentially incapable of tough political decisions and yet
still insufficiently feminine.
Although Gillard studiously avoided gender issues in the 2010 election policies, the
gendered double bind nevertheless played a key role in several campaign events that
attracted a great deal of media attention. First, the way that Gillard had replaced her pre-
decessor, Kevin Rudd, in a party-room ballot on 24th June 2010 was presented in much
of the popular press as evidence of her cold, ruthless ambition, a trait typical of the Iron
Maiden stereotype that was implicitly inconsistent with stereotypical feminine qualities
of warmth, collaboration, and self-effacement. Two days after she assumed the position of
Prime Minister The Courier Mail, for example, noted that ‘the ambitious Gillard did not
hesitate to take up the knife and plant it in Rudd’s back’ (Oakes 2010). Using the same
metaphor, The Age pointed out that since ‘nice girls don’t carry knives’, Gillard would
‘have to be persuasive in explaining how she came to plunge one into Kevin’s neck’
(Grattan 2010). This use of colloquial terms denoting the use of weapons and extreme
violence are typical of the treacherous Iron Maiden stereotype (Baxter 2018: 40). Media
reports such as these, and those that appeared throughout her time in office, typically
presented Gillard’s ambition, authority, and decisiveness as inappropriately aggressive
and evidence of her failure to meet the normative expectations attached to the category
of female leader.

An unintelligible being
While the negative effects of the double bind clearly shaped reports of Gillard’s public per-
formance as a political leader, representations of her personal appearance and her private
life produced across a range of sites and circulated by the media proved even more dam-
aging. Gillard’s private life was subjected to an extraordinary degree of scrutiny, and she
became the target of an ‘unrelenting campaign of vilification and vitriolic sexist abuse by
a loose coalition of shock jocks, bloggers and newspaper columnists’ (Sawer 2013). These
attacks again centred on Gillard’s supposed failure to conform to appropriate norms of femi-
ninity and cast her as an unnatural being who had refused ‘conformity with recognizable
standards of gender intelligibility’ (Butler 1990: 16).
In common with other women in politics, Gillard’s physical appearance – her hairstyle,
her (red) hair colour, her clothes, her body shape, her voice – were the topic of much public
discussion, consternation, and criticism in a regime of continual scrutiny rarely directed at
men. Yet it was more than Gillard’s personal appearance that became the object of derision.
As Summers (2013b: 4) explained, her situation ‘pushes all the buttons that get conserva-
tives exercised: she is not a mother; she is not married, she lives “in sin”, she is an athe-
ist’, and her former partners, invariably referred to as a series of ‘boyfriends’, were often
named in reports of her rise to power. Gillard’s private relationships clearly failed to comply
with the most favoured form of heterosexuality, which is marked by monogamy (signalled
through marriage), reproduction (signalled by children), and conventionally hierarchical
gender roles (Cameron and Kulick 2006).

469
Roslyn Appleby

Prior to her promotion to Labour Deputy Leader, an infamous photograph published

in a major newspaper in 2005 depicted Gillard in the kitchen of her suburban home, in a
tableau that was described the following day in The Sydney Morning Herald as ‘lifeless’,
with ‘bare walls’, ‘stark benchtops’, and, most notoriously, an ‘empty fruit bowl’ (Hornery
2005). According to the reporter, this was an image that sent a ‘chill wind through readers’,
and it became an iconic sign that Gillard was incapable of domesticity and, by implication,
proper femininity. In a similarly well-publicised interview in The Bulletin a senior con-
servative politician, Senator Bill Heffernan, claimed that Gillard was ‘not qualified to lead
the country’ because she was ‘deliberately barren’ (Harrison 2007). Justifying his remarks,
Heffernan explained that ‘If you’re a leader, you’ve got to understand your community. One
of the great understandings in a community is family, and the relationship between mum,
dads [sic] and a bucket of nappies’ (Harrison 2007). Along similar lines, a former leader
of her own Labour Party, Mark Latham, in an interview in The Australian, observed that
Gillard’s decision not to have children meant that ‘by definition you haven’t got as much
love in your life’ and therefore have ‘no empathy’ (Kelly 2011).
The raft of negative images and evaluations came together in the campaigns con-
ducted by the media during the term of Gillard’s prime ministership. An abiding linguistic
theme revolved around the words used to describe Gillard. Prominent amongst these were
‘Ju-LIAR’, coined in an interview between Gillard and radio shock jock Alan Jones, and
the phrase ‘Ditch the witch’, which initially appeared at an anti-Gillard protest rally and
gained wide media coverage. In a parallel campaign, pornographic photoshopped images
of Gillard’s naked form, as well as cartoons of her wearing a giant dildo, were published
on the internet and circulated to politicians and journalists (Keane 2012; Summers 2012),
and in the final year of her prime ministership, the press circulated reports of a menu at
a Liberal Party fundraising dinner that included one dish described as ‘Julia Gillard’s
Kentucky Fried Quail – Small Breasts, Huge Thighs, and a Big Red Box’ (Overington
2013). Through the circulation of these texts and related media commentary, Gillard
became ‘the victim of appalling levels of sexism not seen before in Australian public life’
(Sawer 2013, para. 2).
It was shortly after the menu item was reported in the media, and near the end of her
three-year term, that a radio interview with Perth radio 6PR shock jock Howard Sattler was
broadcast, once again bringing to the surface explicit questions about Gillard’s sexuality.
In the interview, reported by all major media, including ABC News (2013) and The Sydney
Morning Herald (Spooner 2013), Sattler confronted Gillard with questions regarding her
marital status, asking why she wasn’t married, and whether her partner, Tim Mathieson, had
proposed. Sattler then turned to the question of Mathieson’s sexuality:

HS: Myths, rumours, snide jokes and innuendoes, you’ve been the butt of them many
times
JG: Well I think that’s probably right (laughs). We’ve certainly seen that this week [refer-
ring to the Liberal Party dinner menu]
HS: Can I test a few out?
JG: In what way?
HS: Tim’s gay
JG: Well-
HS: No, somebody’s saying that- that’s a myth
JG: Well that’s absurd
HS: But you hear it. He must be gay, he’s a hairdresser

470
 Women as leaders

JG: Oh well isn’t that-

HS: But you’ve heard- you’ve heard it
JG: But ah-
HS: It’s not me saying it, it’s what people say
JG: Well I mean Howard, I don’t know if every ah silly ah thing that get’s said is going
to be repeated to me now
HS: No no but-
JG: But ah, to all the hairdressers out there, including the men who are listening, I don’t
think, in life, one can actually look at a whole profession, full of different human
beings and say ‘gee, we know something about every one of those human beings’. I
mean it’s absurd, isn’t it.
HS: You can confirm that he’s not?
JG: Oh Howard don’t be ridiculous-
HS: No but-
JG: Of course not
HS: Is it a heterosexual relationship, that’s all I’m asking

A range of discursive practices were used in the Sattler interview to generate audience con-
sent and solidarity. One of the most salient in this extract is Sattler’s adoption of the ‘peo-
ple’s tribune’ role (Talbot 2010: 192), whereby he positions himself as relaying, on behalf
of the wider community, concerns about Tim’s ‘true’ identity. This elitist role, appropri-
ated from the genre of serious investigative interviews, enabled Sattler to engage in a form
of aggressive, inquisitorial dialogue, characterised by repetition, interruption, and dogged
grilling, in order to get to ‘the truth’. In his public inquisitor role, the specific tactic Sattler
employs is an ambiguous ‘double voicing’ (Talbot 2010: 193), where he claims ‘It’s not
me saying it, it’s what people say’: he separates himself from the opinions he expresses by
merely acting as an ‘animator’ of the ‘myths, rumours, snide jokes and innuendoes’ that he
insists are circulating in the public domain. This tactic of presenting the views of others is
typically used to protect the journalist’s guise of neutrality, but is also used, as in Sattler’s
case, to insult the interviewee while maintaining a neutral stance.
Despite Gillard’s typically measured and rational responses, demonstrated in the
interview by her calm explanation that warns against stereotyping any particular group
(including, in this instance, hairdressers), Sattler was not deterred from his thinly veiled
accusations. His relentless questioning points to an obvious, yet unspoken, problem: if
Gillard’s partner, Tim, was a hairdresser, he must be gay, and so she must be … what?
Perhaps a lesbian? Most certainly living a lie, and definitely not fit to govern the country.
Although Sattler was later suspended for this line of questioning, the interview lived on,
echoing across the nation and crystallising the problem of Gillard’s questionable sexual-
ity. As Cameron and Kulick (2006) point out, ‘compulsory heterosexuality’ plays a cru-
cial role in the construction of gender, and Sattler’s interview illustrates how normative
regimes of both gender and sexuality were invoked to undermine Gillard’s credibility.
Together, the panoply of sexist ‘myths, rumours, snide jokes and innuendoes’ referred
to by Sattler worked to produce a particular discursive image of Gillard as an unnatural
being: not only as a person unfit to lead the country, but also as a person unfit to be a rec-
ognised as a normal woman. However, opposition to these attacks was building amongst
progressive elements in the community and, in the year leading up to this interview,
Gillard had also begun to articulate a different discursive practice that offered a firmer,
feistier resistance to her foes.

471
Roslyn Appleby

Strategic essentialism
Having initially chosen not to focus explicitly on issues of gender discrimination, perhaps in
fear of being cast as a victim and therefore too weak to govern, Gillard finally started, two
years into her term as Prime Minister, making public references to the ‘very sexist smear
campaign’ circulating against her in the media (Maley 2012). By drawing attention to the
media’s sexist attacks, she began to express something of her ‘murderous rage’ (Gillard and
Summers 2013). In a widely reported press conference, Gillard named and condemned the
‘misogynists’ and ‘nut jobs on the internet’ who continued to produce and disseminate ‘vile
and sexist’ abuse aimed at discrediting her leadership (Grattan 2012). This impassioned
public address presented a dramatic contrast to the ‘sanitised, well-rehearsed’ speeches
that had been scripted by media minders and had been criticised as stilted and inauthentic
(Hargreaves 2010). These robust spontaneous statements recalled the discursive skill of her
earlier, much-admired performances on the floor of parliament and, more importantly, sig-
nalled an agentive turning point in the discursive rendering of Gillard as a credible, coherent
being.
In parallel with this move, an emerging grassroots campaign, operating largely in social
media, began challenging the gendered portrayals of Gillard and other women in leadership
positions by appropriating the sexist language used in the mainstream media broadcasts.
This social media campaign was ignited by radio broadcaster Alan Jones’s claim on 31st
August 2012 that Australian women in positions of political power were ‘destroying the
joint’. In a twitter response using the hashtag #destroyingthejoint, media commentator Jane
Caro mused: ‘Got time on my hands tonight so thought I’d spend it coming up with new
ways of “destroying the joint” being a woman & all. Ideas welcome’ (see Caro 2012). The
tweet elicited hundreds of humorous replies from women sharing how they were ‘destroy-
ing the joint’ and prompted the formation of a popular ‘Destroy the Joint’ Facebook com-
munity that was effective in countering various examples of sexist activities and comments
in the media (see McLean and Maalsen 2013).
A month after the initial #destroyingthejoint tweet, on 8th October 2012, Gillard deliv-
ered an electrifying speech in parliament about sexism and misogyny (Gillard 2012).
Gillard’s speech was initially framed as a response to Tony Abbott’s accusations that she
had hypocritically defended the sexist behaviour of the House Speaker, but was equally an
opportunity to articulate her rage at the deep-seated misogyny she had endured:

I say to the Leader of the Opposition I will not be lectured about sexism and misogyny
by this man, I will not. And the Government will not be lectured about sexism and
misogyny by this man. Not now, not ever.
The Leader of the Opposition says that people who hold sexist views and who are
misogynists are not appropriate for high office. Well I hope the Leader of the Opposition
has got a piece of paper and he is writing out his resignation. Because if he wants to
know what misogyny looks like in modern Australia, he doesn’t need a motion in the
House of Representatives, he needs a mirror. That’s what he needs. Let’s go through
the Opposition Leader’s repulsive double standards, repulsive double standards when it
comes to misogyny and sexism.

With rhetorical flair, Gillard then detailed Abbott’s previous remarks about women’s lesser
appetite and aptitude for leadership. By turning attention to Abbott’s words and behaviour,
Gillard effectively refused the object position proffered by Abbott in his censure motion. In

472
 Women as leaders

doing so, Gillard moved into the subject position, and made Abbott the object of scrutiny;
her labelling of Abbott as a misogynist was strengthened by her confident stance, sweep-
ing arm gestures, and pointing finger (see ABC News 2012). Gillard’s speech then turned
to the ways in which Abbott’s stated views had personally affected her. In so doing, Gillard
identified with a political category defined by gender and, as a woman, situated herself in
‘a group with a distinctive, and distinctively oppressive, history – an ongoing history which
is an appropriate target of social critique and political transformation’ (Stone 2004: 137). In
this pivotal speech, Gillard identified as a woman, and with women, but, at the same time,
refused to conform to an oppressive gender regime that demands, amongst other behav-
ioural norms, an essentialised, passive femininity:

I was very offended personally when the Leader of the Opposition, as Minister of
Health, said, and I quote, ‘Abortion is the easy way out’. I was very personally offended
by those comments. You said that in March 2004, I suggest you check the records. I
was also very offended on behalf of the women of Australia when in the course of this
carbon pricing campaign, the Leader of the Opposition said ‘What the housewives of
Australia need to do – what the housewives of Australia need understand as they do the
ironing’. Thank you for that painting of women’s roles in modern Australia.
And then of course, I was offended too by the sexism, by the misogyny of the Leader
of the Opposition catcalling across this table at me as I sit here as Prime Minister,
‘If the Prime Minister wants to, politically speaking, make an honest woman of her-
self’, something that would never have been said to any man sitting in this chair. I was
offended when the Leader of the Opposition went outside in the front of Parliament and
stood next to a sign that said ‘Ditch the witch’. I was offended when the Leader of the
Opposition stood next to a sign that described me as a man’s bitch. I was offended by
those things. Misogyny, sexism, every day from this Leader of the Opposition. Every
day in every way, across the time the Leader of the Opposition has sat in that chair and
I’ve sat in this chair, that is all we have heard from him.

Gillard repeated, with emphasis, the personal pronoun ‘I’, aligning herself personally and
explicitly with ‘the women of Australia’ in an insurgent position of power rather than sub-
ordination. The authority she adopted in this speaking position was not only the institu-
tionalised authority invested in the discourse of leadership (through her position as Prime
Minister, speaking in parliament), but also the authority – as a ‘woman’ – to name and
condemn the forces of sexist discrimination directed against her. In Butler’s words, this
subversive move demonstrated ‘the performative power of appropriating the very terms by
which one has been abused in order to … derive an affirmation from that degradation …
[and] revaluing affirmatively the category … of “woman”’ (Butler 1997: 158).
Most reporters in the Australian male-dominated parliamentary press gallery interpreted
Gillard’s speech as an act of political opportunism, missing the stronger emotional impact
carried in Gillard’s words and the significance of those words for a wider audience in tune
with gender politics. The political editor of a major newspaper accused Gillard of failing
to meet public expectations that she be a ‘flag bearer for women’; instead, he wrote, she
had proved to be ‘just another politician’ bent on retaining power at any expense (Hartcher
2012). Another columnist described the speech as a ‘bucket of bilgewater’ (Sheehan 2012)
and accused Gillard of revealing her ‘true nature’ by ‘playing the gender card’, a phrase
that was to be taken up in a new round of criticism across the media. Writing for a daily
newspaper owned by Rupert Murdoch, a female columnist opined that ‘Playing the gender

473
Roslyn Appleby

card is the pathetic last refuge of incompetents … It offends the Australian notion of the fair
go’ (Devine 2012). Such commentary implied that it was wrong for a woman in a country
like Australia (that prides itself on egalitarianism yet has deep structural problems of gender
inequality) to be naming and shaming the discourses of sexism and misogyny that continue
to circulate in the public domain.
Yet the significance of the speech was recognised by a more progressive media within
Australia and around the world. In Australia, New Matilda identified this as ‘the most impor-
tant speech’ of Gillard’s prime ministership and applauded the ‘re-emergence of feminism
in public life’ (Eltham 2012). As Eltham observed, the sentiments expressed in the speech
were recognisable by many Australian women who ‘have experienced the dead hand of
misogyny at close quarters, either through sexual harassment, routine sexual vilification in
the workplace, or in the insidious “boys’ club” mentality that still grips many Australian
social environments’. In the USA, Jezebel cheered Gillard’s impassioned ‘smackdown’,
and described her as ‘one badass motherfucker’ (Morrissey 2012); The New Yorker lauded
Gillard’s ‘genuine anger’ and suggested that the ‘real problem’ for the opposition was sim-
ply having a woman ‘running the country’ (Lester 2012). The Guardian recognised that the
event ‘was seen by many women as a defining moment for feminism in the country’ (Rourke
2012). Although Gillard was deposed as Prime Minister within a year, the speech stands as
a testimony to the power of language in politics.

The discursive construction of gender

This brief case study demonstrates how a poststructuralist perspective can reveal a range of
ways in which women in the public domain may be discursively constructed. Through the
use of social category labels and evaluative positioning, Gillard’s rise to power was linguisti-
cally represented in the mainstream Australian media as an act of ruthless personal ambition
that would disappoint those who expected a female leader to display behaviour considered
more appropriate to her gender. Her decisions in office were similarly described and evaluated
according to binary gender norms, with the effect of bringing into question Gillard’s capac-
ity to perform as a leader with both toughness and compassion. As is typical of the attention
applied to women in public domains, her physical appearance became the grounds for further
negative appraisals, and her unmarried and child-free status was referenced to position Gillard
as an abnormal outsider in terms of the appropriate performance of gender and sexuality.
Through the lens of gender normativity, she was discursively portrayed by her detractors as an
unnatural, unintelligible being, and implicitly judged as unfit to govern the nation.
This case has also illustrated how a poststructuralist analysis can expose the operation
of oppositional discourses, such as those articulated in Gillard’s misogyny speech. The dis-
cursive stance taken in this speech demonstrates that women, including those in leadership,
can break institutional silence surrounding issues of discrimination with acts of strategic
essentialism that name and condemn the discourses of sexism and misogyny that circu-
late in politics and the media. These acts of insurrection, more recently apparent in the
#metoo movement, continue to divide the media and are, in turn, matched with expressions
of aggrieved white male entitlement (Kimmel 2013). Gillard’s misogyny speech gave voice
to the rage experienced by many women who have endured sexism in silence and have
observed a global re-emergence of feminism in the political and cultural landscape. In this
sense, a deeply localised, contextually embedded use of language has not only offered a
reconstituted discursive identity for one woman but has also indexed broader social catego-
ries and movements of gender and sexuality in contemporary times.

474
 Women as leaders

Future directions
One future direction for poststructural research is likely to be an expansion of studies that
investigate how language is used in social media to shape, reflect, and contest contempo-
rary politics of gender and sexuality. These practices are arguably most evident in online
campaigns that highlight and denounce violence against women, such as the #metoo move-
ment, and, on the other end of the spectrum, online forums that feature the aggrieved voice
of ‘incels’, that is, people – usually men – who define themselves as involuntarily celibate.
For an insightful commentary on language, gender, and incel discourse, see https://debuk.w
ordpress.com/2018/05/01/is-terrorism-the-right-word/
A second direction for future research has emerged from posthumanism and its relation-
ship with new materialisms. This strand of research is concerned with the ways in which
poststructuralist ideas about the centrality of discourse are augmented by posthumanist
ideas about the liveliness of matter in the more-than-human world. Debates in this area can
be seen in the work of scholars such as Barad (2003), Braidotti (2013), and Bucholtz and
Hall (2016). The implications of feminist posthumanism for language and gender have also
been taken up in the work of Appleby and Pennycook (2017).

Acknowledgements
Thanks go to John Benjamins Publishing Company for permission to reuse parts of my
chapter ‘Julia Gillard: A murderous rage’, which previously appeared in the book Discourse,
Politics, and Women as Global Leaders (edited by Wilson, J. and Boxer D., 2015) https://
benjamins.com/catalog/dapsac.63

Further reading
Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press. Cham, Switzerland:
Palgrave Macmillan.
This book provides a useful guide to discourse analysis and demonstrates, using the tools of
feminist poststructural discourse analysis, how the media constructs leading women in essentialist,
sexualised ways.
Cameron, D. (May 2015 onwards) Language: A Feminist Guide [Web log post]. Available at https://
debuk.wordpress.com (Accessed: 15th February 2019).
This series of regular blog posts is an important scholarly site that offers incisive and witty analyses
of the ways in which language use in contemporary cultural and political events instantiate broader
social and cultural issues of gender inequality and sexism.
Wilson, J. and Boxer, D. (eds.) (2015) Discourse, Politics and Women as Global Leaders. Amsterdam:
John Benjamins.
This edited volume focuses on the discourse practices that pertain to women in global political
leadership and contributes to our understanding of the way language and discourse constitute gendered
identities in political domains. Each chapter investigates a particular national context.

Related topics
Poststructuralist research on language, gender, and sexuality; leadership language of Middle
Eastern women; identity construction in gendered workplaces; leadership and humour at work:
using interactional sociolinguistics to explore the role of gender; analysing gendered discourses
online.

475
Roslyn Appleby

References
ABC News (2012) ‘Gillard labels Abbott a misogynist’. Available at: https://www.youtube.com/watch
?reload=9&v=ihd7ofrwQX0 (Accessed: 11th July 2018).
ABC News (2013) ‘Radio host Howard Sattler asks Julia Gillard if partner Tim Mathieson is gay’.
Available at: https://www.youtube.com/watch?v=flAJmIs1d1I (Accessed: 11th July 2018).
Appleby, R. and Pennycook, P. (2017) ‘Swimming with sharks, ecological feminism and posthuman
language politics’. Critical Inquiry in Language Studies, 14(2–3), pp. 239–261.
Barad, K. (2003) ‘Posthumanist performativity: Toward an understanding of how matter comes to
matter’. Signs: Journal of Women in Culture and Society, 28(3), pp. 801–831.
Baxter, J. (2018) Women Leaders and Gender Stereotyping in the UK Press. Cham, Switzerland:
Palgrave Macmillan.
Braidotti, R. (2013) The Posthuman. Cambridge: Polity.
Bucholtz, M. and Hall, K. (2005) ‘Identity and interaction: A sociocultural linguistic approach’.
Discourse Studies, 4(4–5), pp. 585–614.
Bucholtz, M. and Hall, K. (2016) ‘Embodied sociolinguistics’. In: Coupland, N. (ed.) Sociolinguistics:
Theoretical Debates. Cambridge: Cambridge University Press, pp. 173–197.
Butler, J. (1988) ‘Performative acts and gender constitution: An essay in phenomenology and feminist
theory’. Theatre Journal, 40(4), pp. 519–531.
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. New York, NY: Routledge.
Butler, J. (1997) Excitable Speech: A Politics of the Performative. New York, NY: Routledge.
Cameron, D. (2001) Working with Spoken Discourse. London: SAGE.
Cameron, D. and Kulick, D. (2006) ‘Heteronorms’. In: Cameron, D. and Kulick, D. (eds.) The
Language and Sexuality Reader. London: Routledge, pp. 165–168.
Caro, J. (2012) Cry Havoc and Let Loose the Bitches of War. Available at: https://newmatilda.com/20
12/09/04/cry-havoc-and-let-loose-bitches-war (Accessed: 11th July 2018).
Denmark, D., Ward, I., and Bean, C. (2012) ‘Gender and leader effects in the 2010 Australian election’.
Australian Journal of Political Science, 47(4), pp. 563–578.
Devine, M. (2012) Gender Card is a Loser for Gillard. Available at: http://www.dailytelegraph.com.au
/news/opinion/miranda-devine-gender-card-is-a-loser/story-e6frezz0-1226494961475 (Accessed:
11th July 2018).
Eltham, B. (2012) Gillard Rides the New Wave of Feminism. Available at: https://newmatilda.com/20
12/10/10/gillard-rides-new-wave-feminism (Accessed: 27th June 2018).
Foucault, M. (1972) The Archeology of Knowledge and the Discourse on Language. Trans. A. M.
Sheridan Smith. New York, NY: Harper and Row.
Gillard, J. (2012) Transcript of Julia Gillard’s Speech. Available at: http://www.smh.com.au/federal-
politics/political-news/transcript-of-julia-gillards-speech-20121010-27c36.html (Accessed: 22nd
October 2013).
Gillard, J. and Summers, A. (2013) Julia Gillard and Anne Summers. Available at: http://www.abc.
net.au/radionational/programs/bigideas/julia-gillard-26-anne-summers/4978214 (Accessed: 8th
October 2015).
Grattan, M. (2010) Finessing a Flagrant Backflip. Available at: http://www.smh.com.au/federal-polit
ics/political-opinion/finessing-a-flagrant-backflip-20100625-z9t4.html (Accessed: 22nd October
2013).
Grattan, M. (2012) Misogynists, Nutjobs and Falsehoods: PM Hits Back. Available at: http://www
.theage.com.au/federal-politics/political-news/misogynists-nutjobs-and-falsehoods-pm-hits-back-
20120823-24p8t.html (Accessed: 25th October 2012).
Hall, L. J. and Donaghue, N. (2012) ‘“Nice girls don’t carry knives”: Constructions of ambition in
media coverage of Australia’s first female Prime Minister’. British Journal of Social Psychology,
52(4), pp. 631–647.
Hargreaves, W. (2010) Julia, Dump the Voice Coach. Available at: http://www.heraldsun.com.au/news/op
inion/julia-dump-the-voice-coaching/story-e6frfhqf-1225902500427 (Accessed: 12th October 2013).

476
 Women as leaders

Harrison, D. (2007) I’m Sorry, Heffernan Tells Gillard. Available at: http://www.smh.com.au/news/nat
ional/im-sorry-heffernan-tells-gillard/2007/05/02/1177788206008.html (Accessed: 12th October
2013).
Hartcher, P. (2012) We Expected More of Gillard. Available at: http://www.smh.com.au/federal-polit
ics/political-opinion/we-expected-more-of-gillard-20121009-27bd6.html (Accessed: 12th October
2013).
Hornery, A. (2005) Gillard Bares All. Available at: http://www.smh.com.au/news/Spike/Gillard-bares
-all/2005/01/23/1106415457103.html (Accessed: 11th October 2013).
Keane, B. (2012) Careful! Larry Pickering is Profoundly Offensive. Available at: http://www.crikey
.com.au/2012/04/05/careful-larry-pickering-is-profoundly-offensive/ (Accessed: 25th September
2014).
Kelly, J. (2011) Mark Latham Says Julia Gillard has No Empathy Because She’s Childless. Available
at: http://www.theaustralian.com.au/national-affairs/mark-latham-says-julia-gillard-has-no-emp
athy-because-shes-childless/story-fn59niix-1226033174177 (Accessed: 10th October 2013).
Kimmel, M. (2013) Angry White Men: American Masculinity at the End of an Era. New York, NY:
Nation Books.
Lee, A. and Petersen, A. (2011) ‘Discourse analysis’, in Somekh, B. and Lewin, C. (eds) Theory and
Method in Social Research, 2nd edn. London: Sage, pp. 139–46.
Lester, A. (2012) Ladylike: Julia Gillard’s Misogyny Speech. Available at: http://www.newyorker
.com/online/blogs/newsdesk/2012/10/julia-gillards-misogyny-speech.html?printable=true
&currentPage=all (Accessed: 10th October 2013).
Litosseliti, L. (2006) Gender and Language: Theory and Practice. London: Hodder Arnold.
Maley, J. (2012). With Her Back to the Wall, Real Julia Shows Her Mettle. Available at https://ww
w.brisbanetimes.com.au/politics/federal/with-her-back-to-the-wall-real-julia-shows-her-mettle-2
0120823-24p3p.html#ixzz24PFDIrpe (Accessed: 18th February 2019).
McLean, J. and Maalsen, S. (2013) ‘Destroying the joint or dying of shame? A geography of revitalized
feminism in social media and beyond’. Geographical Research, 51(3), 243–256.
Mills, S. (1997) Discourse. New York, NY: Routledge.
Morrissey (2012) Best Thing You’ll See All Day: Australia’s Female Prime Minister Rips Misogynist a
New One in Epic Speech on Sexism. Available at: http://jezebel.com/5950163/best-thing-youll-see
-all-day-australias-female-prime-minister-rips-misogynist-a-new-one-in-epic-speech-on-sexism
(Accessed: 20th October 2012).
Oakes, L. (2010) Rudd Whacked but Abbott Now Bleeding. Available at: http://www.dailytelegraph
.com.au/news/opinion/rudd-whacked-but-abbott-now-bleeding/story-e6frezz0-1225884444193
(Accessed: 10th October 2013).
Overington, C. (2013) Liberal Menu on Gillard: Small Breasts, Huge Thighs, and Big Red Box.
Available at: http://aww.ninemsn.com.au/news/newsstories/8673507/liberal-menu-on-gillard-
small-breasts-huge-thighs-and-big-red-box (Accessed: 1st July 2013).
Rourke, A. (2012) Julia Gillard’s Attack on Sexism Hailed as Turning Point for Australian Women.
Available at: http://www.theguardian.com/world/2012/oct/12/julia-gillard-sexism-australian-wom
en (Accessed: 20th October 2012).
Sawer, M. (2010) Managing Gender: The 2010 Federal Election. Available at: http://www.australia
nreview.net/digest/2010/10/sawer.html (Accessed: 18th February 2019).
Sawer, M. (2013) Is Australia Still Not Ready for a Female Prime Minister? Available at: https:/
/theconversation.com/is-australia-still-not-ready-for-a-female-prime-minister-15424 (Accessed:
26th February 2019).
Sheehan, P. (2012) Gillard Reveals True Nature in Playing Gender Card. Available at: http://www
.smh.com.au/federal-politics/political-opinion/gillard-reveals-true-nature-in-playing-gender-card
-20121010-27cnq.html (Accessed: 11th October 2012).
Spooner, R. (2013) Shock Jock Suspended for ‘Gay’ Question to PM. Available at: http://www.smh.
com.au/entertainment/tv-and-radio/shock-jock-suspended-for-gay-question-to-pm-20130614-2o6
lm.html (Accessed: 13th June 2019).

477
Roslyn Appleby

Stone, A. (2004) ‘Essentialism and anti-essentialism in feminist philosophy’, Journal of Moral

Philosophy, 1(2). pp. 135–153.
Summers, A. (2012) Her Rights at Work (R-rated Version): The Political Persecution of Australia’s
First Female Prime Minister. Available at: http://annesummers.com.au/speeches/her-rights-at-w
ork-r-rated/ (Accessed: 20th October 2013).
Summers, A. (2013a) The Misogyny Factor. Sydney, Australia: NewSouth Publishing.
Summers, A. (2013b) The Sexual Politics of Power. Available at: http://meanjin.com.au/articles/post/
the-sexual-politics-of-power/ (Accessed: 7th October 2013).
Talbot, M. (2010) Language and Gender, 2nd ed. Cambridge, UK: Polity Press.

478
 Part VII
Semiotic and multimodal
approaches
 32
Gender and sexuality in discourse
Semiotic and multimodal approaches
Michelle M. Lazar (Part VII lead)

Introduction
Multimodality is a field of study in its own right (van Leeuwen 2014). However, it is a field
that is ‘rather fragmented and unconsolidated’ (Machin et al. 2016: 302). Taken together, the
two statements can be read as attesting positively to the recognition, from various quarters
in sociolinguistics and discourse studies, to the fact that meaning-making in texts is a neces-
sarily multiple and heterogeneous semiotic practice. The ‘turn’ towards multimodality (or
sometimes referred to as multisemiotics) in linguistics signals an appreciation for the com-
posite nature of texts, in which the meanings of a text could be realised through the integra-
tion of more than one semiotic mode (Kress and van Leeuwen 2006). As such, the study of
multimodality represents a significant and productive step forward for undertaking a more
holistic analysis of texts and talk. It is a move which widens the semiotic lens of linguists
to accept that written and spoken language are among several modes of meaning and sense-
making in discourses of any kind. The pervasiveness of digitally-mediated communication
nowadays makes the phenomenon of multimodality all the more pronounced and, indeed
inevitable, for discourse participants and discourse analysts alike.
Historically, some perspectives within linguistics have been theoretically predisposed to
the view of a multiplicity of semiosis than others. Social semiotics is one such perspective
(Hodge and Kress 1988; Kress and van Leeuwen 2006; and van Leeuwen 2005) – a body of
work which has taken inspiration from the scholarship of Halliday. Halliday’s (1978) book,
titled Language as Social Semiotic, paved the way for linguists thinking about language as
one, among several, socially embedded semiotic or meaning-making systems; others would
include visual symbols and images, gestures, colour, music, materiality, and so on.
The ‘turn’ towards multimodality in language, gender, and sexuality scholarship, more
specifically, has been comparatively slow and uneven, although the trend is gradually begin-
ning to change, following shifts in wider sociolinguistics and discourse studies. Still, the
relative infancy of multimodal research in language, gender, and sexuality studies seems
to suggest that assumptions about the primacy of language in human communication has
remained strong. This is perhaps quite understandable, given the core orientation of the
field, attested to in the foregrounding of ‘language’ in the name of the field, as well as in

481
Michelle M. Lazar

the titles of related linguistics journals such as Gender and Language and Language and
Sexuality. Widening the lens on semiosis does not eschew or undervalue research about
language(s) and language-use pertaining to gender and sexuality, which are important in
their own right. However, where texts are multimodal, a focus beyond the semiotics of
language alone facilitates a richer and more holistic understanding of the ways gender and
sexuality are represented in discourse. Depending on the contexts and communicative prac-
tices, several different multisemiotic resources could be deployed, among which language
is likely to be one.
In this chapter, five selected ways to carry out multimodal research on gender and sexuality
are discussed under the headings ‘Goffman’s approach to gender representations’, ‘Critical
social semiotic approach’, ‘Approaches to multimodal digital discourses’, ‘Linguistic/semi-
otic landscapes approach’, and ‘Multimodal approaches to conceptual metaphor’. The term
‘approaches’ is used loosely to refer to a range of frameworks, perspectives, and research
foci that has been pertinent to the discursive study of gender and sexuality. In this chapter,
in introducing each of the five strands, two studies shall be selected to briefly describe
how multimodality is used to address a variety of research questions about gender and
sexuality. Among the studies described are the four chapters featured in this section of the
book, namely, under the strands on Goffman, critical social semiotics and multimodal digi-
tal discourses. Although discussion of the five strands will be dealt with separately, the
strands are not mutually exclusive and, in fact, may be combined depending on the nature
of one’s research projects. On a personal note, even though my own multimodal research
falls mainly within critical social semiotics, I have drawn upon the other frameworks and
perspectives as well where relevant in different research projects, and have found them to
be complementary and mutually beneficial.
Before proceeding to outline the five approaches, the concept of ‘affordance’, which
underlies any study of multimodality, needs some explanation. The social semiotic perspec-
tive provides a useful understanding of the concept, which may be applied to multimodal
analyses more generally. Even though all semiotic modes can be deployed for meaning-
making, they have different ‘affordances’ i.e. each has a different set of potentialities and
limitations for meaning-making. Kress and van Leeuwen (2006) explain that semiotic
modes are shaped by the intrinsic characteristics and possibilities of the medium as well
as by the sociocultural values, histories, and communicative purposes that strongly affect
the uses of these potentialities. Not everything that can be expressed in written language,
for example, can be necessarily expressed through visual images, or vice versa. Even when
seemingly the ‘same’ meanings may be expressed in either pictorial form or in writing or
speech, the meanings will be realised differently depending on the medium, which in turn
will affect the actual meanings communicated in context. In multimodal analysis, therefore,
the economy of meaning-making in texts and talk, contributed by various semiotic modes
through their different affordances, becomes pertinent. In what follows, the five approaches
to semiotic and multimodal analyses of gender and sexuality shall be presented.

Semiotic and multimodal approaches to the study of gender and

sexuality
Goffman’s approach to gender representations
Goffman’s Gender Advertisements (1979) presents a classic framework for the study of
visual representations of gender stereotyping in print media. Although some of the findings

482
 Semiotic and multimodal approaches

are now dated and his methodology is controversial (as will be discussed later), his is a
touchstone framework for scholars working on hetero-gendered visual representations, par-
ticularly in advertising. Reflecting the period of study in the 1970s, Goffman identified a
series of sexist representational patterns in the way white, North American women were
depicted vis-à-vis their male counterparts in North American commercial newspaper and
magazine advertisements.
Undergirded by an approach that could be described as social-interactional, a key con-
cept of Goffman’s study is ‘gender display’, which refers to ritual-like portrayals of gen-
dered behaviours and appearances, indexical of male and female social actors’ structural
relationships and identities with regard to each other. Goffman notes that mediatised images
of women (and men) are styled to appear ‘only natural’ (1979: 89) for an audience already
accustomed to such routinised gender expressions in actual social situations. Congruent
with contemporary social constructionist and performative theories, Goffman emphasises
that both curated and actual poses are socially imbibed performances of gender: ‘advertisers
conventionalise our conventions, stylize what is already a stylization’, the only difference
being that mediatised portrayals are ‘hyper-ritualised’ gender displays (1979: 29). In other
words, Goffman is referring to gender representations in advertisements as similar to quo-
tidian performances of socially prevalent hegemonic ideologies, except that in advertise-
ments the displays are even more pronounced.
Goffman’s gender displays are premised upon a mutually constituted hierarchical rela-
tionship between men and women, patterned after parent–child relationships of benign
dominance and subordination, respectively. His analytical framework for interpreting the
social dynamics of asymmetrical gender relations relies on simple, often taken-for-granted
postures and gestures involving social actors’ use of hands, eyes, facial expressions and
head postures, knees, and relative sizing and positioning of bodies. Based on these, he
identified six categories of nonverbal gender display in advertisements: (i) relative size
(women depicted smaller/lower relative to men); (ii) function ranking (men in executive
roles, women in supportive roles); (iii) feminine touch (women’s fingers and hands shown to
lightly caress objects or themselves); (iv) ritualisation of subordination (women in submis-
sive and appeasing postures and gestures); (v) licensed withdrawal (women as psychologi-
cally removed from situations); and (vi) family scenes (women as mothers).
Goffman’s method of data collection, however, has been criticised; instead of drawing
on a random sample, he had selectively chosen data from newspapers and magazines that
would match his specific research objectives (Kang 1997). Notwithstanding this, his study
has been widely adapted by other scholars. Subsequent studies have improved upon his
sampling method, and used his framework to compare his findings with other data sets
across time and audience types (e.g. Hiramoto and Teo 2015; Kang 1997), or combined his
framework with other (social semiotic) approaches (e.g. Bell and Milic 2002; Lazar 2000).
The two studies discussed in this section, Hovland et al. (2005) and Kohrs and Gill (in this
volume), belong to the former category of studies.
Selecting a random (instead of a non-random) sample of advertisements from the highest
circulating US and Korean women’s magazines in the year 2000, Hovland et al. (2005) used
Goffman’s analytical framework to undertake a comparative study of gendered representa-
tions of American advertisements over time, between American and Korean cultures, and
between differently aged women in the US and Korea. Hovland et. al augmented Goffman’s
six analytical categories of gender display by including a few other emergent categories from
the studies of other scholars who had also used Goffman’s framework. Notably, Hovland et.
al incorporated the categories ‘Body Display’ (state of models’ undress) and ‘Independence/

483
Michelle M. Lazar

Self-assurance’ (overall impression of models’ sense of autonomy and poise) introduced by

Kang (1997), and ‘Full Facial Prominence’ (models gazing directly at the viewer) devel-
oped by McLaughlin and Goulet (1999). Their findings revealed that sexist depictions of
women in American magazines had decreased significantly over time, although relative to
Korean magazines, stereotypical images of women in the American sample was found to
be higher. In terms of whether there would be a greater presence of racially diverse models
in ads addressed to younger female audiences than an older target group, Hovland et. al
found the American magazine ads to be ‘disturbingly ethnocentric’ in general. In magazines
addressed to younger Korean women, there was a rising trend of using white Caucasian
models, signalling an emphasis on a white or Euro-American beauty standard. In both cat-
egories then, white women represented normative standards of feminine beauty – a finding
that is consonant with wider feminist literature on the mediatisation of feminine beauty ide-
als (Kilbourne 1999; Wolf 1991).
In using Goffman’s framework, Hovland et. al found that the framework was weighted
predominantly towards negative sexist imagery. In contrast, their study revealed the evolu-
tion of some new positive analytical categories depicting female independence and con-
fidence which warranted inclusion in the study of more contemporary gender displays in
advertising. However, they noted that Goffman’s approach, which focused on visual images
alone, rather than on language as well, proved suitable for undertaking cross-cultural com-
parisons of gender representations. The framework was evidently productive in extending
beyond a focus solely on gender to investigating other intersecting social identity categories
such as race/ethnicity also. In fact, Hovland et. al recommended that race ought to be inves-
tigated more systematically in regard to all of Goffman’s gender display categories in future.
In a similar vein of investigating the extent to which Goffman’s analytical categories
and findings still apply to gender advertisements some 40 years since, Kohrs and Gill’s
study (in this part) is based on 200 advertisements collected from upmarket US and UK
women’s magazines. Using Goffman’s framework as their point of departure, Kohrs and
Gill identify six key units of nonverbal behaviour, which they use as their analytical lens:
gaze, posture, gesture, touch, facial expression, and proxemics. The authors caution though
that, methodologically, one cannot simply derive meaning from the isolated units of body
‘language’; rather, the meanings should be interpreted contextually. When compared to
Goffman’s findings, Kohrs and Gill found that, with the exception of ‘feminine touch’, the
rest of Goffman’s categories did not apply in the present time, especially with regard to
middle- and upper-class Western female audiences. Unlike Goffman’s dataset, Kohrs and
Gill focused on advertisements from women’s magazines only, which featured women pre-
dominantly (rather than women and men), thus rendering some of Goffman’s analytical cat-
egories non-applicable too. Instead, in line with some of the emergent analytical categories
found in Hovland et al.’s (2005) study, Kohrs and Gill noted the rise of a bold and confident
female figure in their advertisements. Referring to this subjectivity as ‘confident appearing’,
it is expressed through a composite of visual signifiers such as an unsmiling direct gaze at
the viewer, a confident stance, heads held high, and striding forward. Referring to this as a
‘postfeminist’ subjectivity, the authors note the representation as a far cry from the diminu-
tive, passive, deferential gender displays of women decades ago.
In sum, these studies, which draw upon Goffman’s framework, use it as a point of depar-
ture to investigate current regimes of media representation. While acknowledging that pock-
ets of overtly patriarchal gender displays remain, newer gender tropes requiring updated
analytical categories are noticed. Also, gender stereotyping in advertising, be that traditional

484
 Semiotic and multimodal approaches

or new, continues to be a productive area of study, especially in regard to intersectional

analyses of race, class, age, sexuality, and culture.
While Goffman’s framework focused solely on visual representations of gender, appli-
cable primarily to advertising data, the next approach, critical social semiotics, facilitates a
multi-semiotic analysis of a wide range of discourse types.

Critical social semiotic approach

In this section, a critical social semiotic approach, which brings together critical discourse
studies and a social semiotic approach to multimodal analysis is discussed. In terms of theoret-
ical orientation, a multimodal social semiotic approach is consonant with a critical discourse
perspective. In fact, the major proponents of multimodal social semiotics, Kress and van
Leeuwen (2006), are also critical discourse analysts. Although another notable social semiotic
framework for visual analysis by O’Toole (1994) exists, Kress and van Leeuwen’s approach
is better known and most frequently used in critical discourse studies (see also van Leeuwen
2014). A critical social semiotic approach (a term which, I suggest, calls attention to the social
semiotic theoretical foundations applied to critical discourse scholarship) is nowadays known
in critical discourse circles also as ‘critical multimodal discourse analysis’ or ‘multimodal
critical discourse analysis’. Unlike Goffman’s framework, critical social semiotics or critical
multimodality typically involve the analysis of language along with a range of other forms of
semiosis, such as visual images, layout, typography, colour, and materiality.
Kress and van Leeuwen’s social semiotic theory of communication offers an explicit and
systematic method for analysing the meanings expressed by syntactic relations between the
people, places, and things depicted in images. Kress and van Leeuwen based their approach
on Halliday’s (1994) notion of ‘metafunctions’, which Halliday had theorised in his sys-
temic functional grammar of language. The metafunctions refer to three basic ways of using
language to communicate, which Halliday termed ‘ideational’ (representing our experience
of the world), ‘interpersonal’ (enacting social interactions), and ‘textual’ (organising mes-
sages within the text and in relation to the wider context) meanings. Kress and van Leeuwen
(2006) adapted the three metafunctions to apply to all semiotic modes, not only language.
Accordingly, therefore, any semiotic mode can typically represent objects and their relations
in the world as experienced by people; project particular kinds of social relation and truth
values (modality) between the producer, the viewer, and the object represented; and make
connections within the text, and between the text and its social context. Central to systemic
functional theory is the notion of a system or network of semantic choices, i.e. within each
metafunction is a set of choices from which producers of texts select different ways to make
meaning. The particular choices made in and across semiotic modes in texts and talk are not
neutral but can convey ideological meanings.
In language and gender studies, the usefulness of incorporating multisemiotic analysis
into a critical discourse perspective was highlighted in an article titled ‘Gender, discourse
and semiotics’ (Lazar 2000). In it was proposed

the uncoupling of the two categories “discourse” and “language” in favour of discourse
encompassing semiosis of various kinds (including language). A critical analysis of
discourse, in other words, involves a commitment to the analysis of various strands
of semiosis that configure in the realisation of particular discoursal meanings in texts.
(Lazar 2000: 177)

485
Michelle M. Lazar

More recently, a call for critical multimodality research on gender, language, and discourse
was made by Machin et al. (2016) in a special issue of the journal Gender and Language
(vol. 10, no. 3), which featured a collection of articles using the approach.
The studies by Caldas-Coulthard and McLoughlin, in Part VII, draw on a multimodal
social semiotics approach in their (feminist) critical discourse analytical (CDA) perspective.
Their two chapters indicate some of the types of research pursued within feminist CDA,
namely, sexist ideological representations of women as well as ‘newer’ popular postfemi-
nist discourses (see Lazar 2018). Caldas-Coulthard’s chapter deals with gender stereotyping
in representations of women criminals in public discourses. Her analysis of photographic
images from the Getty image bank and the language and images in the Brazilian press, espe-
cially of white Brazilian women, reveal the sexualisation of women criminals, even though
the crimes are not sex-related. The intertwining of sexuality and criminality is achieved
representationally through a number of semiotic choices like visual modality and ‘demand’
images (represented participants looking directly at the viewer), fragmentation of the female
body and sexualised poses, and the colour of clothing construed as indexing danger and sex-
iness. The emphasis on sexuality in the Brazilian news reports, Caldas-Coulthard observes,
is intended to titillate readers, while at the same time to judge and condemn the women
criminals even more harshly. More generally, she finds that when women and men are por-
trayed differently for similar kinds of criminal acts, women offenders are doubly condemned
for their criminality and for their sexuality, or simply for the fact that they are women.
In contrast with Caldas-Coulthard’s investigation of overt sexist ideology, McLoughlin’s
chapter examines a purportedly progressive feminist discourse promoted in the commercial
media which, from a feminist critical discourse perspective, proves ideologically problematic
also. McLoughlin’s study shows how articles in a high-fashion magazine (Vogue) and an adver-
tising campaign for a cosmetics brand by a pharmaceutical company (Boots Walgreen Alliance)
marketed Chimamanda Ngozi Adichie, a Nigerian author, as a feminist (or, I would suggest, a
postfeminist) sensation in the West. By analysing the media discourses over a period of time,
McLoughlin reveals the representational transformation of Adichie from an award-winning
author, a popular feminist icon arising from her TED talk on feminism, to a cosmetics brand
ambassador. In the process, the focus gradually shifted from Adichie’s writing to her appear-
ance, as evidenced through her verbal represented discourse and through visual semiotic choices
of modality and depictions of background settings and colourful eye-catching clothing, which
simultaneously indexed trendiness as well as her African heritage. The transformation is not sim-
ply a commercial exploitation of Adichie; rather, her ‘brand’ of neoliberal (post) feminism based
on notions of individual empowerment coupled with self-professed love for make-up and fashion
are in sync with the media industries’ predisposition towards a non-threatening ‘saleable’ femi-
nism. While the Adichie sensation admittedly contributed towards mainstreaming feminism, her
participation in commodity feminism does a disservice as well. As McLoughlin notes, Adichie’s
make-up–wearing stance presented as a feminist act of resistance, ironically, does nothing to
address restrictive media ideals of beauty, while painting feminism in rather narrow terms.
With the arrival of new digital media technologies, and their affordances, approaches to
multimodal investigations of gender and sexuality on social media have arisen, to which we
turn our attention next.

Approaches to multimodal digital discourses

Although digital discourse studies is not a new development – known as ‘computer-medi-
ated discourse analysis’ (Herring 2001) in an earlier period – there has been a surge of

486
 Semiotic and multimodal approaches

interest in this field over the past decade, with the rapid development of digital technologies
(collectively referred to as Web 2.0) (e.g. Barton and Lee 2013; Myers 2010; Tannen and
Trester 2012; Zappavigna 2012). The advent of newer technologies has brought about inno-
vative possibilities for representation and has spawned an intensely interactive participatory
online culture through affordances of generating, networking, and sharing user-generated
content. Much of this research has focused on language on social media platforms, although
studies on visual semiotics and multimodality have begun to emerge more recently (Danesi
2017; Jones et al. 2015; Page 2019, for example).
Similarly, in the field of language, gender, and sexuality, studies in digitally mediated
contexts have focused primarily on language (e.g. Herring 1995, 2003; Mackenzie and Zhao
2020), with a few studies emerging recently on visuality and multimodality of gender and
sexuality on social media (e.g. Thurlow 2017; Zappavigna and Zhao 2017). Given the diver-
sity of types of online discourse and the technological and semiotic affordances, there is no
singular methodological approach for the analysis of digital discourses; instead, scholars
tend towards mixed methodologies (Bolander and Locher 2014). Such is the case, also,
for the two studies discussed in Part VII. As the studies show, digitally mediated contexts
provide individuals, as well as social groups, with a range of technological and semiotic
affordances for performing, sharing, and contesting gender and sexual identities in different
and complex ways
Leppänen and Tapionkaski (in this part) examine the identity work of young men in par-
ticipatory online media culture. Specifically, the authors are interested in how participants
mobilise a range of semiotic resources to perform and stylise gender, sexual, and other
social identities of the self and others on social media platforms. Adopting a combined
discourse analytical and ethnographic approach, Leppänen and Tapionkaski report on the
complex ways identities are constructed by different groups of men in two of their previous
studies. In one of these studies, the performance of a young alternative masculinity termed
‘bronies’ is presented on a discussion forum for young adult male fans of a children’s televi-
sion series targeted at young girls. Through a combination of language styles, verbal narra-
tives, emoji, and images, the multifaceted, lived experiences of the ‘brony’ fan identity at
the intersection of gender, sexuality, and age is enacted, which eschews hegemonic Finnish
masculinity. Reporting on another study, the authors discuss the multimodal social media
practices of young migrant men in Finland. In this study, self-crafted video narratives by
migrant men, shot through smartphones or video cameras, are disseminated on such sites
as YouTube and Vimeo. The videos, which are parodic in nature, deploy semiosis such as
embodied performances, multilingual language practices, cinematic devices, and music to
construct narratives of ‘selves’ and Finnish hostile ‘others’ at the intersection of race, ethnic-
ity, nationality, gender, and sexuality. Leppänen and Tapionkaski’s analyses of intersectional
identities enacted through the multisemiotic affordances on digital media transcend binary
identity categories such as female–male, feminine–masculine, child–adult, heterosexual–
asexual, non-Finnish–Finnish, and humorous–critical, in showing the ways social media
participants consciously transgress norms.
While Leppänen and Tapionkaski focus on how individuals using social media plat-
forms construct identities and interact online, Lazar’s (2020) study deals with the mul-
timodal digital discourse of an LGBTQ social movement in Singapore. Known as Pink
Dot SG (Singapore), the social movement regularly uploads promotional videos featur-
ing Singaporean LGBTQ issues and announcing its annual mass gathering, on the video-
sharing website YouTube for national and transnational consumption. The study examines
the multimodal construction of a homonationalist discourse in the videos. Although in its

487
Michelle M. Lazar

original sense ‘homonationalism’ refers to the instrumental mobilisation of queer identities

to serve depoliticised neoliberal agendas of (particularly Western) nation-states, Lazar’s
study takes a different spin on the concept, which shows how a disenfranchised social group
consciously aligns its queer identity with nationalistic values as a resistive (albeit, assimila-
tionist) political strategy. The enmeshing of sexual/gender non-normativity and nationalism
in Pink Dot’s videos is achieved multi- and inter-semiotically through the co-deployment of
spoken and written language, nationalistic songs, colour symbolism (pink to simultaneously
signify national and gay identities), visual images, photographic footages of the national
flag, national indexicals (national institutions that index nation-ness), iconic urban city-
scapes of Singapore, and embodied performances (or ‘corporeal semiotics’) involving the
aggregation of citizens to form a human ‘pink dot’.
In the next section, our focus shifts to how linguistic landscape approaches that draw on
semiotic and multimodal resources are mobilised for the critical investigation of gender and
sexuality concerns.

Approaches to linguistic/semiotic landscapes

Linguistic landscapes has been a burgeoning field of research for more than over a decade,
drawing on a wide variety of theoretical and methodological approaches that can be quan-
titative as well as qualitative in nature. Shohamy and Ben-Rafael (2015: 1) define linguistic
landscapes as ‘research about the presence, representation, meanings and interpretation of
languages displayed in public places’. Landry and Bourhis’ (1997) work is regarded as the
first major effort to refer to publicly displayed signs (e.g. road signs, advertising billboards,
street and place names, signs on shops and other buildings) as constitutive of the linguistic
landscape of any given place. Over the years, the field has developed in a number of ways.
Although the display of different languages in physical environments has been an important
focus of linguistic landscape scholarship, the scope has widened beyond a focus on multi-
lingualism, language, and ‘languaging’ to study a range of multimodal representations, arte-
facts, and materialities in public spaces. In other words, the term ‘linguistic landscapes’ has
now been extended to cover multi-semiotic landscapes as well. It could be argued that the
terms ‘semiotic landscapes’ (Jaworski and Thurlow 2010) or ‘semioscapes’ (Lazar 2018)
might be more appropriate or accurate descriptors to highlight the role played by all forms
of semiosis in public spaces, whereby language (as one semiotic system) may or may not
play a prominent role as the particular case may be. Whichever is the preferred terminol-
ogy, linguistic/semiotic landscapes is not only about the study of semiotics in place but how
spaces are themselves performed semiotically i.e. spaces accrue particular social meanings
through the process of semiosis (Lazar 2018; Milani 2014). Places and spaces, too, have
been widened in current linguistic/semiotic landscape studies to encompass the internet as a
virtual linguistic landscape (e.g. Biró 2018; Ivkóvic and Lotherington 2009) as well as the
body as a corporeal landscape (Peck and Stroud 2015).
Contributing to the ongoing development of the field in recent years, scholars have begun
to call attention to the gendered and sexualised nature of semioticised public spaces. A spe-
cial issue on ‘Gender, sexuality and linguistic landscapes’ edited by Milani (2018) in the
Journal of Linguistic Landscapes is a case in point (see also Kerry 2017; Poseiko 2016).
These studies not only contribute to linguistic/semiotic landscape studies, but enrich the
field of language, gender, and sexuality by showing how power relations undergirding social
orders of gender and sexuality structure, maintain, and/or contest public spaces semiotically.
The two studies that will be described demonstrate this well.

488
 Semiotic and multimodal approaches

In a study about storefront signages (a conventional object of study in linguistic/semi-

otic landscape studies), which combine words and symbols, Trinch and Snajdr (2018)
focus on how gentrifying women (mostly highly educated, professional, white, and
married) struggle to claim and transform the public space in Brooklyn, USA. Drawing
on visual ethnography, interviews, and digital archival material, the study looks at the
innovative wordplay found on shop signs catered to consumers who are mothers. The
authors argue that through the creative signages, Brooklyn mothers attempt to transform
traditional notions of motherhood, conservative heterosexual mores, conventional gen-
der roles, as well as the linguistic landscape of the urban space itself. For example, a
storefront signage of a local store – ‘boing boing: at your cervix since 1996’ – creatively
alludes to sexual activity and female reproduction through wordplay as well as filling
in the interior of the two ‘o’ graphemes in ‘boing boing’ so that visually the letters
resemble cervical caps or diaphragms. By taking to the streets traditionally private and
taboo topics, the women are seen as asserting control over their sexuality and mother-
ing. However, as Trinch and Snajdr argue, such attempts were met with public back-
lash, prompting pragmatic trade-offs in later developments of signages. For instance,
the authors note that the ambiguity of wordplay led to the production of a signage with
racialised undertones, used as a cover by white women to challenge normative mother-
hood. The crux of the study is that the signages represented more than a semiotics of
creativity; they were, in the words of the authors, ‘formidable social acts’ in a struggle
over space and power.
In a different study, Milani (2014) focused on a small data set of ‘banal sexed signs’ or
mundane semiotic aggregates, which he had observed in three separate cities. Adopting
queer theory as his theoretical lens, Milani discusses the operations of power, in relation to
gender and sexuality in unassuming public spaces. Queer theory problematises any form
of normativity, including naturalised configurations of sex, gender, and sexuality, and as
Milani’s study shows, opens up interpretative ambiguity to some extent. A newsstand at a
US airport constituted Milani’s first set of data, in which he deconstructed the ideology of
gendered difference made manifest through the spatial arrangement of magazines aimed at
women and men, and their respective magazine front cover images. Interpellating brows-
ers in distinctly gendered terms, Milani argues that the visual representations of the models
on the front pages could potentially offer a ‘queer hope of ambiguous desire’. The erotic
ambiguity, however, was quickly dispelled through the heteronormative content of the
magazines. The second data involved the window display of two T-shirts with affective
statements (‘I love [x]’) printed on them in a retail store in Stockholm. From the outside of
the shop, the words and symbols on the two T-shirts conveyed a distinctly heteronormative
gender ideology. Yet, when browsing inside the shop, a more dynamic affective process is
activated through the availability of T-shirts (albeit less visibly placed) that catered to non-
heterosexual desiring persons. Moving from an investigation of material objects in public
spaces to store owners own ‘take’ on space, politics, identity, and desire in relation to gender
and sexuality, Milani’s final data involved signage in a coffee shop in a Johannesburg suburb
as well as interviews with the co-owner of the shop. The signage at the entrance of the coffee
shop spelt out a list of ideologically intolerant types of persons disallowed on the premises,
aimed at fostering a ‘queer space’ of respectful convivial debate. The queerness of the space
was intended by the owners to transcend a reductively ‘gay and lesbian’ space, even though
the owners self-identified as lesbians.
The final approach dealt with in this chapter that is useful for gender and sexuality
research is multimodal conceptual metaphor analysis.

489
Michelle M. Lazar

Multimodal approaches to conceptual metaphor

Unlike popular understandings of metaphor as a literary device using words deliberately
to achieve some artistic or rhetorical effect, a cognitive linguistic approach to metaphors,
originally developed by Lakoff and Johnson (1980), views metaphors as conceptual in
nature, used to facilitate understanding of abstract and unfamiliar ideas in ordinary eve-
ryday situations. In fact, conceptual metaphor scholars regard metaphors as a pervasive
process in human thought, action, and everyday communication (Kövecses 2002; Lakoff
and Johnson 1980). According to conceptual metaphor theory, one conceptual domain is
understood in relation to another conceptual domain. For example, in life is a journey,
‘life’ is the ‘Target’ domain in need of comprehension (something abstract and complex)
which is understood in terms of a ‘journey’, the ‘Source’ domain (something more familiar)
that helps make sense of the former. The process by which metaphoric comprehension is
achieved is called ‘mapping’, whereby elements from the Source domain are transferred
onto the Target domain, usually systematically and unconsciously. Importantly, conceptual
metaphors not only facilitate understanding of a more abstract domain but also mediate and
structure the experience of it (Knowles and Moon 2006; Kövecses 2002).
Linguistic, or the verbal, is the most familiar semiotic mode of metaphoric expres-
sion. However, as metaphor (following the cognitive approach) is conceptual in nature, it
can be expressed nonverbally and multimodally as well (e.g. Forceville 1998; Kövecses
2002). Regardless of whether it is verbal or nonverbal, if only one mode is deployed
for metaphoric expression, it is regarded as a monomodal metaphor; whereas in multi-
modal metaphors the Target and Source domains are predominantly presented in different
modes, which can include visuals, written/spoken language, gestures, nonverbal sounds,
and music (Forceville 2009).
In language, gender, and sexuality studies, a focus on verbal metaphors in discourse has
been more common (e.g. Hobbs 2013; Koller 2004; Velasco-Sacristan and Fuertes-Olivera
2006) than multimodal metaphors. The latter, in fact, remains as yet an untapped research
approach, which can be better utilised for investigating topics on discourse, gender, and
sexuality. The two studies discussed in this section represent one of the earliest in multi-
modal metaphor analysis involving gender and advertising (Lazar 2009) and a more recent
one on sex education material for preschool children (Liang et al. 2017). The studies will be
discussed in reverse order.
Drawing on Forceville’s study of pictorial and multimodal metaphors, Liang, O’Halloran,
and Tan (2016) examine metaphorical representations in sex education picture books aimed
at preschoolers in mainland China. The use of metaphors in this educational discourse genre
ostensibly served the pedagogical objective of facilitating better understanding of scien-
tific knowledge about sex and sexuality in a young audience. Yet, the authors argue that
through the metaphorical representations deployed in the books, ideologies about gender
stereotypes and traditional mainstream values on (heteronormative) sex and sexuality were
being instilled in the young at the same time. One type of metaphor based on personifica-
tion grafted male or female faces and traits onto the target domain concepts of reproductive
organs, sperm cells, and ova, so that binary sex categories and gender stereotyping became
naturalised. Another type, identified as domestication metaphors, facilitated understanding
of distant and abstract biological concepts and processes by associating them with familiar
and concrete objects within the worldview of children. However, the selection of source
domains was sometimes ideologically laden (for example, the ovum was verbally and vis-
ually metaphorised as a treasure quest). Finally, the authors identified ‘cross-experience’

490
 Semiotic and multimodal approaches

metaphors, which were established multimodally through what the authors describe as
mind–body associations. In this way, conjugal love and procreation were foregrounded and
prioritised as the driving force for the physical activity of sexual intercourse.
Set against a general observation that a ‘war mentality’ was becoming a prevalent com-
monsensical mode of rationality, through the utilisation of the domain of war to conceptu-
alise a wide range of non-militarised activities, Lazar’s (2009) study focused on how the
domain of banal feminine beauty practices has been metaphorised in terms of warfare in
advertising discourse. Combining a critical discourse perspective with multimodal meta-
phor analysis, the study based on over 100 print beauty advertisements (ads for cosmetics,
skin and haircare products, slimming and body modification services), uncovered the opera-
tion of the conceptual metaphor beautification is war, expressed through a combination
of language, images, and colour. In tandem with a problem–solution schema commonly
found in advertising, a multi-tiered analysis of the conceptual metaphor was systemati-
cally presented. ‘Problems’ were anything that hindered the achievement of the beauty ideal
(e.g. nature or ageing), which were conceptualised as enemies. ‘Solutions’ were cosmetic
brands that occupied the role of powerful allies to prospective consumers. The ‘consumers’
were women conceptualised both as fighters in the struggle and whose bodies, at the same
time, were the battlegrounds. The study revealed contradictory elements in the construction
of ‘modern’ femininity. While on the one hand, women were represented as ‘empowered’
subjects, on the other hand, militarisation associated women’s exercise of agency with a
hegemonic mode of masculinity. With women’s own bodies cast as sites of struggle, and
a concomitantly radical shift in conceptualisation from seeing the ‘enemy-as-other’ to the
‘enemy-as-self’, the study suggested the possibility that anxiety and alienation in women’s
relationships with their own bodies as threats was exigent.

Conclusion
In this chapter, five approaches or lenses for the study of semiotics and multimodality were
presented. These are by no means comprehensive; rather, they were selected to highlight
productive ways for analysing semiotic representations and performances of gender and
sexuality in a range of contexts. As the studies discussed in this section showed, perpetu-
ation as well as contestation and subversion of gender/sexual norms and identities were
enabled through the mobilisation of a number of semiotic resources (with language play-
ing a salient or not so salient part, depending on the situations). The range of semiosis
involved included gestures, colour, visual images and photography, music and song, dress
and props, materialities, represented affect, typography, and embodied performances. Each
contributed to the meaning-making in texts and talk, in an integrated way, based on their
affordances.
Although the turn towards semiotics and multimodality in language, gender, and sexu-
ality scholarship, on the whole, is at the stage of relative infancy, it is slowly but surely
growing. Fragmentation and unconsolidation of approaches, therefore, can be viewed posi-
tively as a healthy sign of growth and maturation of any field. Messiness and mixing of
approaches, I suggest, disavow foreclosing the development of diverse ideas and insights,
and offer opportunities for learning, borrowing, and integrating from different perspec-
tives. Multimodality itself, as a field, is a product of (multi-disciplinary) learning from such
fields as linguistics, semiology, proxemics, art history, cultural studies, and psychology. If
we extend the concept of affordances beyond semiotic modalities to semiotic and multi-
modal approaches, then we can say that each framework or lens has its own (theoretical and

491
Michelle M. Lazar

practical) affordances for apprehending the phenomena of multi-semioticity. That, I believe,

can only enhance the robustness and diversity of inquiries about gender, sexuality, and other
identities in texts and talk.

References
Barton, D. and Lee, C. (2013) Language Online: Investigating Digital Texts and Practices. London:
Routledge.
Bell, P. and Milic, M. (2002) ‘Goffman’s gender advertisements revisited: Combining content analysis
with semiotic analysis’. Visual Communication, 1(2), pp. 203–222.
Biró, E. (2018) ‘More than a facebook share: Exploring virtual linguistic landscape’. Acta Universitatis
Sapientine, Philologica, 10(2), pp. 181–192.
Bolander, B. and Locher, M. A. (2014) ‘Doing sociolinguistic research on computer-mediated data: A
review of four methodological issues’. Discourse, Context and Media, 3, pp. 14–26.
Danesi, M. (2017) The Semiotics of Emoji: The Rise of Visual Language in the Age of the Internet.
London: Bloomsbury.
Goffman, E. (1979) Gender Advertisements. Cambridge, MA: Harvard University Press.
Forceville, C. (1998) Pictorial Metaphor in Advertising. London: Routledge.
Forceville, C. (2009) ‘Non-verbal and multimodal metaphor in a cognitivist framework agendas for
research’. In: Forceville, C. and Urios-Aparisi, E. (eds.) Multimodal Metaphor. Berlin: Mouton de
Gruyter, pp. 18–42.
Halliday, M. A. K. (1978) Language as Social Semiotic. London: Arnold.
Halliday, M. A. K. (1994) ‘Systemic Theory’, in Asher, R. E. (ed) Encyclopedia of Language and
Linguistics, vol 8. Pergamon Press.
Herring, S. (1995) ‘Men’s language on the internet’. Nordlyd, 23, pp. 1–20.
Herring, S. (2001) ‘Computer-mediated discourse’. In: Schiffrin, D., Tannen, D., and Hamilton, H.
(eds.) The Handbook of Discourse Analysis. Oxford: Blackwell, pp. 612–634.
Herring, S. (2003) ‘Gender and power in on-line communication’. In: Holmes, J. and Meyerhoff, M.
(eds.) Handbook of Language and Gender. Malden, Oxford: Blackwell, pp. 202–228.
Hiramoto, M. and Teo, C. S. L. (2015) ‘Heteronormative love makes a house a home: Multimodal
analysis of luxury housing advertisements in Singapore’. Journal of Language and Sexuality, 4(2),
pp. 223–253.
Hobbs, P. (2013) ‘Fuck as a metaphor for male sexual aggression’. Gender and Language, 7(2), pp.
147–174.
Hodge, R. and Kress, G. (1988). Social Semiotics. Cambridge: Polity.
Hovland, R., McMahan, C., Lee, G., Hwang, J. S., and Kim, J. (2005) ‘Gender role portrayals in
American and Korean advertisements’. Sex Roles, 53(11/12), pp. 887–899.
Irkóvic, D. and Lotherington, H. (2009) ‘Multilingualism in cyberspace: Conceptualizing the virtual
linguistic landscape’. Journal of Multilingualism, 6(1), pp. 17–36.
Jaworski, A. and Thurlow, C. (2010) ‘Introducing semiotic landscapes’. In: Jaworski, A. and Thurlow,
C. (eds.) Semiotic Landscapes: Language, Image, Space. London: Continuum, pp. 1–40.
Jones, R., Chik, A., and Hafner, C. (eds.) (2015) Discourse and Digital Practices: Doing Discourse
Analysis in the Digital Age. London: Routledge.
Kang, M. E. (1997) ‘The portrayal of women’s images in magazine adverts: Goffman’s gender analysis
revisited’. Sex Roles, 37(11/12), pp. 979–993.
Kerry, V. (2017) ‘The construction of hegemonic masculinity in the semiotic landscape of a CrossFit
“Cave”’. Visual Communication, 16(2), pp. 209–237.
Kilbourne, J. (1999) Can’t Buy My Love: How Advertising Changes the Way We Think and Feel. New
York, NY: Simon & Schuster.
Knowles, M. and Moon, R. (2006) Introducing Metaphor. London: Routledge.
Koller, V. (2004) Metaphor and Gender in Business Media Discourse: A Critical Cognitive Study.
Basingstoke: Palgrave Macmillan.

492
 Semiotic and multimodal approaches

Kövecses, Z. (2002) Metaphor: A Practical Introduction. Oxford: Oxford University Press.

Kress, G. and van Leeuwen, T. (2006) Reading Images: The Grammar of Visual Design, 2nd ed.
London: Routledge.
Lakoff, G. and Johnson, M. (1980) Metaphors We Live By. Chicago: University of Chicago Press.
Landry, R. and Bourhis, R. Y. (1997) ‘Linguistic Landscape and Ethnolinguistic Vitality: An Empirical
Study’, Journal of Language and Social Psychology, 16(1), pp. 23–49.
Lazar, M. M. (2000) ‘Gender, discourse and semiotics: The politics of parenthood representations’.
Discourse and Society, 11(3), pp. 373–400.
Lazar, M. M. (2009) ‘Gender, war and body politics: A critical multimodal analysis of metaphor
in advertising’. In: Ahrens, K. (ed.) Politics, Gender and Conceptual Metaphors. Basingstoke:
Palgrave, pp. 209–234.
Lazar, M. M. (2018) ‘Gender, sexuality and semioscapes’. Linguistic Landscape, 4(3), pp. 320–327.
Lazar, M. M. (2020) ‘Semiotics of homonationalism’. In: Hall, K. and Barrett, R. (eds.) Oxford
Handbook of Language and Sexuality. Oxford: Oxford University Press.
Liang, J. Y., O’Halloran, K., and Tan, S. (2016) ‘Where do I come from? Metaphors in sex education
picture books for young children in China’. Metaphor and Symbol, 31(3), pp. 179–193.
Machin, D., Caldas-Coulthard, C., and Milani, T. (2016) ‘Doing critical multimodality in research on
gender, language and discourse’. Gender and Language, 10(3), pp. 301–308.
Mackenzie, J. and Zhao, S. (eds) (2020) ‘Special issue - Doing motherhood online: Parenting, identity
and digital interaction’, Discourse, Context, & Media, 37/38.
McLaughlin, T. L. and Goulet, N. (1999) ‘Gender Advertisements in Magazines Aimed at African
Americans: A Comparison to Their Occurrence in Magazines Aimed at Caucasians’, Sex Roles,
40, pp. 61–71.
Milani, T. (2014) ‘Sexed signs – Queering the scenery’. International Journal of the Sociology of
Language, 228, pp. 201–225.
Milani, T. (ed) (2018) Gender, Sexuality and Linguistic Landscapes. Special Issue: Journal of
Linguistic Landscapes. Amsterdam: John Benjamins.
Myers, G. (2010) The Discourse of Blogs and Wikis. London: Continuum.
O’Toole, M. (1994) The Language of Displayed Art. London: Leicester University Press.
Page, R. (2019) ‘Group selfies and snapchat: From sociality to synthetic collectivisation’. Discourse,
Context and Media, 28, pp. 79–92.
Peck, A. and Stroud, C. (2015) ‘Skinscapes’. Linguistic Landscape, 1(1/2), pp. 133–151.
Poseiko, S. (2016) ‘Images of women in the semiotic landscape of the Baltic states’. Miscellanea
Anthropologica et Sociologica, 17(3), pp. 111–128.
Shohamy, E. and Ben-Rafael, E. (2015) ‘Introduction – Linguistic landscape: A new journal’.
Linguistic Landscape, 1(1/2), pp. 1–5.
Tannen, D. and Trester, A. M. (eds.) (2012) Discourse 2.0: Language and New Media. Washington,
DC: Georgetown University Press.
Thurlow, C. (2017) ‘Forget about the words? Tacking with language, media and semiotic ideologies of
digital discourse: Case of sexting’. Discourse, Context and Media, 20, pp. 10–19.
Trinch, S. and Snajdr, E. (2018) ‘Mothering Brooklyn: Signs, sexuality, and gentrification under
cover’. Linguistic Landscape, 4(3), pp. 214–237.
Van Leeuwen, T. (2014) ‘Critical discourse analysis and multimodality’. In: Hart, C. and Cap, P. (eds.)
Contemporary Critical Discourse Studies. London: Bloomsbury, pp. 281–297.
Van Leeuwen, T. (2005) Introducing Social Semiotics. London: Routledge.
Velasco-Sacristan, M. and Fuertes-Olivera, P. (2006) ‘Towards a critical cognitive-pragmatic approach
to gender metaphors in advertising English’. Journal of Pragmatics, 38(11), pp. 1982–2002.
Wolf, N. (1991) The Beauty Myth: How Images of Beauty are Used against Women. New York, NY:
Anchor Books.
Zappavigna, M. (2012) Twitter and Social Media: How We Use Language to Create Affiliation on the
Web. London: Continuum.
Zappavigna, M. and Zhao, S. (2017) ‘Selfies on “mommyblogging”: An emerging visual genre’.
Discourse, Context and Media, 20, pp. 239–247.

493
 33
Multimodal constructions of
feminism
The transfiguration of Chimamanda Ngozi
Adichie in Vogue

Linda McLoughlin

Introduction
In this chapter I outline a technique which lies at the interface of feminist critical discourse
analysis and social semiotics. Its focus is on how language and other modes of communi-
cation combine to create meaning in two multimodal texts: articles in Vogue, a high-end
fashion magazine1 and an advertising campaign, titled ‘Ready to speak up’, launched by
the international pharmaceutical company Boots Walgreen Alliance to promote its cosmet-
ics brand No. 7 Match Made range.2 Both texts feature the award-winning Nigerian author
Chimamanda Ngozi Adichie whose TED talk,3 ‘We should all be feminists’, received much
critical acclaim and established Adichie as a leading feminist, according to popular media
sources. The chapter will examine the global marketing of this contemporary feminist icon
and consider exactly what discourses (ways of seeing) of feminism are being globally dis-
tributed and how. Following Goankar and Povinelli (2003),4 I explore the matrix of elements
that enable the transnational circulation of Adichie as a leading feminist icon. The aim is to
examine the significations assigned to Adichie in these multimodal texts in order to gauge
how her subject position (i.e. the way she is positioned by the text) is defined in relation to
feminism. It will also make visible how feminist work has provided a way of understanding
‘postfeminist’ female subjectivity as mediated in mass-media texts.
The chapter is divided into four sections. In the first, the appropriation of feminism in
popular culture is traced with reference to relevant literature. Next, the feminist critical
discourse analysis (CDA)/social semiotics approach taken is set out. The third and largest
section comprises the analysis, organised around the transformation of Adichie in Vogue
from award-winning author to feminist icon whose model-like looks and self-avowed love
of make-up helped her to secure the lucrative cosmetics deal. The analysis is contextual-
ised by attempting to pinpoint the shift in subject positioning to the period when Adichie
was announced as the ‘new face’ of Boots No. 7. Finally, there is a discussion which pulls
together the main findings in relation to constructions of feminism in popular culture and
argues that Adichie’s particular ‘brand’ of feminism was promoted because it resonates with

494
 Multimodal constructions of feminism

investments in feminism which govern commodified feminism, especially in Western loca-

tions. As Goldman (1992: 131) explains:

The pun, commodity feminism, is a reminder that commodity relations turn the relations
of acting subjects into the relations between objects. The process of turning feminism
into sign values fetishises feminism into an iconography of things. When advertisers
appropriate feminism, they cook it to distil out a residue – an object: a look, a style.
Women’s discourses are thus relocated and respoken by named objects like Hanes
Hose, Nike Shoes, Esprit Jeans. Sign objects are thus made to stand for, and made
equivalent to feminist goals of independence and professional success. Personality can
be represented, relationships achieved and resources acquired through personal con-
sumer choices.

Here, Goldman highlights the fact that producers of commodities have just one goal in sight,
namely the sale of goods; therefore any apparent headway in improving women’s position
in society is merely a patina of benevolence.

The rise of commodity feminism in popular culture

This section aims to establish how commodified feminism, as represented by Adichie, is cir-
culated and transfigured in a transnational context. Questions are considered relating to how
ideals of commodity feminism shift as well as entrench, include, and exclude multicultural
cosmopolitan feminism(s). I want to clarify here that I am not against make-up wearing,
nor do I have anything against Adichie; indeed, I admire her writing and support many of
the widely quoted endorsements of ‘We should all be feminists’. What I find problematic
is the amount of time and energy spent in asserting a connection between feminism, make-
up wearing, and fashion. It seems the politics of feminism are transfigured in commodity
feminism by foregrounding selected key tenets, for example, women’s right to have their
voices heard, and making modified self-presentation and aestheticisation a condition of their
fulfilment. Whilst Adichie’s success in generating public dialogue about feminism is com-
mendable, her acceptance of the transformative properties of make-up may well reflect a
valuing of things associated with women but is politically void as it does nothing to chal-
lenge or even subvert mainstream notions about femininity. Rather, this ‘cherry-picking’
from feminism, namely celebrating individual agency and disguising conformity as choice,
merely encourages consumption, making women sources of profit for large corporations.
This project brings together two key sites which have been the objects of considerable
attention from feminist scholars during the last four decades; the beauty industry which
has attracted criticism for its exploitation of women in reinforcing male dominance and
setting unattainable standards (Orbach 1978; Wolf, 1990); and women’s magazines with
critics ‘pointing to them as a locus of ideological messages that serve to legitimise and
naturalise unequal relations, and which offer a narrow and restrictive template of feminin-
ity constructed around fashion, beauty and “how to get a man”’ (Gill 2009: 346–347). As
Machin and Thornborrow (2003) explain, discourses are globally marketed by powerful
multinational corporations. It can be no accident that two such corporations appear to have
joined forces in the global marketing of Adichie. It is well documented that magazines
are underpinned by the revenue from advertising space which gives advertisers a powerful
voice in relation to the content and ethos of magazines (Gough-Yates 2003). There can be no

495
Linda McLoughlin

doubt that Adichie’s self-avowed love of make-up is receptive to advertisers because they
would rather see their cosmetics advertisements beside an article that complements what
they are selling.
It is difficult to pinpoint how and when feminism became entwined with consumerism, par-
ticularly as Adichie points out in an interview with vogue.com, ‘it’s obvious that feminism for
many people is a bad word, even if you believe in it, the word is off-putting’. Gill (2009: 346)
refers to a moment of flux and contestation in relation to constructions of feminism and post-
structuralism ‘in which many young women actively disavow or repudiate a feminist identity’.
Popular culture’s rationale for embracing what is colloquially referred to as the ‘F’-word can
be traced to historical shifts within feminism. A number of theorists have tracked different tra-
jectories regarding the emergence of a global discourse of popular (post)feminism as evident
in the Boots’ ‘Speak up’ campaign and the selected Vogue articles. Wolf (1994), for instance,
reflects on the gains brought about by the second wave of the feminist movement which took
place between 1960 and 1980, focusing on the workplace, sexuality, family, and reproductive
rights. This led to material changes within society that have helped to improve women’s posi-
tion, but Wolf now argues that women should renounce ‘victim feminism’ which reinforces the
stereotype of them as fragile and vulnerable as outlined in her earlier work. As an alternative,
she advocates ‘power feminism’ which sees women as equal to men, celebrates female sexual-
ity, and encourages women to claim their individual voices through a variety of tactics. These
include consumer campaigns and putting pressure on the media to alter their sexist portrayals
of women. Given the period in which she was writing, her revision is devoid of any reference
to the importance of intersectionality (the study of intersections between forms or systems of
oppression) but her message that women are gaining power seems to have resonated with text
producers of advertisements and women’s magazines merely because women can now be seen
as a more valuable demographic in the capitalist market place who can no longer be spoken
to in patronising ways. Gill (2007) proposes a postfeminist sensibility to analyse contempo-
rary cultural products. She goes on to explicate the themes that characterise this sensibility
and ‘to emphasise the contradictory nature of postfeminist discourse and the entanglement of
both feminist and anti-feminist themes within them’ (2007: 148). Gill’s identified themes are
relevant to this chapter, particularly those relating to self-surveillance, individualism, choice
and empowerment, the dominance of a makeover paradigm and an emphasis on consumer-
ism, and the commodification of difference. Unlike Wolf, (1994), Gill (2007: 148) relates the
themes to intersectionality, claiming that they ‘co-exist with and are structured by stark and
continuing inequalities and exclusions that relate to “race” and ethnicity, class, age, sexuality
and disability – as well as gender’.
Lazar (2006: 505), in relation to beauty advertising in Singapore, refers to a global dis-
course of popular (post)feminism known as ‘power femininity’ which, she explains:

incorporates feminist signifiers of emancipation and empowerment as well as

circulat[ing] popular postfeminist assumptions that feminist struggles have ended viz.,
that full equality for all women has been achieved and that women of today can “have
it all”; indeed, that it is becoming a woman’s world, with a celebration of all things
feminine, including the desire for self-aestheticisation.

In a review of Adichie’s book We Should All Be Feminists, based on the above-mentioned

TED talk, the British The Telegraph journalist Rupert Hawksley gives an insight into why
Adichie’s ‘brand’ of feminism might prove attractive to proponents of the neo-capitalist5
global economy:

496
 Multimodal constructions of feminism

The 37 year old author of the Orange Prize-winning novel Half of a Yellow Sun has, in
those brief 52 pages, transformed what is so often a deeply divisive, volatile and con-
frontational subject into a clear-headed, honest and beautifully argued statement. It has
forced me to reconsider my opinions in a way that more militant feminist writing never
has. Adichie’s is not the language of warfare, it does not seek to berate men or set one
sex against the other.
(Hawksley 2014)

Clearly, Hawksley finds this less threatening form of feminism more acceptable than other
feminisms and in part explains Adichie’s wide-ranging popularity in the mainstream and
why she was chosen as a brand ambassador. Furthermore, in their selection of a Black
woman in an industry that is notorious for its racism, Boots appear to address another issue
of concern to feminists, namely relating to the politics of skin colour and the prevalence of
white models in advertisements (Asuri 2008; McLoughlin 2013). The analysis will critique
Boots’ positioning as benefactor and Vogue’s apparent altruism, namely, their validation of
dark skin, celebration of diversity and difference, and championing of women’s rights to
reveal the strategies that enable the transnational circulation and strategic exoticisation of
Adichie in the chosen texts. This chapter necessarily draws on all the above discussions, but
it also brings these trajectories together to discuss the significance and transnational circula-
tion of Adichie as a specific cultural form – a celebrity feminist icon and brand ambassador.

Method and approach

The analysis presented here is based upon examination of texts from the Boots’ ‘Speak
up’ campaign, including screenshots and voiceover from the television advertisement and
images and text from print advertisements in magazines and hoardings. Images, text, dia-
logue, clothing, visual design, and layout of Vogue articles featuring Adichie, both before
and after the announcement of her as the ‘new face’ of Boots No. 7, were also examined.
Since visual, verbal, and textual elements are interwoven, a multimodal approach is an
important tool for deconstructing the texts. As Machin et al. (2016: 304) explain, multimo-
dality includes not just language but all the semiotic modes that make up a social context.
They usefully point to the compatibility of a multimodal and CDA approach as both aim
to ‘reveal[s] the discourses buried in texts, which may not be apparent to a casual viewer’.
I employ feminist CDA since the nexus of CDA and feminist studies provides a way to
‘advance a rich and nuanced understanding of the complex workings of power and ideology
in discourse in sustaining (hierarchically) gendered social arrangements’ (Lazar 2007: 141).
This is particularly important, as Lazar points out, due to the complex and subtle forms in
which asymmetrical power relations presently operate particularly in corporate and con-
sumerist ideologies. In deconstructing texts, the aim is to disrupt and render problematic
that which is passed off as ‘common sense’, for example, in neoliberal discourse,6 there
would seem to be nothing inherently wrong in having as a role model a ‘real’ woman who
has achieved commercial success. This is no doubt intended to critique the lack of repre-
sentation of women who challenge the stereotypical depictions of beauty but what exactly
counts as a ‘real’ woman is never questioned and is problematic because it implies that some
women qualify as legitimate whilst others do not.
Machin et al. (2016: 304) emphasise the importance for gender, language, and discourse
in understanding the effects of particular semiotic selections out of a pool of potential alter-
natives, since it is in these choices that ideology and power are encoded. A social semiotic

497
Linda McLoughlin

approach is therefore crucial to this analysis to establish the affordances of signs within the
text. To this end, Kress and van Leeuwen (2006) offer a systematic framework for examin-
ing the formal elements and structures of visual design. They set out four coding orienta-
tions that inform the way in which texts are coded by specific social groups, or within
specific institutional contexts as follows:

1. Technological coding orientations, which have as their dominant principle, the ‘effec-
tiveness’ of the visual representation as a ‘blueprint’.
2. Sensory coding orientations, which are used in contexts in which the pleasure principle
is allowed to be the dominant.
3. Abstract coding orientations, which are used by sociocultural elites – in ‘high’ art […]
In such contexts modality is higher the more an image reduces the individual to the
general, and the concrete to its essential qualities.
4. The commonsense naturalistic coding orientation, which remains, for the time being,
the dominant one in our society.
(Kress and van Leeuwen 2006: 170)

According to Kress (2010), all modes of representation can be harnessed for meaning-mak-
ing, which is contrary to Adichie’s claim, made in the Boots’ television advertisement, that
‘make-up is just make-up’. A key tenet of poststructuralist theory is that texts and discourses
are constructive phenomena, shaping the identities and practices of human subjects. With
this in mind, the following analysis will illustrate that make-up does indeed have a context,
purpose, and symbolic meaning. When Adichie confides that for a period she hid her high
heels and stopped wearing lipstick (Boots’ television advertisement), she alludes to two dis-
tinctive signs which act as signifiers of femininity in a Western European system of cultural
representation.
A feminist postcolonial framework is also beneficial to the analysis since it developed in
response to a criticism that feminism focused on the experiences of women in Western cul-
tures and thus seeks to avoid homogenising and systematising the experiences of different
groups of women. This framework will aid discussion regarding the strategic exoticisation
of Adichie. Intersectionality, mentioned earlier, is a useful approach to denote that Adichie
is simultaneously positioned as a woman, Black, heterosexual, and middle-class.
Goankar and Povinelli (2003) advocate a form-sensitive analysis of cultural phenomena
which I find useful in mapping how forms, such as a feminist icon, become palpable and
recognised as such. Instead of thinking about meaning and translations, they believe we
should instead think about circulation (the cultural process that motivates movement) and
transfiguration (or makeover). Globalisation has led to changes in the modes of communi-
cation; leading to a vast web of intertwined social, economic, cultural, and technological
changes. According to Kress (2010: 5):

Globalisation is not one ‘thing’; it is differently constituted in different places, as are

its effects and impacts, interacting with the vastly varied cultural social, economic and
political conditions of any one specific locality. Yet the deep effects are constant and
recognisable.

One of the mechanisms of global connectivity, relating to Adichie, is the widely quoted
endorsement of her as a feminist, as cited above in the extract from The Telegraph, which
in turn are further quoted in media sources, adding to the flow and dynamics of circulation.

498
 Multimodal constructions of feminism

How Adichie became feted as a feminist icon and imbued with a celebrity aura, it seems,
is contingent on the audience and appears to relate to a complicity in the more acceptable
‘face’ of feminism she proposes. Her self-avowed love of make-up and fashion, coupled
with model-like looks (which offer difference sufficiently within the stereotypical standard),
make her a value-bearing commodity in the context into which she is inserted. The context,
of course, is a consumer culture and the mechanisms of circulation, modernity’s various
modes of communication as mentioned above. Consumption activities such as the Boots’
advertising campaign and frequent inclusion in Vogue further motivates Adichie’s move-
ment across global space. According to Goankar and Povinelli (2003), occasion is a further
condition of circulation. In terms of timing, feminism’s topicality with magazine editors is
explained by Keller (2011) who links this to the emergence of women who grew up with
third-wave feminist values taking up editorial positions at magazines. She interviewed four
New York–based magazine editors who identified as feminist, to uncover contradictions
embedded in their identifying as feminist whilst creating a cultural product often deemed
anti-feminist. Her findings suggest that editors combine practical strategies with a distinc-
tively ‘third-wave ethic’ to navigate between corporate and cultural expectations in order
to integrate a popular feminism into the magazine’s content. I mentioned above the inter-
dependence of magazines and advertising, which can be illustrated by the following com-
ment to Vogue from the Vice President of Skincare and Global brands for Walgreens Boots
Alliance (http://www.vogue.co.uk/article/chimamanda-ngozi-adichie-for-boots-no7):

At No. 7 we believe that when women know their make-up is just right they feel great,
they feel ready to show up in the world in the way they want. Chimamanda was the
perfect choice for us, as not only is she an inspirational woman, we share the same
philosophy about beauty.

I emphasise the slogan-like phrase ‘philosophy of beauty’ because it suggests advanced

knowledge and scholarship which imbues it with authority. The Vice President endorses
aspects of third-wave feminism such as women’s agency: they are ‘ready to show up’
and have independence ‘in the way they want’, whilst validating his company’s choice of
Adichie. This thus conjoins the two in the minds of the reader/viewer.

The transfguration of Adichie in Vogue

This chapter aims to show a gradual shift in Adichie’s subject positioning in Vogue follow-
ing the announcement of her as the ‘new face’ of Boots, from an initial focus on her writing
to an increased concentration on her appearance. Therefore, the analysis and discussion will
deal with texts in the order in which they appeared chronologically.

Article 1
The first article I found in Vogue relating to Adichie, in the Culture > Books section, cor-
responds with the endorsements of Adichie previously mentioned: ‘Chimamanda Ngozi
Adichie on her “flawless” speech, out today as an eBook’. The article begins:

If anyone has the skills to make a speech about feminism go viral, it’s Chimamanda
Ngozi Adichie, the Lagos-based writer whose ideas are as complex as her language is
straightforward […] Adichie’s oration weaved together human stories from her youth

499
Linda McLoughlin

in Nigeria with a complicated discourse about gender roles in the modern world and a
literal textbook definition of “feminism” […]

The term ‘viral’ refers to a phenomenon characterised by mass dissemination of an item

brought about due to advanced technology and the ease with which people now share infor-
mation. Since the writer refers to this activity as a ‘skill’, it can be taken that Adichie’s
ability to make her speech go viral is seen as a positive attribute. The article is presented as
a telephone interview – ‘Reached by phone in Lagos, Adichie spoke to Vogue.com about
the overwhelming success of her speech and what it means to talk politics with the whole
world’. The one accompanying image shows Adichie at the centre of the frame delivering
her speech marked by the lectern and background.
Thus, the image records a reality as it connects Adichie with a particular location –
Euston, the setting for the TED talk, and its specific moment in time, December 2012.
According to Kress and van Leeuwen’s (2001) coding orientation, modality is realised natu-
ralistically as the image presents a true picture of reality. The visual choice of the text pro-
ducer indexes Adichie’s intellect, the meaning associated with academia. This is in contrast
with subsequent images in the magazine that follow the announcement of Adichie as the
new face of Boots, which are focused on indexing her femininity. Signifiers of Adichie’s
African identity are the flawless braid crowning her head and the Dutch Wax print,7 which is
recognisable by its super-bright colours and oversized patterns, of her clothing. The purpose
of the interview is to gauge Adichie’s response to the success of her TED talk and to place
her as spokesperson on global sexual politics. The question ‘Has it [the talk] resonated dif-
ferently in Nigeria than in America?’ places her as a Nigerian subject in the global context.

Article 2
The second article titled ‘Chimamanda says “forget being likeable”’ reports on Adichie’s
acceptance of an award from Girls Write Now, a New York based organisation which seeks
to transform the lives of inner-city girls. The focus of the article is the advice Adichie pur-
ports to give young women which taps into the third-wave popular feminism referred to
earlier:

I think that what our society teaches young girls, and I think it’s also something that’s
quite difficult for even older women and self-professed feminists to shrug off, is that
idea that likeability is an essential part of you, of the space you occupy in the world, that
you’re supposed to twist yourself into shapes to make yourself likeable, that you’re sup-
posed to hold back sometimes, pull back, don’t quite say, don’t be too pushy, because
you have to be likeable.

In prefacing ‘older women and self-professed feminists’ with the adverb ‘even’, Adichie
distances herself from them. The advice she gives young women corresponds with third-
wave feminist discourses of individual empowerment which to some extent repudiates sec-
ond-wave conceptions of essentialism and the notion of a sisterhood. Whilst it is beneficial
to point out that there is nothing inherent in women that would cause them to seek likeabil-
ity, she omits to mention that there may be consequences if they follow her advice and go
against society’s prescriptions regarding appropriate conduct for women. That is not to say
that young women should adhere to society’s prescriptions, but they might be prepared for
the fact that not everyone will appreciate their newly acquired assertiveness. There are two

500
 Multimodal constructions of feminism

Figure 33.1 Adichie in a trendy loft-style apartment.

accompanying images – the first, showing a perfectly groomed Adichie looking off into the
distance, could equally have been used to illustrate a fashion item as the upper body shot,
in a decontextualised setting, is characteristic of fashion shoots. Adichie is styled in cloth-
ing that fuses African and Western influences – a demure black collar reminiscent of the
serviceable ‘little black dress’ injected with animal print and shocking pink layering. The
second image (Figure 33.1) shows Adichie in a trendy loft-style apartment setting looking
confidently into the camera.
The viewer is encouraged to imagine the apartment belongs to Adichie due to what appear
to be personal possessions on display: books, African-themed art and furniture. Thus, the
image could be taken as a visual representation of her success. In Kress and van Leeuwen’s
(2006) framework, it is a ‘demand’ photograph, one in which Adichie directly addresses the
viewer who is presupposed to be naive on matters relating to feminism and women’s rights
compared to the authoritative and knowledgeable Adichie.

Article 3
In April 2015, Vogue published a further interview with Adichie, titled “I wanted to claim
my own name”. This is the longest of the texts analysed signalling the magazine’s grow-
ing investment in Adichie. The interviewer, Erica Wagner, recounts her first meeting with
Adichie on her home territory which is described in the opening lines:

I’m on the shore of Lagos Lagoon with Chimamanda Ngozi Adichie on a late afternoon
in January. It is harmattan season, when a hot wind blows across the Sahara, bringing
dust that makes the sun glow dark gold as it hangs over the palm trees on the opposite

501
Linda McLoughlin

shore. Adichie, in a neat-waisted patterned dress and teetering in lavender heels that
are utterly unsuited to the sandy ground, is about to pose for Vogue’s photographer …

The introduction of Adichie in the context of her homeland is reminiscent of postcolonial

representations of Africa where the unfamiliar landscape to Western eyes, different seasonal
patterns (‘Harmattan’ season88) and food (‘jollof rice’, ‘moin-moin’, and ‘garri’ – referred to
later in the article) are foregrounded. Poverty appears to be eliminated in Adichie’s middle-
class environment although there are references, later in the article, to a cook, driver, and
make-up artist who attend to her needs and, presumably, are not members of the middle
class. Much emphasis is placed on Adichie’s familiarity with the setting, but this is juxta-
posed with signs, for example, ‘lavender heels’, to suggest that Adichie is straddling two
cultures. This makes her the perfect choice as brand ambassador for cosmopolitan femi-
nism. The rationale for the ‘exotic’ landscape (‘dust that makes the sun glow dark gold’)
becomes apparent; it seems that Adichie is modelling for the magazine’s fashion shoot
(high-end magazines are known for, and have been criticised for, their exotic backdrops).
Photographs, presumably from the fashion shoot, are interspersed at various points in the
article. The first shows Adichie in the foreground beside a waterfront with a city scape in
the background, presumably Lagos Lagoon. Adichie is confidently posed, looking into the
camera. Her dress is extremely colourful, with a striking, unusual design. The interviewer,
Wagner, is clearly in awe of Adichie, which is reflected in the number of positive evalua-
tions, not merely in relation to her writing and feminism but also to the fact that they have
bonded through a shared love of orange nail polish. The signification of ‘orange’, since
arguably, this is not a stereotypical nail colour such as ‘nude’ (muted), ‘pink’ (classic), or
‘red’ (bold) in Western contexts, suggests something fashionable and daring. Later, when
her ‘favourite make-up artist’ arrives, Adichie ‘suggests I [Wagner] might like a go in the
make-up chair’. In Eckert and McConnell-Ginet’s (1992) framework, these activities of
polishing nails and putting on make-up, locate the two women within a ‘community of
practice’. Thorne (1994), cited by Eckert (2010), refers to colouring nails and lips as the
technology of femininity where, through the engagement of common practices, gender is
co-constructed; participants come to develop and share ways of doing through engagement
in a common enterprise. In this way, the practice of painting nails and make-up wearing are
normalised for Vogue’s readership and establish Adichie, the novelist, as a member of this
fashion-conscious and make-up–wearing community of practice. The interview is inter-
spersed with references to Adichie’s feminism ‘[b]ecause it’s known in my family, you don’t
want to demean women in my presence!’; ‘The oppression of women’, she says, ‘Makes
me angry … My family says to me, “Oh you’re such a man!”’ We begin to see an emergent
brand of feminism which, though feisty, ‘makes me angry’, is not at odds with an industry
which has been cited as a source of women’s oppression. The title of an article can illustrate
the prominence given to particular topics by the text producer and the title of this article,
‘Chimamanda Ngozi Adichie: “I wanted to claim my own name”’, is no exception, although
the story behind the quotation is only revealed much later in the article:

She got into trouble for speaking her mind in Nigeria: when an interviewer addressed
her as Mrs Chimamanda Adichie, she corrected him, saying she wished to be known as
“Ms”, which the journalist reports as “Miss”. Her insistence on her own family name
was all over the news here last spring. She should be happy to be addressed as “Mrs”,
she was told, since she was, after all, married.

502
 Multimodal constructions of feminism

The interviewer goes on to offer insights into Adichie’s personal life. For example, it seems
that she is quite protective of her privacy, particularly her relationship with her husband but
‘occasionally she lets something slip, like telling me about a pack of crayons he gave her
recently. Crayons?’ This interactive style of writing, anticipating questions from the reader,
creates a sense of the reader eavesdropping on a private conversation, although the less
familiar address term, Adichie, is used throughout. As with the foregoing articles, propor-
tionately, the content is more about Adichie’s writing, feminism, and the accolades she has
received for both, than matters relating to aesthetics. However, an issue is raised regarding
the difficulties for women of colour in finding a foundation to match their skin tone which
acts as a precursor to why Adichie is considered to be the perfect choice in the Boots ‘Match
Made’ campaign. This leads to a discussion about race in which Adichie claims ‘I only
became black when I came to America, […] in Nigeria I’m not black, […] We don’t do race
in Nigeria’; what follows is an explanation that inequalities relating to gender, rather than
race, are the salient issue in Nigeria. As mentioned above, Wagner’s appreciation of Adichie
permeates the interview but her comment that the make-up artist ‘highlights Adichie’s truly
extraordinary beauty perfectly, a shimmering gold on her eyelids the only really glossy
touch’ gives a foretaste of the more focused attention on Adichie’s physical appearance in
subsequent editions.

Article 4
The next article titled ‘African novelist Chimamanda Ngozi Adichie delivers a powerful
message on Dior’s front row’, does much to confirm Adichie’s status as a celebrity:

Under a flutter of camera flashes announcing the front row arrivals of Rihanna, Jennifer
Lawrence, and Kate Moss at Dior this morning, one may not have immediately noticed
that the show’s true guest of honor had arrived. Quietly slipping past photographers,
leading African author and modern feminist icon Chimamanda Ngozi Adichie took her
front row perch alongside fashion power players.

Having elevated Adichie’s status relative to other ‘A-list’ celebrities, the report goes on to
explain Adichie’s presence at the fashion show, namely, that she was the inspiration for
a new designer’s ‘female-empowered designs’ (T-shirts emblazoned with ‘We Should All
Be Feminists’ the soundtrack from which is played as the models walk the runway). The
report also mentions that Adichie’s talk was sampled in a hit by Beyoncé. These endorse-
ments of Adichie’s feminism from the designer (whose success, according to the report,
emanates from the fact that she is Dior’s first female creative director in 70 years) and pop
icon Beyoncé (whose involvement in campaigns such as ‘ban bossy’ links her with popular
feminism) are further elements in the enabling matrix that place Adichie firmly in the realm
of celebrity. There are two accompanying images – one where Adichie is posed, according
to the caption, in ‘long-sleeved sculptural frock splashed with a monochromatic graphic
pattern and accessorised with a pair of green and magenta shoes that matched the colourful
patchwork pockets of her dress. It delivered as strong a message as her words’ (my empha-
sis). The other image shows a slightly uncomfortable-looking Adichie in the front row, with
clasped hands, intently looking up to the runway. How Adichie’s clothing signals a message
and what the message is, is not quite clear but we see a link starting to form connecting
Adichie’s appearance with what she says.

503
Linda McLoughlin

Boots No. 7 advertising campaign

The appropriation of this feminist icon for commercial purposes is most visible in the Boots
No.7 ‘Match Made’ advertising campaign, which can be viewed through the link in Vogue’s
article: ‘Chimamanda Ngozi Adichie is No. 7’s new face’ (18th October 2016). The televi-
sion advertisement signals optimism with its opening frame showing a lush green field with
clear blue skies above. The soothing background music is accompanied by melodic female
voices humming harmoniously to the rhythm. A furrow acts as a vector (the lines which
link elements) leading the viewer’s eye to a distant figure – as it comes nearer, we see the
figure is Adichie in freely flowing pale pink dress. Adichie speaks, introducing a problem,
namely that due to society’s restrictions on women’s behaviour, she was not taken seriously
because of her love of make-up. The solution, as mentioned earlier, was to stop wearing
make-up and hide her high heels. The camera points up to the bare branches of a tree – it
then pans out to show the full tree to the right of the frame. Presumably, the viewer is to
connect the desolate tree with the feelings Adichie had on being make-up free and without
high heels. Far from resolving the issue, this abandonment of feminine accoutrements led
to a further problem – ‘I became a false version of myself’. It is not clear what led to her
transformation, but Adichie claims that she ‘woke up’. The spoken discourse is accompa-
nied throughout by visual signifiers corresponding with the words spoken. For example, the
scene in which Adichie moves is autumnal, signifying optimism, rebirth, and renewal; a tree
sheds its leaves then they magically reconnect to the branches in a colourful display of red
and gold. As the music becomes more dramatic, a young animal is startled, corresponding
with Adichie’s claim that she ‘woke up’. The signifiers promote the notion of transforma-
tion and enjoyment of a self-made new look. Adichie’s stance is that make-up wearing can
be an act of resistance, ironically, not to challenge restrictive ideals of female beauty but to
feminism itself. Her claim in the advertisement – that make-up is just make-up, it’s about
how she feels when she gets it right, what makes her happy when she looks in the mirror,
what makes her walk ever so slightly taller, the face she chooses to show the world and what
she chooses to say – is as much about psychological and emotional well-being as it is about
empowerment.
A comment from the novelist, Salman Rushdie, cited in Article 3, on meeting Adichie
at a literary festival, gives some indication as to why Adichie was seen as an appropriate
ambassador to front a campaign with the slogan ‘ready to speak up’:

[…] but what was so striking was her own confidence and authority. She very much
held her own, and spoke fluently and powerfully, and all of us there that day could see
that someone very remarkable had just arrived. A star is born, I remember thinking, and
so it was.

Adichie’s eloquence and force in speaking are the basis for Rushdie’s metaphorically
charged positive evaluation. One cannot criticise any attempt to credit a woman’s verbal
proficiency since it does much to dispel negative stereotypes of public-speaking men
and silent women (Cameron 2007). Furthermore, Adichie’s novels are celebrated for the
strong voices of female characters. As seen above, Adichie seems to adopt the adage
of second-wave feminists by making the personal political, namely in reprimanding the
person who insisted on calling her ‘Mrs’. Taken together, it is therefore difficult to imag-
ine that Adichie appears to see no contradiction in co-opting feminism for commercial
purposes.

504
 Multimodal constructions of feminism

Conclusion
In this chapter, I have shown how feminist CDA, informed by social semiotics, is a useful
framework for analysing the transnational circulation of a feminist icon. In their selection of
Adichie, Vogue and Boots appear to have ceded to pressure on the media to alter their negative
portrayals of women. However, offering a ‘real’ woman as a role model, making platitudinous
remarks about rights and choices, making more shades available to match a wider range of skin
tones etc., hardly makes these multinational companies feminist. On the surface, the choice
of Adichie as a motivational and inspirational role model is difficult to contest – her relaxing
smile is safe and reassuring and she exudes confidence, albeit as a result of transformation and
reinvention. Her claims about make-up are presented as a positive aspect of modernity and pro-
gress but there is little consideration of the implications for non-make-up wearers who, presum-
ably, are lacking in traits such as the confidence required to speak up for themselves. Adichie
comes across as rational and well-informed; her claims are presented as incontrovertible facts
– ‘make-up is just make-up’ – though never supported by research. Her neoliberal discourse on
self-aestheticisation is worrying to say the least; suggesting that a pleasing appearance can fos-
ter psychological and emotional well-being does nothing to tackle the material obstacles that
women still face in today’s society. Furthermore, she creates an illusion that make-up wearing
is an expression of women’s freedom; that women are making rational decisions of their own
volition to attain an appearance of their own choosing. However, her own appearance, though
a refreshing change from the customary white, Western models of beauty advertisements, typi-
fies narrow cultural ideals of beauty and it seems little thought has been given to how her con-
ventionally attractive model-like looks, curvaceous figure, and well-endowed breasts might be
difficult for many women to measure up to, potentially causing anxiety and insecurity.

Future directions
Fraser (2013) reminds us that at its roots, the feminist movement’s raison d’être was to dis-
mantle patriarchy and, with it, capitalism. Her advice that feminism should end its dangerous
liaison with marketisation and that we should be wary of actors who push for it offers a way
forward. It seems Adichie’s neoliberal stance on feminism appeals to commodity culture since
it exacerbates perceived tensions between second- and third-wave feminism. In espousing
that ‘we should all be feminists’, she makes a distinction between conservative and neoliberal
feminism. Her references, cited in the texts analysed, to ‘self-professed’ feminists, ‘moralising
make-up’, etc. suggest that she seeks to distance herself from and to challenge the perceived
oppressive ideology of more ‘militant’ feminists who stereotypically are believed to frown
upon make-up wearing, are male-bashing, confrontational, etc. Misrepresenting and dismiss-
ing other forms of feminism in this way disavows the inroads gained but is also disingenuous
since how else does she imagine women came to have the income needed to purchase the
beauty products she promotes? In reclaiming beauty practices in terms of pleasure inducement
and its transformative properties, then making tenuous links to factors related to women’s pro-
gression – agency, individualism, empowerment, the right to speak up and be heard – it plays
into the hands of commodity producers and thus should be avoided.

Acknowledgements
I am very grateful to Akintunde Akinleye for kind permission to use the image of
Chimamanda Ngozi Adichie.

505
Linda McLoughlin

Notes
1 This is how the magazine’s ethos is described to advertisers: ‘a globally renowned women’s luxury
magazine on fashion, trends, beauty, people and lifestyle’. http://www.themediaant.com/magazine/
vogue-magazine-advertising
2 Boots, established in 1849, formed an alliance with Walgreens in December 2014, making it the larg-
est retail pharmacy in America and Europe.
3 TED (Technology, Entertainment and Design) are influential talks by expert speakers on a range of
topics. For more information go to: https://www.ted.com/about/our-organization
4 I am also indebted to Goldie Asuri (2008), whose article on the transfiguration of the former Miss
World and Bollywood star Aishwarya Rai, was a source of inspiration.
5 Neo-capitalism is an economic ideology which blends some elements of capitalism with other sys-
tems. The capitalist doctrine becomes deeper due to technological developments and the internation-
alisation of markets; ‘new’ compared to forms of capitalism before World War II.
6 Neoliberal discourse refers to discourses that emphasis individual agency in the form of taking
responsibility for one’s behaviour and lifestyle choices.
7 Dutch Wax print, originally produced in Europe and sold in West Africa, has become a marker of
identity for some West African women.
8 The Harmattan is a season in the West African subcontinent, which occurs between the end of
November and the middle of March.

Further reading
Chandler, D. (2017) Semiotics: The Basics, 3rd ed. London: Routledge.
This book introduces the key concepts of semiotics in accessible and jargon-free language.
Gill, R. and Scharff, C. (eds.) (2011) New Femininities: Postfeminism, Neoliberalism and Subjectivity.
London: Palgrave Macmillan.
This edited collection looks at the way in which experiences and representations of femininity are
changing, and explores the possibilities for producing ‘new’ femininities in the twenty-first century.
Kress, G. (2010) Multimodality: A Social Semiotic Approach to Contemporary Communication.
London: Routledge.
This accessibly written book examines multiple modes of communication and meaning-making
with lots of illustrations to demonstrate the points made.
Kress, G. and van Leeuwen, T. (2006) Reading Images: The Grammar of Visual Design, 2nd ed.
London: Routledge.
This book examines the ways in which images convey meaning. It offers a systematic and
comprehensive account of the grammar of visual design.
Lazar, M. M. (ed.) (2005) Feminist Critical Discourse Analysis: Gender, Power and Ideology in
Discourse. London: Palgrave Macmillan.
This collection brings together well-known scholars writing from feminist perspectives within
CDA. The theoretical structure of CDA is illustrated with empirical research from a range of locations
and domains.

Related topics
Gender and sexuality in discourse: semiotic and multimodal approaches; multimodal constructions
of feminism; revisiting ‘gender advertisements’ in contemporary culture; doing gender and
sexuality intersectionally in multimodal social media practices; the South African news media and
representations of sexuality.

References
Asuri, G. (2008) ‘Ash-coloured whiteness: The transfiguration of Aishwarya Rai’. South Asian
Popular Culture, 6(2), pp. 109–123.

506
 Multimodal constructions of feminism

Cameron, D. (2007) The Myth of Mars and Venus. Oxford: Oxford University Press.
Eckert, P. (2010) ‘Vowels and nail polish: The emergence of linguistic style in the preadolescent
heterosexual marketplace’. In: Meyerhoff, M. and Schleef, E. (eds.) The Routledge Sociolinguistics
Reader. London: Routledge, pp. 441–447.
Eckert, P. and McConnell-Ginet, S. (1992) ‘Communities of practice: Where language, gender and
power all live’. In: Hall, K. and Bucholtz, M. (eds.) Locating Power: Proceedings of the Second
Berkeley Women and Language Conference. Berkeley, CA: Women and Language Group, pp. 88–99.
Fraser, N. (2013) Fortunes of Feminism from State-Managed Capitalism to Neoliberal Crisis. London:
Verso.
Hawksley, R. (2014) ‘Why we should all be feminists’. Available at: http://www.telegraph.co.uk/cul
ture/books/11201682/Why-we-should-all-be-feminists.html (Accessed: 29th March 2018).
Gill, R. (2007) ‘Postfeminist media culture; elements of a sensibility’. European Journal of Cultural
Studies, 10(2), pp. 147–166.
Gill, R. (2009) ‘Mediated intimacy and postfeminism: A discourse analytic examination of sex and
relationships advice in a women’s magazine’. Discourse and Communication, 3(4), pp. 345–369.
Goankar, D. P. and Povinelli, E. (2003) ‘Technologies of public forms: Circulation, transfiguration,
recognition’. Public Culture, 15(3), pp. 385–397.
Goldman, R. (1992) Reading Ads Socially. London: Routledge.
Gough-Yates, A. (2003) Understanding Women’s Magazines. London: Routledge.
Keller, J. (2011) ‘Feminist editors and the new girl glossies: Fashionable feminism or just another
sexist rag?’ Women’s Studies International Forum, 34(1), pp. 1–12.
Kress, G. (2010) Multimodality: A Semiotic Approach to Contemporary Communication. London:
Taylor and Francis.
Kress, G. and van Leeuwen, T. (2001) Multimodal Discourse the Modes and Media of Contemporary
Communication. London: Arnold.
Lazar, M. (2006) ‘“Discover the power of femininity!” Analysing global “power femininity” in local
advertising’. Feminist Media Studies, 6(4), pp. 505–517.
Lazar, M. M. (2007) ‘Feminist critical discourse analysis: Articulating a feminist discourse praxis’,
Critical Discourse Studies, 4(2), pp. 141–164.
Machin, D. and Thornborrow, J. (2003) ‘Branding and discourse: The case of Cosmopolitan’.
Discourse and Society, 14(4), pp. 453–471.
Machin, D., Caldas-Coulthard, C. R., and Milani, T. M. (2016) ‘Doing critical multimodality in
research on gender, language and discourse’. Gender and Language, 10(3), pp. 301–308.
McLoughlin, L. (2013) ‘Crystal clear: Paler skin equals beauty – A multimodal analysis of Asiana
magazine’. Asiana, 11(1), pp. 15–29.
Orbach, S. (1978) Fat is a Feminist Issue the Anti-Diet Guide to Permanent Weight Loss. New York,
NY: Paddington Press.
Thorne, B. (1994) Gender Play. New Brunswick, NJ: Rutgers University Press.
Walgreen Boots Alliance. Available at: http://www.walgreensbootsalliance.com/about/ (Accessed: 3rd
February 2019).
Wolf, N. (1990) The Beauty Myth. London: Chatto and Windus.
Wolf, N. (1994) Fire with Fire: New Female Power and How It Will Change the Twenty-First Century.
New York, NY: Vintage.

The texts
Vogue Culture > Books (July 2014) ‘Chimamanda Ngozi Adchie on her “flawless” speech, out today
as an eBook’. Available at: http://www.vogue.com/article/chimamanda-ngozi-adicihie-feminism-b
eyonce-book (Accessed: 3rd February 2019).
British Vogue (June 2015) ‘Chimamanda says “forget being likeable”’. Available at: http://www.
vogue.co.uk/article/chimamanda-ngozi-adichie-likeability-women-strive-to-be-liked (Accessed:
3rd February 2019).

507
Linda McLoughlin

British Vogue (November 2015) ‘Chimamanda Ngozi Adichie: “I wanted to claim my own name”’.
Available at: http://www.vogue.co.uk/article/chimamanda-ngozi-adichie-novelist-ted-speaker-int
erview (Accessed: 3rd February 2019).
Vogue (September 2016) ‘African novelist Chimamanda Ngozi Adichie delivers a powerful message
on Dior’s front row’. Available at: http://www.vogue.com/article/dior-chimamanda-ngozi-adichie
-front-row-we-should-all-be-feminists-beyonce (Accessed: 3rd February 2019).
Vogue (21 October 2016) ‘Boots’ No. 7 match made “ready to speak up” television advertisement’.
Available at: https://www.vogue.co.uk/article/chimamanda-ngozi-adichie-for-boots-no7 (Accessed:
3rd February 2019).

508
 34
Judged and condemned
Semiotic representations of women criminals
Carmen Rosa Caldas-Coulthard

Introduction1
This chapter will explore how women perpetrators of crime are resemiotised in public dis-
courses, focusing primarily on media narratives with particular attention to the Brazilian
press. Iedema (2003: 29) refers to ‘resemiotisation’ as the means

to provide the analytical means for (1) tracing how semiotics are translated from one
into the other as social processes unfold, as well as for (2) asking why these semiotics
(rather than others) are mobilised to do certain things at certain times.

I argue that the semiotic resources (language, images, colour, among others) which are used
to represent women criminals are most of the time associated with their gender and have
therefore important cultural and political gendered meanings. I define gender here

along two key dimensions. First, social relationships (and representations) are infused
with assumed differences between the sexes. These assumptions are ideological and
can be shown to be constructed in language and other forms of representation and in
social practices in institutions. … Second, gender imbalance never operates alone but
intersects with other axes of inequality.
(Machin, Caldas-Coulthard, and Milani 2016: 306)

For my analysis, I take a multimodal, discourse-analytical perspective since I believe that

gendered identities are mainly constructed through multiple semiotic modes.
The notion of discourse adopted in this chapter relates to a set of attitudes or values of
the producers of the semiotic signs. Inevitably, our choices in text and image reflect these
attitudes and values, and also constitute them. I have drawn on the concept of discourse
following authors such as Fairclough 1992; Foucault 1978; Hall 1997; Kress 2010; Kress
and Van Leeuwen 1996, 2001; and Wodak 2000 for whom discourses are knowledges of
practices which are at the same time:

509
Carmen Rosa Caldas-Coulthard

1. Knowledges of how things are or must be done, together with specific evaluations and
legitimations of, and purposes for, these practices (Caldas-Coulthard and Van Leeuwen
2001: 158);
2. Knowledges which are linked to the context of specific communicative practices mate-
rialised in texts or other semiotic resources (Van Leeuwen 2005).

By using discourse and multimodal analysis as methodological tools, I am in fact consid-

ering the ways semiotic resources create and recreate meanings reflecting the views and
ideology of speakers/writers.
Crime as a topic has been extensively studied in many disciplines: criminology (Jewkes
2004; Jewkes and Letherby 2002; Newburn 2017; Williams 2012), sociology (Carradine
et al. 2009; Giddens 1989), psychology (Howitt 2011), media and cultural studies (Brown
2003), and more recently, cultural criminology (Ferrell, Hayward, and Young 2015). These
disciplines have shed light on the many ways media can distort crime and how some events
are defined as crimes and others are not. There is however, very little systematic enquiry
into the mediatised semiotics of crime. A critical approach focusing on semiotic resources
is a powerful tool to understand the intricacies of media representation of crime and its
ideological implications.
According to the Map of Violence 2015 (Mapa da Violência), produced by the Latin-
American Social Sciences Faculty (Flacso), in Brazil, where I myself live, 4,762 women
were murdered in 2013. There was an increase of 54 per cent in the number of Black women
killed every year in the last ten years.
In 2013, according to the same source, Brazil had a figure of 4.8 murders per a 100,000
women. It is in fifth place out of 83 investigated countries for feminicide. There is one case
of rape in Brazil every 11 minutes, according to the Brazilian Yearbook of Public Safety
(Folha de São Paulo 2017).
Although crimes of violence are much more committed by men and against women than
the other way round, women also kill and commit crimes. There is nevertheless, a general
essentialist tendency in postmodern societies to perceive women as fragile and submissive
and less associated with crime. This tendency is based on values of a positivist or traditional
criminology (Espinoza 2002), which recognises intrinsic qualities in men that make them
more prone to crime:

… in their female ‘essence’, women have a low tendency for delinquency. If a criminal,
the motives for the deed would be related to early sexuality, puberty, the menopause,
labour or hormone alterations.
(Ratton, Galvão, and Andrade 2011: 3)

For Bianchini (2011), common-sense views associate women with crimes of passion or
crimes related to motherhood (abortion or infanticide). Women are not considered as agents
of premeditated or planned crime, like men. Statistics however point to a significant rise in
female criminality. Between the years of 2000 and 2011, according to the Brazilian National
Department for Prisons (Departamento Penitenciário Nacional), the number of women pris-
oners grew by 252 per cent.
These figures were a starting point for my research into the representation of female
criminals in the press. And the initial questions were: of all these female criminals, who
gets reported in the press? And how? What are the ideological implications of these

510
 Representations of women criminals

representations? In other words, in stories about crime and gender, what makes a story
‘reportable’ in the first place?
Crime stories, where sexualised women are represented as main characters, are a signifi-
cant arena in which social conflict and gender discrimination are symbolically explored. I
will discuss here therefore how the very few female characters that get into the main press
are recontextualised in relation to their male counterparts and condemned doubly – for their
deeds and for their gender. For my analysis, I draw on images from image banks, from
newspapers, and online news.

Analytical tools
I adopt a critical semiotic multimodal approach to my data. The main proponents of this
approach to social interaction, Kress and Van Leeuwen (2001), follow the works of systemic
functional linguistics (Halliday 1978; Halliday and Hasan 1989; Halliday and Matthiessen
2014) and critical discourse analysts (Fairclough 1989, 1992, 2003, 2010; Weiss and Wodak
2003; Wodak and Meyer 2001). Their main innovation nevertheless is to include in their
analyses not just language but all the semiotic modes that make up a social context. They
also raise critical questions about the ‘affordances’ and buried ideological purposes used by
communicators in a context of usage.

Affordances are the potential uses of a given object, stemming from the perceivable
properties of the object. Because perception is selective, depending on the needs and
interests of the observer, different perceivers will notice different affordances. But those
that remain unnoticed continue to exist objectively, latent in the object, waiting to be dis-
covered. The same can be said of meaning. People will derive meaning from the material
qualities of an object, for example, its colour, and which qualities they notice and what
significance they give to them will depend on their needs and interests, whether these are
individual and contingent and/or social and cultural, shared with a community.
(Van Leeuwen 2011: 59)

Van Leeuwen continues to say that the term ‘meaning potential’ (Halliday 1978) is related to
the ‘affordances that have become part of the acknowledged semiotic resources of a culture’
(ibid.). Colour or typography, for example, have a theoretical meaning potential ‘consisting
of all their past uses and an actual meaning potential constituted by those past and present
uses that are acknowledged and considered relevant by the users of that resource in a spe-
cific context’ (ibid.). Think, for example, the Eiffel Tower covered by the French flag after
the Football World Cup and the meanings of victory it afforded.
This critical and semiotic approach is very appropriate for my purpose of investigating
gender stereotyping in the representation of criminal women since it helps me to under-
stand the effects of particular semiotic selections chosen from a pool of potential alterna-
tives, which lead to discrimination. It is in the semiotic choices that ideology and power are
encoded. Differently from other discursive analytical approaches, an important part of this
affordance-led approach is the possibility to have something to say about particular kinds of
semiotic material, how they are used, and what is special about it.
For the multimodal analysis, I use semiotic categories put forward by Kress 2010; Kress
and Van Leeuwen 1996, 2001; Machin 2007; and Van Leeuwen 2005, 2007, 2008, and 2011
to consider photographs, colours, and language as meaning potential.

511
Carmen Rosa Caldas-Coulthard

To complement the visual multimodal analysis, I also use insights from ‘appraisal theory’
(Martin and White 2005) in order to discuss ‘labelling’ and ‘evaluation’ in the linguistic
mode.
My main thesis is that, as with language, semiotic modes can produce evaluative mean-
ings that will influence the ways we receive information. Despite all the advances in soci-
ety in terms of gender relations, the women in my data are contextualised as transgressive
actors, not only because they are criminals but also because they perform behaviours highly
inappropriate for their ‘sex’. They are trivialised, derogated, and consequently judged and
condemned.
My conclusions point to processes of social devaluation: sexism is the pervasive and
underlying ideology recurrent in many representations of female criminals.

Crime and the media

Portrayals of crime and deviance are prominent and proliferate in all sorts of media.
Deviancy, particularly in the form of ‘fictionalised’ or ‘real life’ stories, fascinates people.
Crime reports, written and spoken, and their accompanying images are entertaining for their
dramatic nature and their appeal to morbid curiosity, but they are also compelling since they
connote a sense of insecurity and risk. Repackaged as a mass-media entertainment com-
modity, crime reports blur the boundaries between fact and fiction. They are a source of fear
and escapism. Their popularity is attested in the many crime, serial killer, and sex-crime
books sold at airports, bookshops, and even at supermarkets. In the visual domain, the most
popular TV series in the last decades are all about crime and deviance: the American The
Wire (2002–2013), Breaking Bad (2008–2013), the British produced Prime Suspect (1991–
2006), Wallender (2008–2016), The Fall (2013–2016), Line of Duty (2012–2017), Peaky
Blinders (2013–2017), among others. The outcome of these representations of crime life
through powerful episodic, realist narration is that viewers, myself included, get involved
with characters and addicted to the development of the story.
The treatment of any crime topic will always depend, however, on who is chosen to com-
ment and whose opinions and definitions are sought. Choice and selection, therefore, will
determine how a certain event or a certain person will be reported/represented and the impli-
cations derived from this choice will have ideological consequences. Clearly, therefore,
media crime stories do not tell us about society. They show us, as Hartley (1982) suggests,
certain aspects of society.
Each crime or tragic story is written against the background of other similar stories and
they become part of a larger myth about values. Crime is outside common sense and it can
only be explained in terms of ‘concepts and mythologies of dominant ideology (corruption,
criminal infection, foreign agitation)’ (Chibnall, 1977: 115). I also add ‘gender’ to this list.
The framework of ideologies is revealed and reflected in the semiotic choices made by
producers. Because semiotic resources are an instrument of communication, but also an
instrument of control, a particular choice from a range of options can convey a fact or distort
it and readers and viewers can be informed or manipulated. That is why particular choices
can be so significant in terms of discourse interpretation.
The processes articulated in crime narratives focus on acts of violence committed by
social actors and the actions taken by the authorities involved in the contexts of crime –
police procedures, the court proceedings judgement, and prison. An interesting and very
subtle identity choice in terms of gender in several current TV series is that the many ‘new’
female heroines, although powerful in their functional representations (they are lawyers,

512
 Representations of women criminals

politicians, detectives, investigators, or even police superintendents – roles most of the time
performed by men), are frequently associated with mental problems (either they are autistic,
or bipolar, or even drug addicts). Contrastively and not surprisingly, their male counterparts
do not seem to have such mental problems!
In current written media stories about crime, a great number of participants and contexts
are represented not only linguistically but also visually. Women as perpetrators of crime are
construed very specifically in terms of gender and produce an important link to the social
negative imaginary about women. This is what I will discuss next.

Visual representation and gender

Figure 34.1 is a picture from a crime story, and Figure 34.2 is a picture from another. What
could a reader deduce from these images as part of a story?
Whereas the reader looking at the second picture might reasonably guess that the story
was about boys on the run, the reader of the first is more likely to assume it was about pros-
titution. In fact, both news stories are concerned with theft, robbery, and motorcycles.
The first story is a piece of news published on line in 2015, which tells the story of two
female ‘adolescents’ who stole a motorbike and were put in prison. The accompanying
text has a typical narrative structure (situation, problem, solution, evaluation). The action
involves the theft/stealing of a motorbike and the police taking hold of the situation (not
only was the robbery against the law but also drugs were found in a house where the adoles-
cents had been before the robbery).
The second crime story, which also follows the same narrative structure, does not sexu-
alise the boys to imply criminality in the visual representation, because sexuality is not a
semiotic resource to be drawn upon in the characterisation of male criminality. Props that

Figure 34.1 An adolescent is arrested for the seventeenth time for stealing a motorbike in
Anápolis, Goiania (since these pictures appeared in open access blogs, permission
for reproduction is not applicable).

513
Carmen Rosa Caldas-Coulthard

Figure 34.2 A motorboy is filmed during the robbery and the thief is shot by the police.

are present in portrayals of male criminals (guns, glasses, particular hats, hairstyle, jewel-
lery, etc.) do, however. But their signification is very different from the ways the girls are
made to signify (or are resemiotised) through their bodies.
The photos of these two stories do not add any new narrative meaning to the recount-
ing of facts. They are simply evaluative and connotative – the first represents unnamed,
fragmented, semi-naked bodies. The decision to show only parts of the whole person – the
upper legs and hips – sexualises the adolescents even though this has nothing to do with
the reported story. And it associates people’s view of prostitution being linked to the deeds
committed by the actors. Therefore, the underlying message is that girls who ‘dress’ in that
way are likely to be criminals and the consequent implication is that because they ‘look’ like
this, they must be guilty.
My point therefore is that, as these first examples demonstrate, there is a clear gender
difference in the way women and men are portrayed in ordinary crime stories and that the
visual elements add hidden meanings not made explicit in the textual form.
This preliminary point will be further expanded in my analysis of women criminals in the
section ‘Female criminals – a question of gender’ below.

Image and meaning: theoretical categories

Denotation and connotation
According to Hall (1968) images show particular events, particular people, places, and
things. They ‘document’, or in semiotic terminology, they ‘denote’. The author proposes
‘pointers of denotation’ in visual communication, which are very important tools for the
understanding of visual messages. Hall also suggests that people can be categorised accord-
ing to specific social types (dress, objects, or physiognomic features). They can also be
categorised as a group or as an individual.
The social distance between participants in the image and the viewer affords different
meanings. ‘Intimacy’ or ‘distance’ is produced according to the ways bodies are depicted in
pictures. If the viewer interacts visually with only a face or a head, s/he feels intimate and

514
 Representations of women criminals

near the represented actor. A focus on head and shoulders produces meanings of personal
closeness while a more distant focus on the upper, lower body or complete body distance the
viewer from the represented participant.
In the picture (Figure 34.3), the five children document or denote middle-class children
(for their clothes and the setting they are placed in) in a close social distance (we can see
their whole bodies) in relation to the viewer. They are happy, amazed children looking at
something interesting and the viewer empathises with them.
Other images however depict people, places, things, and events to get general or abstract
ideas across. They connote ideas and concepts.
The representation of beautiful people, smiling, placed in interesting and attractive con-
texts (Figure 34.4) draw us to the represented participants and our reaction is positive.
Negative representations (Figure 34.5), however, produce the opposite effect.
Both positiveness and negativeness are, however, construed through the semiotic
resources authors choose in order to deliver their message. In Figure 34.4, for example, the
social actors (the connoted grandmother and grandson) are represented as happy, smiling,
and cuddling and surrounded by lush green trees – all these choices connote ‘affection’
and ‘intimacy’. In Figure 34.5, by contrast, we see the representation of an old lady, ugly
because of her exaggerated physical characteristics. The semiotic evaluation is negative
since she is construed as ‘odd’ for her age: in the colour image (not reproduced here)
her glasses are green, her hair is blue, and her shoes are red. She connotes transgression,
therefore.
According to Machin (2007: 25) asking what an image connotes is asking:

What ideas and values are communicated through what is represented, and through the
way in which it is represented? Or, from the point of view of the image-maker: How
do I get general or abstract ideas across? How do I get across what events, places and
things mean? What concrete signifier can I use to get a particular abstract idea across?

Figure 34.3 Cousins (author photo).

515
Carmen Rosa Caldas-Coulthard

Figure 34.4 Grandma with child (courtesy of iStock/Getty images, reproduced with
permission).

Figure 34.5 Naughty old woman (courtesy of iStock/Getty images, reproduced with permission).

516
 Representations of women criminals

For Barthes (1977: 23), ‘connotation’, is realised through the choice of ‘poses’, ‘objects’
(these are inducers of ideas, in other words, a bookcase equals an intellectual person); or
‘settings’ (landscapes, rivers, mountains, the sea, etc. can be used as metaphors for changes
in life, the passing of time, or peace); and, finally, ‘photogenia’ (the ways pictures are framed
by distance, focus, and illumination). And of course, the same object or pose can have dif-
ferent connotations for different people. A woman in a short skirt can be viewed as ‘modern’
in Western cultures, for example, but could be arrested for immorality in the Middle East.
All visual choices, therefore, like their linguistic counterparts, are never neutral but
always presented through an ideological point of view and they are always ideologically
motivated. Fairclough (2010: 26) proposes that ‘ideologies are a significant element of pro-
cesses through which relations of power are established, maintained, enacted and trans-
formed’. He also points out that ideological positioning is most effective when its workings
are least visible. In visual representation, viewers tend to believe what they see without
critical questioning of how the choices are made, which is a big problem.
Kress and Van Leeuwen (2001: 20) argue that the same meaning, within a social cultural
domain, can be expressed in different semiotic modes and modes are always resources for
making meaning.
Language is, of course, one of the main semiotic resources we use. All semiotic systems
allow us to negotiate social and power relationships. The fragmented female bodies in the
pictures that accompany the crime narrative in Figure 34.1 ‘afford’ meanings of gendered
sexuality, for example. Therefore, if we want to understand how crime is visually repre-
sented and how some actors are more judged or discriminated against than others in visual
communication, we have to consider the potentiality and affordances of their representation
and how semiotic categories are used to realise particular meanings.
Another important category in visual communication proposed by Kress and Van
Leeuwen (1996) and based on functional linguistic theory is ‘modality’, a subsystem of
the interpersonal function proposed by Halliday (1978) or the relation between viewer and
image. Kress and Van Leeuwen (1996: 160) suggest that one of the most important aspects
of visual communication is the question of the truthfulness of the message. What do we see
that is real, factual, true, or a lie, a fiction, or something out of reality? Choices in ‘modality’
therefore signal degrees of truth or probability expressed by ‘certainty’, ‘possibility’, and
‘probability’ about the world that can be:

• Irrealis – it can be like this;

• Realis – it is like this.

Visual images can also represent the world as though it is real, naturalistic, or as though it
is fantastic or imaginary. The authors call these aspects of representation ‘code orientation’.
For Kress and Van Leeuwen (ibid.), a social semiotic theory cannot establish the absolute
truth about visual representations. It can however, demonstrate that a given proposition is
represented as true or not. Modality is therefore essential for the analysis of multimodal rep-
resentation since it can show people, places, and things as real or as unreal (blurred pictures,
for example) or as caricatures of a proposition or even copies of the real. In crime news, as
I will discuss below, modality and code orientation play a very important role since reality
is always presented according to who sees or recounts events. The general tendency, how-
ever, is for the vast majority of crime representation to be naturalistic and high in modality
choices, while in advertising discourse, sensory orientation plays a very important role since
persuasion is the final aim.

517
Carmen Rosa Caldas-Coulthard

Female criminals – a question of gender

According to the historical and judicial research into female crime developed by Faria
(2010), women, throughout history, have been victims of a social trap that puts them into
a position of being fragile and docile and these characteristics are responsible for the con-
struction of the stereotype of women being less competent. This supposed lack of com-
petence has helped to construe women’s universe as inferior to men’s. As a consequence,
women have been considered, because of their recognised docility, a lot less capable of
committing crimes, and when they do so, they are said to be under the influence of a man or
involved in matters of the heart.
Beauty and seduction have also been constantly evoked to justify female crimes – the
assumption that the most attractive women would be able to trick and deceive men bet-
ter. Traditionally therefore, ‘views of why women turn to or refrain from crime have been
rooted in biological and psychological explanations that focus on women’s nature, which is
supposedly different from men’s (Mayr and Machin 2012: 112).
Lombroso (1895/2004), the founder of anthropological criminology, was one of the first
authors to categorise women criminals according to stereotypical views and evaluative gen-
dered constructions. ‘For him, there are links between the nature of a crime and the person-
ality or physical appearance of the offender’ (Mayr and Machin, ibid). Women criminals
were either:

• Sexual offenders;
• Born criminal – monsters;
• Hysterical offenders;
• Lunatic criminals – mad;
• Epileptic and morally insane.

It is very interesting to note that these constructions still persist in contemporary narratives
about women criminals and visual representations emphasise these assumptions. Evaluations
attached to them (both linguistic and visual) are very similar to Lombroso’s lexicalisation
(or choice of words) of women criminals (Jewkes 2014: 113) – they are described by their
sexuality or sexual deviance, their absence of physical attractiveness or as femmes fatales,
as bad wives, as bad mothers, as mythical monsters, or even as evil manipulators. As Mayr
and Machin (2012) demonstrated in their analysis of ‘women and crime’, mainstream media
has the ‘tendency to depict female criminals in terms of a few standard narrative frame-
works’ (2012: 135) and lexical choices are (hysterical, mad, monster) ‘ways of linguisti-
cally othering which works to remove the offender from their society, thereby avoiding the
uncomfortable truth that they are produced by that society’ (2012: 116).
Using the tools of visual grammar, I first examined a series of Getty images which repre-
sent visually ‘women criminals’. Images in ‘image banks’ as Machin (2004: 781) suggests,
do not record reality, but put forward ideas about certain people or types and ‘convey par-
ticular kinds of scripts, values and identities’. They are a powerful heuristic tool. By typing
the words ‘women criminals’ a series of pictures of particular kinds of women appear on
the screen and they reveal very deep-seated prejudices and values. Because of copyright
restriction, only two examples are shown from a database of 2,605 pictures (Figures 34.5
and 34.6, gettyimages.com).
In this big database, there are many kinds of pictures of women criminals but in general,
in terms of denotation, I can generalise and say that they present ‘specific social types’

518
 Representations of women criminals

Figure 34.6 ‘Woman criminal’. Free istock images at photo.com by Getty.images (royalty free).

Figure 34.7 ‘Woman criminal’. Free istock image at photo.com by Getty.images (royalty free).

519
Carmen Rosa Caldas-Coulthard

placed most of the time in vague contexts with the intention of ‘documenting’ a reality. In
terms of distance, the pictures are shot from close or medium distance and at horizontal
angles, so the distance between the participants is mitigated.
The visual choices used by the photographers of women criminals index meanings of
transgression and confrontation because they transmit, in terms of code orientation, a natu-
ralistic (we believe that the women are criminals because of their threatening faces) and
sensory code orientation (indicated, in the colour image – not reproduced here – by red and
pink lipstick, and red and black clothing, and by the accompanying gestures).
These transgressive identities are materialised through the presentation of ‘demand’(the gaze
to the viewer) pictures, which interpellate the viewer. The gaze of the represented participant is
directly and horizontally addressed to the viewer, connoting danger and in a sense, power.
As pointed out earlier, just like in language, image producers can present positive or
negative meanings, and images of people evoke emotional reactions. Through the semiotic
system of evaluation which is the attachment of values to people, things, and action, Martin
and White (2005: 44–56), in their ‘appraisal theory’, label the world around us as an ‘evalu-
ative disposition of stance’. One of the main functions of evaluation is attitudinal: we use
language or other semiotic resources to assign values of praising or blaming. For the author,
there are three types of attitude:

affect or the emotional evaluation of things, processes or states of affairs (e.g., like/
dislike);
judgement or the ethical evaluation of human behaviour (e.g., good/bad)
appreciation or the aesthetic or functional evaluation of things, processes and states of
affairs (e.g., beautiful/ugly, useful/useless).

Social judgement is particularly important in visual evaluation since it is the assessment of

human behaviour and it is based on systems of rules, of ethics, of social norms, and of acces-
sibility. We can extrapolate the original goal of appraisal theory concerned with linguistic
interaction and apply it to the visual mode in order to claim that evaluation is in fact a kind
of connotation. Crime, of course, is the breaking of social rules and ethical behaviour and it
is always extremely evaluated both linguistically and visually, as discussed below.
In the Getty data, it is through social judgement and aesthetic appreciation that women
criminals are evaluated visually. The semiotic resources employed are many, all connot-
ing rule-breaking through the presentation of props included in the pictures: guns, money,
drugs, lingerie (which connotes intimacy and sexuality), etc.
Colour choices are also very telling. We know through the study of colours (Van Leeuwen
2011), that they have many different meanings according to time, culture, history, etc.
‘Colour systems relate to religions or empires … or the global system that underlies today’s
use of colour in the creation of a lifestyle and corporate identities’ (2011: 15). Colour there-
fore, together with the other semiotic systems, is an integrative part of linguistic and visual
resources that make up a message and all colours have meaning potential and affordances
which will only be ‘narrowed down, made more specific in specific cultural and situation
contexts’ (Van Leeuwen ibid.: 58–59).
Another very important aspect of colour meaning is its association with particular identi-
ties and style as raised by Van Leeuwen (ibid.). I want to claim that even the choice of colour
is evaluative.
In the Getty data analysed, it is interesting to note that criminal women are most of
the time young and their bodies are fragmented. The dressing colour code relies on

520
 Representations of women criminals

non-chromatic colours, grey and black, and chromatic primary reds and yellows (Van
Leeuwen 2011: unnumbered page, between 52 and 53). Because of the choice of saturated
reds (which have a symbolic value of intensity and danger) and black, the overall message is
that women criminals are sexy (the famous little black dress or grey dress connotes sexuality
and elegance). And these of course convey danger and transgression. The choice of colour
no doubt affords these meanings.
Another important point is that, although the topics associated with the representation
have to do with deviance and drugs, the overall meaning potential is related to sexuality.
Sex, in one form or another (bodies, gestures, clothes), is always alluded to. As a conse-
quence of these choices, the viewer can interpret the representations as problematic.
In the popular Brazilian press, the situation is very similar to the Getty images world.
The news stories and the images I examined are of real people involved in real crimes. See
examples in Figures 34.8 and 34.9.
A woman called Elize Matsunaga killed her abusive husband, first by shooting him, then
sawing him into pieces, and putting his body parts into a suitcase. This was a national news
event specially because it was committed by a woman in a terrifying way. The interesting
aspect of the case for us is the ways the murderer was presented linguistically and visually.
Elize Matsunaga2 was described linguistically in the headline of the news as:

SexElize Matsunaga, who killed her husband Marcos Matsunaga, was a call girl.
Really. Eliza Matsunaga was a high-class prostitute registered on the Prostitution web
site McClass.

Elize is linguistically classified as a prostitute and as a ‘femme fatale’. The implication

therefore is that sexuality and criminality are intertwined. Because she is a prostitute, she
is connotatively placed as a member of the criminal fraternity, so murder is just one more
step although her sexual profession was not at issue in the crime event. Ironically, those who
read the story will discover it was the husband’s sexual betrayal that occasioned his murder.
Interestingly, there is no picture of the mistress.
It is also important to note that the picture of Elize in her lingerie (Figure 34.8) taken
from another context, a call-girl site, and designed originally as an advert for sexual ser-
vices, is resemiotised and used in the news story. This is a very persuasive strategy used by
the newsmakers to convince viewers of her criminality.
Visually, modality is realis (this is her!), her body is fragmented and semi-naked, her gaze
is directly pointing to the eye of the viewer (a demand picture), her pose is sexualised, her lips
are red and she is blonde! In the news story, the murderess, who is interacting with the reader,
presents those very interactive features of Getty images (Figure 34.9). This is clearly an excuse
to present a picture of her in her underclothes with the function of selling the magazine and
attracting sensually the viewers. This also invites the reader to construct an alternative narra-
tive. The pictures published are not chosen primarily for their news values or story relevance,
but to titillate readers who incidentally judge and condemn the represented actor.
A point that is worth mentioning here is that the Brazilian population is very mixed in
terms of skin colour, but whiteness and blondness are desired aspects in the construction
of positive identities. Paradoxically therefore, although in reality Black women are in the
majority in the criminal population in Brazil, the few criminal women who are made into the
news, are white and blonde, extremely sexualised, and are labelled as such. So the perverse
construction of stereotypical positive features join criminality and sexuality as given fact.
The next pictures (Figure 34.10) illustrate this point further.

521
Carmen Rosa Caldas-Coulthard

Figure 34.8 Elize in her lingerie (reproduced with the permission of Veja at [email protected]).

These mugshot pictures are of a gang of young women referred to as ‘blonde’ (no names
or other identity markers) who committed ‘lighting kidnapping’ in São Paulo, Brazil. All are
white, blond, and glamorised. Once again, race here intersects with gender, and discrimina-
tion is symbolically explored.
When their identity is described linguistically, they are either described by their physical
attributes (blonde, beautiful) or are given names such as ‘muse’.

Blonde was used to corrupt mayors in a big corruption scheme.3

A last point is that there is a constant association with male criminals and women criminals
are lexicalised through their family connections as ‘misses’, ‘girl-friends’, and ‘lovers’:

Little auntie from Rocinha, ex-wife and sister of drug dealer is put in jail for her asso-
ciation with drugs.4

As Caldas-Coulthard and Moon (2010, 2016) and Caldas-Coulthard (2010) pointed out,
women in news stories are needlessly and standardly reported in terms of their physical
attributes (age, beauty) or classified according to female roles in society (mother, wife,
prostitute, grandmother, etc.). Women criminals, in crime news stories, continue to be

522
 Representations of women criminals

Figure 34.9 Elize the murderess (reproduced with the permission of Veja at [email protected]).

categorised as social actors in relation to their male partners as wives, girlfriends, sisters,
etc. Even in crime, the constant dependency on a male partner continues to be reinstated.
To sum up the analysis, in visual representation of female actors in news as narrative
texts, women criminals tend to be represented in realis modalities, interacting directly with
viewers and therefore connoting anti-social behaviour.
Through sexualised images, the underlying ideology is that female criminals break social
norms and should be condemned twice – by their criminal acts as well as by their sexuality
or simply by the fact that they are women.

Conclusions
The examples discussed in this chapter have demonstrated a strong bias against female
suspects since their representation is construed through very sexist points of view. Women
continue to be associated with men, and in this sense, they are yet again agentless. Sexuality,
youth, and beauty are given aspects of the construction of characters not because criminality
is a social issue that needs to be tackled by state policies but simply because it sells stories.
The outcome of this representation is the fictionalisation of criminal activity. They in fact
portray a very different resemiotisation of the ‘real’ women criminals that are in prison –
these are poor, Black, not physically attractive, and hopeless.
The visual analysis has suggested that the representation ‘encourages readers to draw
misleading conclusions about female offenders and limit critical inquiry of the story’ (Mayr

523
Carmen Rosa Caldas-Coulthard

Figure 34.10 Members of a gang of young women, referred to as ‘The Blondie Gang’
(‘Guangue das Loiras’) (DHPP/Divulgação/VEJA.) veja.abril.com.br (22nd March
2012, 16h30).

and Machin 2012: 135), especially in the Brazilian context where criminality is an enor-
mous social problem that should never be glamorised.

Future directions
Gender and language, as an area of discourse studies, has evolved tremendously in recent
decades and our research, as linguists and semioticians, has contributed in many ways to the
dissemination of awareness about gender and discourse issues, focusing especially on the
discursive constructions of the social. Although it is undeniable that significant advances
have been made through the work of many different social movements, there are still many
examples of domination and exclusion based on people’s gender identities in our societies
and their discursive practices.
Sexism, racism, and ageism in discourses in different private and institutional contexts
have not disappeared. In many parts of the world, there is a considerable backlash in terms
of conservative views on gender issues, materialised in different ways for meaning-making,
in other words, in different semiotic resources, and a growing political influence which aims
at tightening restrictions on issues such as abortion, gay marriage, and secular education,
among many others. Gender inequalities continue to exist and violence is practised against
minorities on a daily basis.
Linguistically mediated violence for us is a central topic and we must continue to explore
the different and subtle ways sexism and violence are materialised in discourse.
As I hope to have demonstrated in my analysis, semiotic resources like: location of ele-
ments and props, modality, proximity from the viewer, gaze, angles, and linguistic evaluation
help to create fictional personas that are not congruent with the brutal reality of daily crime.
The serious stereotypes constructed through modes of representation produce derogation

524
 Representations of women criminals

and exclusion. In the representation of crime, gender imbalance continues to operate with
other axes of inequality. Future critical readings of other institutional discourses related to
crime (in the legal and the medical professions, among others) can help to question and
destabilise these axes of injustice. And more research is still needed.

Notes
1 All the translations from Portuguese into English are my own.
2 SExElize Matsunaga, que assassinou o marido Marcos Matsunaga, era garota de programa!Isso
mesmo! Elize Matsunaga era prostituta de luxo do site de prostituição Mclass!
3 Loira era usada para aliciar prefeitos para esquema de corrupção. https://www.em.com.br/app/noti
cia/nacional/2013/09/24/interna_nacional,452505/loira-era-usada-para-aliciar-prefeitos-para-esq
uema-de-corrupcao.shtml
4 Tiazinha da Rochinha, ex mulher e Irma de traficantes é presa por associacao ao trafico. https://www
.terra.com.br/noticias/brasil/policia/musas-do-crime/

Further reading
Machin, D., Caldas-Coulthard, C. R., and Milani, T. (2016) ‘Doing critical multimodality in research
on gender, language and discourse’. Gender and Language, 10(3), pp. 301–308.
This introductory chapter of Gender and Language opens a discussion of the role of multimodality
in gender studies. For the authors, multimodality is still a fragmented and unconsolidated area but
crucial because it raises critical questions about meanings and gendered identities.
Ruiz, P. (2017) ‘Power revealed: Masking police officers in the public sphere’. Visual Communication
Special Issue: Picturing Protest – Visuality, Visibility and the Public Sphere, 16(3), pp. 299–314.
This paper discusses how journalists constructing news narratives about ‘public demonstrations’
reproduce the ways police distinguish between good and bad/criminal protesters and how images of
violent demonstrators tend to be used to delegitimise political claims. The focus of the paper is on
the ways images highlight acts of concealment in popular print and online narratives and ‘the police’
rather than the protesters become the object of public attention.
Van Leeuwen, T. (2018) ‘More evaluation in critical discourse analysis’. Critical Discourse Studies,
15(2), pp. 140–153.
According to the author, CDA must ‘evaluate’ the results of discourse analysis taking into
consideration the questions of misrepresentation. A critical reading can establish that misrepresentation
results in social inequality. This paper considers therefore very important issues related to any critical
analysis and ethics.

Related topics
Language, gender, and sexuality: reflections on the field’s ongoing critical engagement with the
sociopolitical landscape; anthropological discourse analysis and the social ordering of gender ideology;
gender and sexuality in discourse: semiotic and multimodal approaches; doing gender and sexuality
intersectionally in multimodal social media practices; XML mark-up for nomination, collocation, and
frequency analysis of language of the law.

References
Barthes, R. (1977) Image, Music, Text. London: Fontana.
Bianchini, A. (2011) ‘O crescente aumento do papel da mulher no universto criminal’. Available
at: https://professoraalice.jusbrasil.com.br/artigos/121814124/o-crescente-aumento-do-papel-da
-mulher-no-universo-criminal (Accessed: 28th November 2017).

525
Carmen Rosa Caldas-Coulthard

Brown, S. (2003) Crime and Law in Media Culture. Buckingham: Open University Press.
Caldas-Coulthard, C. R. (2010) ‘Women of a certain age: Life styles, the female body and ageism’. In:
Holmes, J. and Marra, M. (eds.) Femininity, Feminism and Gendered Discourse. Newcastle upon
Tyne: Cambridge Scholars, pp. 21–40.
Caldas-Coulthard, C. R. and Moon, R. (2010) ‘Curvy, hunky, kinky: Using corpora as tools in critical
analysis’. Discourse and Society, 21(2), pp. 1–35.
Caldas-Coulthard, C. R. and Moon, R. (2016) ‘Grandmother, gran, gangsta granny: Semiotic
representations of grandmotherhood’. Gender and Language, 10(3), pp. 309–339.
Caldas-Coulthard, C. R. and Van Leeuwen, T. (2001) ‘Baby’s first toys and the discursive construction
of childhood’. In: Wodak, R. (ed.) Critical Discourse Analysis in Post Modern Societies. Folia
Linguistica, XXXV(1–2), pp. 157–182.
Carradine, E., Cox, P., Lee, M., Plummer, K., and Nigel, S. (2009) Criminology: A Sociological
Introduction. London: Routledge.
Chibnall, S. (1977) Law and Order News. London: Tavistock.
Departamento Penitenciário Nacional. Available at: http://www.depen.pr.gov.br (Accessed: 20th
November 2017).
Espinoza, O. (2002) ‘A prisão feminina desde um olhar da criminologia feminista’. Revista
transdisciplinar de ciências penitenciárias, 1(1), pp. 35–59.
Fairclough, N. (1989) Language and Power. London: Longman.
Fairclough, N. (1992) Language and Social Change. Cambridge, England: Polity Press.
Fairclough, N. (2003) Analyzing Discourse and Text: Textual Analysis for Social Research. London:
Routledge.
Fairclough, N. (2010) Critical Discourse Analysis: The Critical Study of Language. Upper Saddle
River, NJ: Prentice Hall.
Faria, T. D. (2010) ‘A mulher e a criminologia: Relações e paralelos entre históra da criminologia e
a história da mulher no Brasil’. In: do Anais XIX Encontro Nacional do CONPEDI de Pesquisa
e Pós-graduação em Direito, Fundação Boiteaux, Florianópolis, Brazil. Available at http://www
.conpedi.org.br (Accessed: 15th October 2017).
Ferrell, J., Hayward, K., and Young, J. (2015) Cultural Criminology. London: SAGE.
Folha de São Paulo (2017) ‘Opinion: In cases of rape, Brazil resembles India’. Available at: http:/
/www1.folha.uol.com.br/internacional/en/opinion/2017/08/1912251-opinion-in-cases-of-rape-bra
zil-resembles-india.shtml (Accessed: 10th August 2017).
Foucault, M. (1978) The History of Sexuality, Vol. 1. New York, NY: Pantheon Books.
Giddens, A. (1989) Sociology. Cambridge: Polity.
Hall, E. (1968) ‘Proxemics’. Current Anthropology, 9(2/3), pp. 83–108.
Hall, S. (ed.) (1997) Representation: Cultural Representations and Signifying Practices. London:
Sage in association with The Open University.
Halliday, M. A. K. (1978) Language as Social Semiotic. London: Edward Arnold.
Halliday, M. A. K. and Hasan, R. (1989) Language, Context and Text: Aspects of Language in a
Social-Semiotic Perspective, 2nd ed. Oxford: Oxford University Press.
Halliday, M. A. K. and Matthiessen, C. M. I. M. (2014) Halliday’s Introduction to Functional
Grammar, 4th ed. New York, NY: Routledge.
Hartley, J. (1982) Understanding News. London: Methuen.
Howitt, D. (2011) Introduction to Forensic and Criminal Psychology. Harlow: Pearson Education
Limited.
Iedema, R. (2003) ‘Multimodality, resemiotization: Extending the analysis of discourse as multi-
semiotic practice’. Visual Communication, 2(1), pp. 29–57.
Jewkes, Y. (2004) Media and Crime. Key Approaches to Criminology. London: SAGE.
Jewkes, Y. and Letherby, G. (2002) Criminology: A Reader. London: SAGE.
Kress, G. (2010) Multimodality: A Social Semiotic Approach to Contemporary Communication. New
York, NY: Routledge.

526
 Representations of women criminals

Kress, G. and Van Leeuwen, T. (1996) Reading Images: The Grammar of Visual Design. London:
Routledge.
Kress, G. and Van Leeuwen, T. (2001) Multimodal Discourse: the Modes and Media of Contemporary
Communication. London: Arnold.
Lombroso, C. (1895/2004) ‘Criminal anthropology: Its origins and application’, Forum, 20, pp. 33–49,
in Horton, D. and Rich, K. (eds) The Criminal Anthropological Writings of Cesare Lombroso.
Lewiston: The Edwin Mellen Press, pp. 63–82.
Machin, D. (2004) ‘Building the world’s visual language: The increasing global importance of image
banks in corporate media’. Visual Communication, 3(3), pp. 316–336.
Machin, D. (2007) Introduction to Multimodal Analysis. London: Hodder Education.
Machin, D., Caldas-Coulthard, C. R., and Milani, T. (2016) ‘Doing critical multimodality in research
on gender, language and discourse’. Gender and Language, 10(3), pp. 301–308.
Mapa da Violência (2015) ‘Homicídio de Mulheres no Brasil’. Available at: http://www
.mapadaviolencia.org.br (Accessed: 1st August 2017).
Martin, J. R. and White, P. R. R. (2005) The Language of Evaluation: Appraisal in English. New York,
NY: Palgrave Macmillan.
Mayr, A. and Machin, D. (2012) The Language of Crime and Deviance. London: Continuum.
Newburn, T. (2017) Criminology, 3rd ed. London: Routledge.
Ratton, J. L., Galvão C., and Andrade, R. (2011) Crime e gênero: Controvérsias teóricas e empíricas
sobre a agência feminina. Anais do XV Congresso Brasileiro de Sociologia, 26–29 July, Curitiba,
Brasil, pp. 1–15.
Van Leeuwen, T. (2005) Introducing Social Semiotics. London: Routledge.
Van Leeuwen, T. (2007) ‘Legitimation in discourse and communication’. Discourse and
Communication, 1(1), pp. 91–112.
Van Leeuwen, T. (2008) Discourse and Practice: New Tools for Critical Discourse Analysis. New
York, NY: Oxford University Press.
Van Leeuwen, T. (2011) The language of Colour: An Introduction. Abingdon: Routledge.
Weiss, G. and Wodak, R. (eds.) (2003) Critical Discourse Analysis: Theory and Interdisciplinarity.
London: Palgrave.
Williams, K. S. (2012) Textbook on Criminology. Oxford: Oxford University Press.
Wodak, R. (2000) Discourse and Discrimination. London: Routledge.
Wodak, R. and Meyer, M. (eds.) (2001) Methods of Critical Discourse Analysis. London: SAGE.

527
 35
Confdent appearing
Revisiting Gender Advertisements in
contemporary culture

Kirsten Kohrs and Rosalind Gill

Introduction
It is now more than 40 years since Goffman published his landmark study Gender
Advertisements (1979), which changed the way that questions of power and identity in
advertising were understood. Goffman broke with traditional content analytic approaches to
analysing images and instead developed a framework for examining the way that nonverbal
signals communicate messages about gendered power. Working with a largely print-based
sample of adverts, he argued that features such as relative size, posture, and touch conveyed
important messages about social value and authority. For example, relative to men, women
were typically presented as smaller, lower, and adopting deferential or ‘canting’ postures.
Moreover, whilst men were shown using the products they advertised – and hence their
touch was coded as ‘functional’ – women were frequently depicted in a dreamy far-away
kind of ‘licensed withdrawal’. Far from being idiosyncratic, Goffman argued, such repre-
sentations were highly patterned and deeply connected to unequal gendered power relations.
Indeed they operated symbolically as a kind of ritualisation of female subordination.
Goffman’s work has become a reference point for almost all subsequent scholars interested
in gender and visual language, including those in linguistics, semiotics, and the developing
field of multi-modal analysis. His analyses had the force of an instant ‘recognition factor’,
along with vivid examples, and, once apprehended, these patterns of visual domination could
readily be observed to be widespread across consumer culture. Yet Goffman was writing in
a particular place and time – he was engaging with the habitat of images that defined 1970s
North America. How relevant are his analyses today? To what extent do the patterns of ritu-
alised gender subordination he discussed still apply? How might his approach be deployed in
contemporary culture to ‘notice’ – analytically and politically – other gendered patterns?
In this chapter we address these questions in a critical appreciation of Goffman’s work.
Drawing on a sample of 200 advertisements (‘adverts’) placed in upmarket women’s maga-
zines we examine whether his claims still hold true, questioning his arguments about size,
posture, touch, and gaze, and raising new questions about how gender is ‘done’ and com-
municated today. While there are still many examples of adverts that adopt conventions
that Goffman would recognise, the visual landscape of gender is changing, we suggest, in

528
 Revisiting Gender Advertisements

the wake of increasingly well-documented shifts to what is variously understood as post-

feminism, commodity feminism, or neoliberal feminism, i.e., a mediated multiplicity of
contradictory new and revitalised feminist ideas which range from construction of feminism
as stylish, cool, and fashionable identities to political activism. Our focus here is on one
particular iteration of this – the figure of the ‘confident woman’, who, rather than being pre-
sented as small, passive, or deferential, is depicted as bold and powerful. Drawing on a wider
literature in gender studies we seek to interrogate the focus upon ‘confident appearing’, ask-
ing how – i.e. by what visual and textual means – it is produced, what it communicates,
and what it does ideologically and performatively. The chapter’s aims are thus threefold: to
critically engage in debates about how gender materialises in contemporary advertising; to
examine a novel but increasingly dominant trope for the representation of (some) women,
locating this in debates about new femininities; and to contribute both theoretically and
methodologically to analysing the vocabularies of gender in use in visual culture.

Gender and advertising

Advertising has long been a focus of feminist analysis. One reason for this is the sheer vol-
ume of adverts in contemporary culture. As long ago as 2000, it was estimated that the aver-
age North American sees approximately 3,000 advertising messages every day (Kilbourne
2000) or, to put it another way, spends around three years in an average lifespan engaging
with this material. Such figures represent a substantial under-estimate today where adverts
greet us from every conceivable platform and where their ubiquity signals a blurring – if
not a complete breakdown – of the boundaries between advertising and other media. In our
‘promotional culture’ (Wernick 1991) and world of omnipresent branding, commercial con-
tent has ‘gone native’, seamlessly blurring into editorial content in magazines and newspa-
pers, listicles on Buzz Feed, tweets, vines, and blog posts ‘that never mention their corporate
connections’ – in what Einstein (2016) calls ‘black ops advertising’. Advertising is viral,
heavily watched as entertainment via YouTube videos, is endlessly discussed and spoofed,
or may not even be recognised as a commercial message – as when the drink company, Red
Bull, set up a space dive that was reported as news.
The influence of advertising has been compared to that of education and organised reli-
gion, a vast body of material that shapes our media and is at the heart of social existence
(Jhally 1987). However, it is not simply its ubiquitous nature that has generated feminist
interest: adverts have also garnered critique because of their reliance upon gender ideolo-
gies (Jhally 1987); their condensed nature makes them sites of crude stereotypes and some
of the most egregious examples of sexism in media culture. They are also the locus of new
meanings and figurations of gender – deeply implicated in constructions of the ‘new man’
and ‘new father’, for example, and in contemporary representations of trans and non-binary
genders, as well as in popularising distinctively postfeminist depictions of women as active,
empowered sexual subjects (as we discuss below).
Early research on gender and advertising used a quantitative content analytic approach.
This involves coding, then counting, instances of particular portrayals – for example whether
women were portrayed as working outside the home, where they were depicted, whether
they spoke, etc. A major study published in the New York Times magazine (Hennessee and
Nicholson 1972) prefigured decades of research which showed women were predominantly
shown in the home, depicted as housewives and mothers, often in subservient roles; and rarely
provided argument in favour of the advertised products. By contrast, men were portrayed in

529
Kirsten Kohrs & Rosalind Gill

a wide range of settings and occupational roles; as independent and autonomous; and were
presented as objective and knowledgeable about the products being advertised.
Content analytic studies were immensely valuable in documenting striking patterns of
portrayal and in furnishing quantitative evidence for arguments about sexism in advertis-
ing. However, their weaknesses included the superficiality of the analyses provided – only
focused on manifest content, not underlying ideas or ideologies – and the tendency to gener-
ate very similar findings. This was in part an artefact of the methodology itself, since in order
to document instances of something (e.g. a stereotype) it has to have been pre-identified. In
this sense such studies could only tell us what we already ‘knew’, supplying evidence about
the extent of use of a representation rather than anything about it.
Goffman’s (1979) book – alongside emerging work in the semiotic tradition (Coward 1984;
Williamson 1978) – interrupted this depressing litany of statistics about the preponderance of
women depicted as housewives, shown only in the kitchen or bathroom, and portrayed solely
for their physical attractiveness, offering a new approach grounded in an interest in nonverbal
communication. According to Goffman’s dramaturgical perspective individuals are constantly
and routinely putting on a performance in public space to convey culturally distinctive infor-
mation, which allows strangers a glimpse of their lives. Goffman called these performances
displays: ‘all of an individual’s behaviour and appearance informs those who witness him
[sic], minimally telling them something about his social identity, about his mood, intent, and
expectations, and about the state of his relation to them’ (1979: 1). As displays become well
established in a particular sequence, they can be taken out of the original context and ‘quoted’
in the ‘make-believe scenes in advertisements’; advertisers are thus able ‘to use a few mod-
els and props to evoke a lifelike scene’ (Goffman 1979: 3, 23). Advertisements are carefully
choreographed, thereby providing a heightened, aspirational version of reality grounded in a
broad cultural context. Thus, the representation of identity in print advertisement can be con-
sidered a rich source of data for social analysis of the performance of gender.
In Gender Advertisements (1979), Goffman explores the depiction of women, primarily
in print advertising, from a fresh perspective offering insights on taken-for-granted gender
codes based on his understanding of display, i.e. an individual’s nonverbal behaviour (body
language) and appearance. His analysis of gender display focuses on and reveals gender role
stereotyping as well as patterns of submission. He argued that adverts frequently depict ritu-
alised versions of the parent–child relationship, in which women are largely accorded child-
like status. In the adverts he analysed, women were typically shown lower or smaller than
men and using gestures that ‘ritualised their subordination’, for example lying down, using
bashful knee bends, and deferential postures and smiles. Women were also often depicted
in licensed withdrawal (dreamy self-absorption). Clear differences in gendered touch were
also identified. While men’s touch was functional and instrumental, women’s was light and
caressing, often having no purpose at all. As Gill (2007a: 80) has argued elsewhere, this
key difference can be seen vividly in adverts for shower gel: men would be shown as lath-
ering up busily, while women were – and arguably still are – routinely depicted ‘making
only a small circular movement on one shoulder’. Similarly, women were shown touching
themselves frequently, particularly on the face, and were also depicted running their fingers
gently along a range of products from perfume bottles to sanitary pads.

Reading Goffman: a critical appreciation

Goffman’s genius lay in framing the reader’s perception of gender display in advertis-
ing in skilfully chosen samples of mainly print advertisements. He was diffident about

530
 Revisiting Gender Advertisements

quantifying his insights. He acknowledged freely that in terms of the methodological

question of ‘discovery, presentation, and proof’ (Goffman 1979: 24), he only addresses
the first two. Yet ‘his observations were extremely insightful and full of resonance for
most readers’ (Bruce and Yearly 2006: 126). A ‘shock of recognition’ (Manning and Smith
2010) testifies to the relevance of his findings. Guided by Goffman’s observations, the
reader performs an ‘instructed reading’ (Smith 1996), scrutinising the visual evidence
provided in the form of pictures which visually corroborate Goffman’s written descrip-
tions of the underlying pattern.
Forty years after its publication, the book remains central to studies of gender in adver-
tising. Goffman’s analytic framework is a point of departure for the analysis of gender and
the body’s presentation in advertising. One of the major insights of subsequent feminist
work has centred on how ‘cropping’ is used in adverts. Many studies have highlighted
the way in which women’s – and now increasingly men’s – bodies are fragmented in
adverts, visually dissected so that the viewer sees only the lips, the breasts, the bottom,
etc. Kilbourne (1999: 278) compellingly argues that this works as a strategy of objectifi-
cation, and that ‘turning a human being into a thing, an object, is almost always the first
step towards justifying violence against that person’. Cortese (1999) further extended
Goffman’s ideas by documenting examples of what he calls the display of ‘mock assault’
in adverts, such as an advert for a Karl Lagerfeld perfume in which a frightened looking,
sexy woman is portrayed backed against a wall, while a large, muscular man (seen only
from behind) bears down on her. Similar examples remain abundant, particularly in a
context in which advertisers have to use ever more visually arresting imagery in order to
stand out in a crowded mediascape.
But are Goffman’s arguments still relevant today? After decades of feminist activism
and large-scale social transformation, to what extent are these patterns of ritualised subor-
dination of women still evident in advertising? Meta studies (i.e. studies about other stud-
ies) have produced contradictory findings (Eisend 2010; Hall et al. 2005; Wolin 2003).
Goffman’s framework is not easy to operationalise and his categories are not necessarily
mutually exclusive or exhaustive (Smith 1996). For example, the physical placement of a
woman behind another figure may signal coyness and dependence but the effect would be
different if it is a man who is placed in the rear position. Goffman does not specify how the
different effects are achieved but notes that the ‘same verbal description of relative “physi-
cal” position could be equally applied to cover radically different effects. For the effective
reading of his text, the writer depends upon effective viewing by his reader’ (Goffman 1979:
74). Nonetheless, two studies (Belknap and Leonard 1991; Kang 1997) set out to conduct a
conceptual replication of Goffman’s work applying his precise framework. The studies ana-
lysed 1,000 images in 1985, and 500 (approximately 250 in 1971 and 250 in 1991) adver-
tisements respectively. Both found continued evidence of the feminine touch, ritualisation
of subordination, as well as a degree of licensed withdrawal while relative size and function
ranking were infrequent.

Forty years on
Our own research, conducted in 2015, analyses 200 double-page adverts in the upmarket
glossy magazines Vogue (US and UK editions) and Vanity Fair (UK edition). The adverts
featured fashion, accessories, jewellery, perfume, and beauty products and depicted almost
exclusively women. (In order to compare women’s and men’s body language, the sam-
ple was complemented by approximately 100 adverts from the upmarket men’s lifestyle

531
Kirsten Kohrs & Rosalind Gill

magazines Esquire (UK edition) and GQ (UK edition) in which the adverts were equally
highly targeted featuring fashion, accessories, and perfume depicting almost exclusively
men.) Our research takes as point of departure a grounded theory approach, developed
from the work of Glaser and Strauss (1967), which offers a rigorous inductive approach
to building theory from data. In our research we sought to identify the basic units of non-
verbal behaviour in six main categories. These are gaze, posture, gesture, touch, facial
expression, and proxemics, that is, the culturally determined ‘invisible, variable volume
of space surrounding an individual that defines that individual’s preferred distance from
others’ (Griffin 2012: 105). It explores their constituent parts as components of visual
meaning-making.
Before presenting our findings it is important to make some methodological observations
on developing building blocks of meaning-making in visual language. What becomes abun-
dantly clear from an analysis of this type, is the difficulty of deriving meaning from isolated
units of body language in complex social situations. Context is everything. For instance,
contradictory results have been found in interpreting the smile. Less powerful individuals
have been found to both smile more and to smile less – in order to please others, reduce
threat by gathering information, or indicate role conformity in the absence of social control
and status (for excellent detailed analysis see Hall et al. 2005: 901).
Clearly, in advertisements the genre itself also limits the number of meaningful interpre-
tations of a smile, as one would expect it to indicate pleasure or delight regarding the adver-
tised object. Interestingly, the smile is the most frequent facial expression in our sample, but
only presents in just over 10 per cent of the adverts, while in most images there is a neutral
facial expression. When a smile does present, it is cued by proximal indicators, i.e. pointers
that indicate the state of something closely related, such as a feeling of happiness or interest
and liking in social encounters. Interestingly, only 1 image of the 200 in upmarket women’s
magazines presented a facial expression that was not a smile or neutral, namely a frown. In
the absence of contextual information, it is not possible to assign meaning to it – it might
be tiredness or irritation. In other words, facial expression is only readable in the context of
other cues.
Individual units of meaning-making in isolation are polysemic, that is, they can be inter-
preted in multiple different ways. Building blocks of visual meaning such as intimate physi-
cal proximity and bodily contact can signal a close personal relationship. However, this
reading is only valid in conjunction with a cluster of other key units of meaning-making
such as body orientation towards the other, a leaning forward posture, and, frequently,
mutual gaze and friendly facial expression as well as contextual variables. For example, an
intense intimate relationship between a woman and a girl in a Patek Philippe advertisement
(Vogue, UK September 2015), is cued by mutual gaze, smile, body orientation towards one
another, leaning forward, intimate physical proximity, and touch.
The relationship between the actors may thus be read as mother and daughter based on
the socio-psychological notion of a script, frame, or schema, all of which describe ‘struc-
tures of expectation based on past experience’ which ‘help us process and comprehend
stories [and] serve to filter and shape perception’ (Tannen 1984: 179). However, intimate
physical proximity and bodily contact in a different context, such as a crowded urban envi-
ronment, reveal no such meaning, but can be observed among strangers, as for example in
the ‘Rush Hour Commute’ (Vanity Fair, UK September 2015) or ‘Afternoon Sun’ (Vogue,
UK November 2015) executions of the MiuMiu Eyewear ‘Subjective Reality’ campaign
(Figure 35.1).

532
 Revisiting Gender Advertisements

Figure 35.1 MiuMiu ‘Subjective Reality Afternoon Sun’ (Vogue, UK 2015). Image courtesy of
The Advertising Archives.

In this context, the cluster is made up of variables dissonant with a close personal rela-
tionship, namely, lack of eye contact, neutral facial expression, and body orientation away
from one another.
Our sample does not confirm Goffman’s analysis in terms of height, relative size, gen-
dered postures, or canting (deferential) positions, though it does point to some persistence
of a particular form of ‘feminine touch’ – light, caressing, and frequently directed at the self.
There are no incidences of function ranking as none of the images show one person instruct-
ing or serving another, superior roles, or even professional/working environments as indica-
tors of the ritualisation of subordination. Nor are there incidences of licensed withdrawal, a
dreamy self-absorption. The direct gaze in most images levelled at the viewer, the confident
stance or gait, heads held up high, give the women in the advertisements a strong presence
and signal confidence: ‘holding the body erect and the head high is stereotypically a mark
of unashamedness, superiority, and disdain’ (Goffman 1979: 40). This is the case in almost
all of the advertisements. In fact, the women might even be almost naked, as for example in
the YSL Rouge Pur Couture advertisement (https://cdn.newsapi.com.au/image/v1/4576b83
b1cd81b6c3517c91ec481e4bc) where the female model, Cara Delevigne, is only partially
covered by a jacket, yet her relaxed posture, head held high, line of sight looking down on
the viewer, signal unashamedness, superiority, and disdain.
One exception to this bold, confident representation of women might be the Gucci
2015/16 ‘Cruise’ campaign (Vogue, US/UK November 2015) which appears to show a
social setting in grand interiors (Figure 35.2). However, the avoidance of eye contact and the
overall stiff body language of all the female actors makes them appear like puppets rather
than humans. This suggests an edgy, ‘arty’ set of meanings.
The visuals Goffman analysed revealed a ritualisation of subordination which was
partly effected via height relationships: ‘A classic stereotype of deference is that of
lowering oneself physically in some form or other of prostration’ (Goffman 1979: 40);

533
Kirsten Kohrs & Rosalind Gill

‘elevation seems to be employed indicatively in our society, high physical place symbol-
ising high social place’ (Goffman 1979: 43). However in our sample, relative size (height
relationships) was not relevant as there were few advertisements that pictured both men
and women. When advertisements featured a diversity of genders men were taller than
women only in as far as men typically have a greater body height than women, such as
in the group portraits of the Dolce & Gabbana #DG Mamma, the MiuMiu ‘Subjective
Reality’ campaigns, and a Moncler and Kate Spade advertisement. Their placing and the
hold of their bodies did not accentuate this. Indeed it is more frequently the man rather
than the woman who physically lowers himself. In a Gucci advertisement, cropped so that
one can only see the upper bodies, a man with a naked torso is lying on a bed, while a fully
dressed woman is leaning in with her arm resting across his body (GQ, UK September
2015; Vogue, UK September 2015), inverting what might be expected from Goffman’s
analysis (Figure 35.3).
This trend is reflected across our dataset in which, overall, rather than appearing small,
passive, or deferential, women are presented as bold, confident, and powerful, with strong
and assertive patterns of looking. We want to argue that a new kind of visual language is
being developed to address (particularly, though not exclusively) the magazines’ readership
of middle- and upper-class female subjects. These women are being hailed through a com-
posite of signifiers of assertiveness, boldness, and power that together comprise a kind of
‘confident appearing’. It is to this that we turn in the remainder of this chapter, first locating
this new figure in debates about postfeminism, new femininities, and confidence culture
(Gill and Orgad 2015), and then in the next section examining examples in contemporary
advertising.

Figure 35.2 Gucci (Vogue, US 2015). Image courtesy of The Advertising Archives.

534
 Revisiting Gender Advertisements

Figure 35.3 Gucci (GQ, UK 2015). Image courtesy of The Advertising Archives.

Advertising and postfeminism

Since the 1990s, a number of different writers have sought to make sense of changing
constructions of gender in advertising, with a significant body of scholarship interested in
exploring the impact that feminism has had on advertising. Goldman (1992/2000) coined
the term ‘commodity feminism’ to refer to the way that advertising has sought to harness
and appropriate the cultural power of feminism, while neutering its critique of media. From
this perspective, advertising can be thought of as simultaneously using, incorporating, revis-
ing, attacking, and depoliticising feminism. Feminism may be presented as a ‘style’ or used
as a branding strategy. Ideas about – for example – bodily autonomy or reproductive rights
can be emptied of their political significance and sold back to us as things to consume. As
Douglas (1994: 247–248) put it:

[A]dvertising agencies had figured out how to make feminism – and anti feminism –
work for them […] the appropriation of feminist desires and feminist rhetoric by Revlon,
Lancome and other major corporations was nothing short of spectacular. Women’s lib-
eration metamorphosed into female narcissism unchained as political concepts like lib-
eration and equality were collapsed into distinctly personal, private desires.

Relatedly, advertising often seeks to make a suture or articulation between feminist ideas and
more traditional versions of femininity. Lazar’s (2006: 505) work has been valuable in explor-
ing what she calls ‘power femininity’ – a ‘subject effect’ (Butler 1990) of popular (post) femi-
nism ‘which incorporates signifiers of emancipation and empowerment as well as circulates
the assumptions that feminist struggles have ended … full equality has been achieved and
that women of today can “have it all”’. Her recent analyses of advertising (Lazar 2017) have
centred on the presentation of femininity as labour which must nonetheless be presented as
freely chosen, playful, and fun. In these tropes, feminist critiques of power are repudiated and
an imaginary feminism is indicted as ‘censorious’, ‘uptight’ and ‘old school’, and set against
the empowered, pleasurable, feminine subjectivities on offer in advertising.

535
Kirsten Kohrs & Rosalind Gill

McRobbie’s (2009) important research on the ‘aftermath of feminism’ develops a similar

line of argument, exploring the entanglement of feminist ideas with other ideas and prac-
tices, including commercially produced femininities. In turn, Gill and Scharff’s (2011) book
New Femininities examines the way a postfeminist and neoliberal sensibility offers up novel
constructions of gender, structured by notions of agency and entrepreneurship.
All this research contributes to an understanding of the distinctive form of ‘confident
appearing’ we have identified in contemporary advertising. Two other developments are
also illuminating. One is the shift, discussed by Gill (2003, 2007b, 2008) towards tropes of
empowerment in advertising targeted at women, in which women are invited to purchase
everything from sanitary protection to coffee as signs of their power and independence
(from men). This kind of advertising has four themes: ‘an emphasis upon the body, a shift
from objectification to sexual subjectification, a pronounced discourse of choice and auton-
omy, and an emphasis upon empowerment’ (2008: 41).
What is key for our argument here is the way that this shift contributes to the depiction
of a new independent female subject who is presented as playful and in control. Where once
sexualised representations of women in the media presented them as passive, mute objects
of an assumed male gaze, today women are presented as active desiring sexual subjects, it
would seem, in adverts targeted at both men and women. For instance, the left-hand side of
the Gucci advert (Figure 35.3) showing the man in the ‘deferential’ position was used both
in the men’s lifestyle magazine GQ and in British Vogue.
Bra advertising was a key site of this shift, moving dramatically from earnest discus-
sions of the hold and firmness of girdles and from constructions in which women were
depicted as wrapped in lingerie as a gift for men, to a novel feisty tone from the mid-
1990s onwards. For example, model Eva Herzigova, clad in a push-up bra, hailed us
provocatively with the slogan ‘Or are you just pleased to see me?’ – leaving out the punch-
line of this famous Mae West quotation for viewers to fill in themselves (Figure 35.4).

Figure 35.4 Wonderbra (1990s). Image courtesy of The Advertising Archives.

536
 Revisiting Gender Advertisements

This was no passive, objectified sex object, but a woman who was knowingly playing with her
sexual power. Similarly, the confident, assertive tone of a Triumph bra advert from the same
period is quite different from most earlier representations: ‘New hair, new look, new bra. And
if he doesn’t like it, new boyfriend’. Following suit, other ads stressed women’s independ-
ence from men (‘If he’s late you can always start without him’) and their power (‘I pull the
strings’).While there were numerous problematic features of this trend within advertising –
their heterosexual exclusiveness, the emphasis upon the ‘perfect’ desirable body as women’s
source of power – they were distinctive in instantiating a new more assertive vocabulary in
advertising that is a crucial part of the genealogy of what we call ‘confident appearing’.
Another key trend central to the emerging modality of ‘confident appearing’ is the rise of
what has become known as ‘Love your body’ (LYB) advertising. This exhorts women to get
‘comfortable in their own skin’, to ‘believe in yourself’, and materialises with taglines such as
‘You are more beautiful than you think’(Dove) or ‘Awaken your incredible’(Weight Watchers).
LYB discourses are, in part, a response to feminist anger at what has been seen by many as
unrealistic and harmful images of female beauty, and as such, join the list of ways in which the
advertising industry has been forced to adapt. They interrupt the almost entirely normalised
hostile judgement of women’s bodies, with more positive and celebratory messages focused
upon forging links between body confidence, self-esteem, and the products being advertised.
While the mediated public sphere is full of posts testifying to women’s relief and joy
at these messages, including the emotional power of being encouraged – for once – to feel
okay about themselves (Lynch 2011), there is also a growing critical literature that ques-
tions the supposedly benign ideas upon which it rests (Banet-Weiser 2014; Murphy 2013;
Rodrigues 2012). Critics have pointed to a variety of problems with LYB advertising. These
include the ‘fakeness’ of the LYB visual regime. Many of the companies adopting the ico-
nography of ‘natural’, ‘real’ women and passing it off as ‘authentic’ have been exposed
for using precisely the techniques they claim to reject: photoshop, make up, professional
models, etc. The claimed diversity of the images has also been interrogated. Do they really
depart from most other advertising in terms of showing different ages, ethnicities, body
sizes? Moreover, the adverts have been accused of ‘re-citing’ hate speech e.g. endlessly
circulating fat-phobic discourses in the apparent interest of shutting down ‘fat talk’ (Special
K), and of effacing their own complicity with precisely the negative discourses they claim to
reject. What is striking for our argument here, however, is to note the way that these adverts
have evolved a distinctive visual language for representing the ‘natural’ female body, with
a simple, pared-back aesthetic look, as well as a bold and defiant language centred on self-
esteem as a route to beauty. This can be seen to feed directly into the forms of confident
appearing we have identified.

Confdent appearing
In the visual regime we dub ‘confident appearing’, the elements of LYB have dissemi-
nated beyond bodily appearance to construct a new female subject for whom confidence
is central. This can be seen across popular culture directed at women: women’s magazines
publish confidence issues, smartphone apps promote ongoing confidence programmes, and
policymakers talking about gender inequality highlight a female ‘confidence deficit’ as a
central problem in equalising gender pay and status. As Gill and Orgad (2015) have argued,
confidence has emerged in the second decade of the twenty-first century as a gendered ‘tech-
nology of self’ (Foucault 1988), calling on women to act upon themselves to refashion their
subjectivity around ideas of self-belief. Confidence operates across multiple domains from

537
Kirsten Kohrs & Rosalind Gill

body love to international development and from workplace to finance, and is increasingly
diffused across social life.
A growing body of work examines the ‘confidence cult’ (Banet-Weiser 2015, 2017;
Favaro 2017; Gill and Orgad 2015, 2017; Wood 2017) as well as related formations such
as the ‘happiness industry ‘(Binkley 2014; Davies 2015),‘state of esteem’, or ‘wellness’
syndrome (Cederstrom and Spicer 2015). What this work has not examined, however, is the
way these notions materialise not simply as discourses or ideologies evident in language,
but also as particular visual regimes. It is this that we look at here, since it is highly evident
in our sample of adverts in which, rather than appearing to be subordinated, women seemed
to be bold and confident actors.
The visual elements of this mode of confident appearing involve several repeated fea-
tures: head held high, face turned forward, eyes meeting the gaze of the viewer and looking
directly back at her or him. Smiling is rare, and sometimes the gaze has an almost defiant
aspect. These visual motifs are underpinned by linguistic elements, which highlight female
independence, empowerment, self-belief, and entitlement. A good example is Clinique’s
advert for soap, clarifying lotion, and moisturiser.
Using the face of US feminist, Tavi Gevinson, known for founding the feminist blog Rookie
while a teenager, the advert declares ‘FACE FORWARD’(Vogue, US September 2015). Facing
forward, like ‘facing the world’, is a synonym for confidence. Gevinson’s visage, made up in
a naturalistic style, with her hair swept away from her face and tucked behind her ears, exem-
plifies this idea: her ‘bare’ face looks straight at us with a neutral expression. The confidence
message, in a visual language that would even appear to transcend national borders and oper-
ates across multiple media platforms (https://www.youtube.com/watch?v=QrB4hY4Cel8), is
underscored by the written text which commands:

‘Dress for yourself. Dream big. Find your voice. Put it out there’

These imperatives express autonomy, boldness, empowerment, engagement, or taking on the

world. They encode what Dobson (2014) has called a kind of ‘performative shamelessness’
and what Gill and Kanai (2018) dub ‘hollow defiance’. Like other ‘feeling rules’(Hochschild
1983) circulating in contemporary culture they offer powerful messages of hope and pos-
sibility, wrapped in an upbeat and vaguely defiant sense of self-belief – whose argumenta-
tive target is never specified (i.e. What or who it is that stops you – without Clinique’s help
– from dreaming big or finding your voice?). They sell us a kind of ‘feminist feeling’ (Gill
and Orgad 2017), but one that is devoid of political thinking and action.
Other tropes in the visual construction of ‘confident appearing’ involve control or move-
ment, for example with the figure of the woman striding confidently forward through an
urban landscape. In such representations the stride is typically exaggerated – much longer
than is actually typical of walking – to highlight a sense of a forward-moving woman.
An example found at the time of writing is an advert for Geox (Respira) shoes in which a
woman is depicted striding out, her hands in her trouser pockets, her head held high. M&S
also use this motif in a 2017 campaign, the woman’s hair blowing behind her to signify the
speed and purposefulness of her gait, underpinned by the copy, ‘Always walk tall, even in
flats’. Swatch uses a different motif of movement – this time a highly controlled acrobatic
dance move, that mimics the shape of the watch, several times life size, around the woman’s
body. The slogan declares ‘YOUR MOVE’.
What we see, then, across our corpus is evidence of the way that a distinctive visual
language is being developed for depicting female self-confidence – in ways that break

538
 Revisiting Gender Advertisements

markedly from Goffman’s analysis. This is underscored by a multiplicity of beauty and

grooming commercial confidence messages which exhort: ‘Confidence is the new sexy’
(Bobbi Brown), ‘Take control’ (Braun), ‘Command style’ (Great Lengths), or ‘Feel confi-
dent everyday’ (Charles Worthington). In the latter the confidence message is hardly less-
ened by the inclusion, in much smaller font, of the phrase ‘washing your hair’ between ‘feel
confident’ and ‘everyday’. Confident appearing has become a central and recognisable trope
in contemporary advertising to women, and demands further analysis.

Conclusion
In this chapter we have revisited the important work of Goffman, which offered a com-
pelling way of analysing the visual language of gender in advertisements. Using a sam-
ple of 200 adverts taken from Vogue and Vanity Fair we systematically examined whether
Goffman’s arguments still hold true, 40 years on. Our analysis showed marked disjunctures
from Goffman’s account of the ritualised subordination of women in advertising. Indeed,
looking across a range of features of his analysis, little evidence was found of function
ranking, height marking or lower placing, canting, or deferential postures. ‘Licensed with-
drawal’ was rarely present, and women were rarely shown smiling or in any other way
implying lower status. Indeed, only the sharp differentiations in touch that Goffman identi-
fied were to be seen clearly in our sample.
What was evident, however, was what appeared to be a new visual trope for representing
women: one that we have named ‘confident appearing’. In this, women are depicted with their
heads held high, looking directly at the viewer, with a neutral expression, or pictured striding
purposefully forward, or holding themselves in controlled movement. We have shown how
this has become an established visual motif for representing women, and one that is under-
pinned by written texts that exhort confidence, self-belief, and empowerment. In our analysis
we have sought to locate this in a broader understanding of the way in which advertising has
responded to and engaged with feminism. It seems to have become a way for advertisers to
appropriate some of the meanings and contemporary cultural power of feminism, while locat-
ing it in individualised expressions of self-belief and entitlement rather than calls for social
transformation. In this way a vague sense of feminist defiance can be appropriated, yet emp-
tied out of its political force. Confident appearing has become a key trope for enacting this.

Future directions
One limitation of the analysis presented here is the clear class bias of the magazines we
examined. They are targeted at an upper-class and upper-middle-class readership of women,
predominantly college-educated with above average professional, managerial, or execu-
tive employment living in households with above-average incomes, as the breakdown of
audience demographics shows. The brands advertised are upmarket and aspirational, and
the adverts themselves have very high production values connoting wealth and exclusiv-
ity. Perhaps our general findings about Goffman’s work may not be replicated in media
targeted at a different class demographic. This remains a task for future research to exam-
ine. However, the figure of the confident woman whom we have identified as a key fea-
ture of advertising targeted at women does indeed extend beyond the wealthy demographic
addressed by Vogue and Vanity Fair. It can be seen in adverts by high street brands such as
M&S, Boots, and Braun. Confident appearing, we suggest, is a widespread representational
practice that cuts across class, race, age, and sexuality, and it has become a key way that

539
Kirsten Kohrs & Rosalind Gill

advertisers express a ‘lite’, individualised version of feminism, while offering little chal-
lenge to a wider, patriarchal capitalist culture. Documenting these ideological shifts – which
operate at both visual and discursive levels – remains important work for feminist analysts.

Further reading
Banet-Weiser, S. (2018) Empowered: Popular Feminism and Popular Misogyny. Durham, NC: Duke
University Press.
In Empowered, Banet-Weiser explores the relationship between popular feminism and misogyny in
advertising, online and multi-media platforms, and nonprofit and commercial campaigns. Investigating
feminist discourses that emphasise self-confidence, body positivity, and individual achievement
alongside violent misogynist phenomena such as revenge porn, toxic geek masculinity, and men’s
rights movements, the author traces how popular feminism and popular misogyny are co-constituted.
Elias, A., Gill, R., and Scharff, C. (eds.) (2016) Aesthetic Labour: Beauty Politics in Neoliberalism.
London: Palgrave.
Aesthetic Labour considers questions about gender and the politics of appearance from a new
perspective by developing the notion of aesthetic labour. Bringing together feminist writing regarding
the ‘beauty myth’ with recent scholarship about new forms of work, the authors suggest that in
this moment of ubiquitous photography, social media, and 360-degree surveillance, women are
increasingly required to be ‘aesthetic entrepreneurs’, maintaining a constant state of vigilance about
their appearance.
Hall, S. (1997) Representation: Cultural Representations and Signifying Practices. London and
Thousand Oaks, CA: SAGE.
Hall offers a comprehensive outline of how visual images, language, and discourse work as
‘systems of representation’. Individual chapters explore: representation as a signifying practice in
diverse social contexts and institutional sites; the use of photography in the construction of national
identity and culture; other cultures in ethnographic museums; fantasies of the racialised ‘Other’ in
popular media, film, and image; the construction of masculine identities in discourses of consumer
culture and advertising; and the gendering of narratives in television soap operas.
Orgad, S. and Gill, R. (2019) The Confidence Cult. Durham, NC: Duke University Press.
Confidence Cult explores how confidence has become a ‘technology of self’ that invites girls
and women to work on themselves. It demonstrates the extensiveness of what Orgad and Gill call
the ‘cult(ure) of confidence’ across different areas of social life, identities in discourses of consumer
culture, and advertising; and the gendering of narratives in television soap operas.
Rose, G. (2016) Visual Methodologies: An Introduction to Researching with Visual Materials, 4th ed.
London and Thousand Oaks, CA: SAGE.
Visual Methodologies: An Introduction to Researching with Visual Materials is a critical guide
to the study and analysis of visual culture which examines individual methodologies in a clear and
structured fashion.

Related topics
Multimodal constructions of feminism; gender and sexuality in discourse: semiotic and multimodal
approaches; gender and sexuality normativities; analysing gendered discourses online; feminist
poststructuralism – discourse, subjectivity, the body, and power.

References
Banet-Weiser, S. (2014) ‘Am I pretty or ugly? Girls and the market for selfesteem’. Girlhood Studies,
7(1), pp. 83–101.

540
 Revisiting Gender Advertisements

Banet-Weiser, S. (2015) ‘“Confidence you can carry!”: Girls in crisis and the market for girls’
empowerment organizations’. Continuum: Journal of Media and Cultural Studies, 29(2), pp.
182–193.
Banet-Weiser, S. (2017) ‘“I’m beautiful the way I am”: Empowerment, beauty, and aesthetic labour’.
In: Elias, A. S., Gill, R., and Scharff, C. (eds.) Aesthetic Labour: Dynamics of Virtual Work.
London: Palgrave Macmillan, pp. 265–282.
Belknap, P. and Leonard, W. M. (1991) ‘A conceptual replication and extension of Erving Goffman’s
study of gender advertisements’. Sex Roles, 25, pp. 103–118.
Binkley, S. (2014) Happiness as Enterprise: An Essay on Neoliberal Life. Albany, NY: State University
of New York Press.
Bruce, S. and Yearly, S. (2006) ‘Goffman, Erving (1922–82)’. In: The Sage Dictionary of Sociology.
London: SAGE Ltd (UK).
Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. London: Routledge.
Cederstrom, C. and Spicer, A. (2015) The Wellness Syndrome. Cambridge: Polity Press.
Cortese, A. J. (1999) Provocateur: Images of Women and Minorities in Advertising. Lanham, MD:
Rowman & Littlefield Publishing Group.
Coward, R. (1984) Female Desire: Women’s Sexuality Today. London: Paladin Grafton Books.
Davies, W. (2015) The Happiness Industry: How the Government and Big Business Sold Us Wellbeing.
London: Verso.
Dobson, A. S. (2014) ‘Performative shamelessness on young women’s social network sites: Shielding
the self and resisting gender melancholia’. Feminism & Psychology, 24(1), pp. 97–114.
Douglas, S. J. (1994) Where the Girls Are: Growing Up Female with the Mass Media. Harmondsworth:
Penguin.
Einstein, M. (2016) Black Ops Advertising: Native Ads, Content, Marketing, and the Covert World of
the Digital Sell. New York, NY: OR Books.
Eisend, M. (2010) ‘A meta-analysis of gender roles in advertising’. Journal of the Academy of
Marketing Science, 38(4), pp. 418–440.
Favaro, L. (2017) ‘“Just be confident girls!”: Confidence chic as neoliberal governmentality’. In:
Elias, A., Gill, R., and Scharff, C. (ed.) Aesthetic Labour. Dynamics of Virtual Work. London:
Palgrave Macmillan, pp. 283–299.
Foucault, M. (1988) ‘Technologies of the self’. In: Technologies of the Self: A Seminar with Michel
Foucault. London: Tavistock Publications, pp. 16–49.
Gill, R. (2003) ‘From sexual objectification to sexual subjectification: The resexualisation of women’s
bodies in the media’. Feminist Media Studies, 3(1), pp. 99–106.
Gill, R. (2007a) Gender and the Media. Cambridge, UK: Polity.
Gill, R. (2007b) ‘Postfeminist media culture: Elements of a sensibility’. European Journal of Cultural
Studies, 10(2), pp. 147–166.
Gill, R. (2008) ‘Empowerment/sexism: Figuring female sexual agency in contemporary advertising’.
Feminism & Psychology, 18(1), pp. 35–60.
Gill, R. and Kanai, A. (2018) ‘Mediating neoliberal capitalism: Affect, subjectivity and inequality’.
Journal of Communication, 68(2), pp. 318–326.
Gill, R. and Orgad, S. (2015) ‘The confidence cult(ure)’. Australian Feminist Studies, 30(86), pp.
324–344.
Gill, R. and Orgad, S. (2017) ‘Confidence culture and the remaking of feminism’. New Formations: A
Journal of Culture/Theory/Politics, 91, pp. 16–34.
Gill, R. and Scharff, C. (eds.) (2011) New Femininities: Postfeminism, Neoliberalism and Subjectivity.
Basingstoke: Palgrave.
Glaser, B. G. and Strauss, A. L. (1967) The Discovery of Grounded Theory: Strategies for Qualitative
Research. New Brunswick: Aldine Transactions.
Goffman, E. (1979) Gender Advertisements. London: Macmillan Press Ltd.
Goldman, R. (1992/2000) Reading Ads Socially. London: Routledge.
Griffin, E. (2012) A First Look at Communication Theory, 8th ed. New York, NY: McGraw Hill.

541
Kirsten Kohrs & Rosalind Gill

Hall, J. A., Coats, E. J., and LeBeau, L. S. (2005) ‘Nonverbal behavior and the vertical dimension of
social relations: A meta-analysis’. Psychological Bulletin, 6(131), pp. 898–924.
Hennessee, J. A. and Nicholson, J. (1972) ‘NOW says tv commercials insult women’. New York Times,
p. SM12.
Hochschild, A. (1983) The Managed Heart: Commercialization of Human Feeling. Berkeley, CA:
University of California Press.
Jhally, S. (1987) The Codes of Advertising: Fetishism and the Political Economy of Meaning in the
Consumer Society. New York, NY: St. Martin’s.
Kang, M.-E. (1997) ‘The portrayal of women’s images in magazine advertisements: Goffman’s gender
analysis revisited’. Sex Roles, 37(11/12), pp. 979–997.
Kilbourne, J. (1999) Deadly Persuasion: Why Women and Girls Must Fight the Addictive Power of
Advertising. New York, NY: Free Press.
Kilbourne, J. (2000) Can’t Buy My Love: How Advertising Changes the Way We Think and Feel. New
York, NY: Simon and Schuster.
Lazar, M. M. (2006) ‘“Discover the power of femininity!” – Analysing global “power femininity” in
local advertising’. Feminist Media Studies, 6(4), pp. 505–517.
Lazar, M. M. (2017) ‘“Seriously girly fun!”: Recontextualising aesthetic labour as fun and play in
cosmetics advertising’. In: Elias, A. S., Gill, R., and Scharff, C. (eds.) Aesthetic Labour. Dynamics
of Virtual Work. London: Palgrave Macmillan, pp. 51–66.
Lynch, M. (2011) ‘Blogging for beauty? A critical analysis of operation beautiful’. Women’s Studies
International Forum, 34(6), pp. 582–592.
Manning, P. and Smith, G. (2010) ‘Symbolic interactionism’. In: Elliott, A. (ed.) The Routledge
Companion to Social Theory. Abingdon, Oxon: Routledge, pp. 37–55.
McRobbie, A. (2009) The Aftermath of Feminism: Gender, Culture and Social Change. London:
SAGE.
Murphy, R. (2013) (De)constructing “Body Love” Discourses in Young Women’s Magazines.
Unpublished PhD Thesis. Victoria: University of Wellington.
Rodrigues, S. (2012) ‘Undressing homogeneity: Prescribing femininity and the transformation of self-
esteem in how to look good naked’. Journal of Popular Film and Television, 40(1), pp. 42–51.
Smith, G. (1996) ‘Gender advertisements revisited: A visual sociology classic’. Electronic Journal of
Sociology, 2(1), pp. 1–27.
Tannen, D. (1984) ‘What’s in a frame? Surface evidence for underlying expectations’. In: Freedle, R.
O. (ed.) New Directions in Discourse Processing. Norwood, NJ: Ablex, pp. 137–181.
Wernick, A. (1991) Promotional Culture: Advertising, Ideology, and Symbolic Expression. London:
SAGE.
Williamson, J. (1978) Decoding Advertisements: Ideology and Meaning in Advertising. London:
Marion Boyars.
Wolin, L. D. (2003) ‘Gender issues in advertising—An oversight synthesis of research: 1970–2002’.
Journal of Advertising Research, 43(1), pp. 111–129.
Wood, R. (2017) ‘Look good, feel good: Sexiness and sexual pleasure in neoliberalism’. In: Elias,
A., Gill, R., and Scharff, C. (eds.) Aesthetic Labour. Dynamics of Virtual Work. London: Palgrave
Macmillan, pp. 317–332.

542
 36
Doing gender and sexuality
intersectionally in multimodal
social media practices
Sirpa Leppänen and Sanna Tapionkaski

Introduction
In this chapter, our focus is on how members in participatory cultures, such as fan groups
and communities around user-generated content on social media, engage in informal and
interest-driven social and discursive practices primarily on social media, but in ways that
intertwine with their offline lives. In particular, we will show how social media participants
construct their gendered and sexual identities through cultural practices at specific intersec-
tions with other sociocultural categories.
Before this, however, it is important to understand what we mean by social media, and
why social media practices matter in terms of research on language, gender, and sexual-
ity. Social media can be defined as digital applications that build on the ideological and
technological premises and foundations of Web 2.0 (e.g. Herring 2013), allowing the crea-
tion, exchange, and circulation of user-generated content (Kaplan and Haenlein 2010) and
enabling interaction between users. Social media encompass both applications explicitly,
building on the idea of mutual exchange of content, and digital environments in which the
main content can consist of single-authored or monophonic discourse but that also offer an
opportunity to authors and recipients to interact with one another (such as discussion sec-
tions of blog sites).
Social media are sites for everyday social interaction that intertwine with and comple-
ment physical activities and interactions. In terms of gender and sexuality they also matter
a great deal. They can provide individuals and groups with ‘affordances’ (i.e. enablings and
constraints for behaviour provided by some object or environment to some agent (Gibson
1986)) for doing gender and sexuality in different ways. Sometimes these ways can be novel
and liberating, at other times they can replicate distinctions and power relations that are
typical of other contexts, and at yet other times involve and impose forms of constraint that
are specific to the technological activities and interactions in digital media contexts (see e.g.
Herring and Kapidzic 2015).
In research on how gendered and sexualised identities are constructed on social media,
an approach is needed that can capture their complexity and embeddedness in relations of
power. We suggest that this approach can rely on the notion of ‘intersectionality’. Gender

543
Sirpa Leppänen & Sanna Tapionkaski

and sexuality are always constructed at intersections with other identity categories, and
these multifaceted identity constructions are often part of particular identity politics in rela-
tions, contexts, and formations of power (Hill Collins and Blige 2016). While intersection-
ality has various meanings, in our approach focusing on identity work in digitally mediated
contexts, we argue that an anticategorical approach to intersectionality (McCall 2005) is
particularly useful. In brief, this approach does not take identity categories such as gender or
sexuality for granted but examines how a complex array of various intersecting categories
– and norms related to categories – emerge in everyday contexts of social life. Here, this
will involve studying how participants in social media perform, negotiate, and struggle over
gendered and sexualised identities multimodally by drawing on various meaning-making
resources such as language, other semiotic symbols, and (sub-) cultural scripts that index
specific understandings of gender and sexuality, in particular digital contexts. This happens,
first, by paying attention to the ways in which, in the context of specific participatory cul-
tural practices, people orient to norms of gender and sexuality, by adhering to or rejecting
particular norms, or by positioning themselves critically in relation to them. Second, our
analyses pay attention to the availability of and access to various multimodal meaning-mak-
ing resources: despite the relatively broad access to social media (in the West), not everyone
has the same access to or knowledge of resources used in online identity construction. That
is, it matters who is able to do what kind of performances, where, and when. To examine
the potential inequalities in these practices, we suggest that an ethnographic understanding
of people’s lived realities on- and offline should support the analyses of discourse practices
in social media.

Social media activities and interactions call for a transdisciplinary

framework
In such investigations, we argue, a transdisciplinary framework that draws on discourse
studies, ethnography, sociolinguistics, and the study of multimodality can be particularly
useful. A detailed analysis of social media discourse practices, and social meanings con-
structed in them, combined with the ethnographic understanding of the studied phenom-
enon, can enable the scrutiny of the complex intersectional identity work that can occur
even in a very limited amount of social media data.
Discourse studies offer tools to study the ways in which people do their identities online
through their linguistic and discursive choices. In such analyses attention can be paid to
what kind of language resources participants draw on, what is made visible and what not,
and what kind of norms and discourses related to gender and sexuality they mobilise – dis-
courses here referring to the regulatory practices that constitute for us what counts as truth,
reality, and knowledge (Baxter 2003; Foucault 1978). For its part, immersive ethnographic
fieldwork that can either range from discourse-centred (Androutsopoulos 2008) to multi-
sited (on/offline) ethnography (Staehr 2015) enables a nuanced understanding of the studied
phenomenon from both the emic (insider) perspective of the participants and the analyst’s
etic (outsider) perspective. That is, without understanding the specific social, cultural, and
media contexts of the participatory practices, it may be impossible for the researcher to rec-
ognise the particular sociolinguistic contextual cues (or indices, Silverstein 2006) that the
specific choices have and how these contribute to identity work.
Paying attention to contextual cues (i.e. indexicality) can explain how language users
interpret each other’s, and index meanings in their own linguistic and discursive identity

544
 Doing gender & sexuality intersectionally

construction. Language use, or verbal and textual means – such as languages, varieties,
styles, and genres – are important resources in this. Further, communication can be mono-
lingual or multilingual in nature; and it can also involve code switches as a means for creat-
ing specific discourse-level meanings (Leppänen 2012).
However, besides verbal and textual resources, social media participants can also mobi-
lise resources provided by other modes. In this task, the study of multimodality (e.g. Kress
and van Leeuwen 2006) is necessary as an analytic perspective: it allows the investigation
of how social media users also draw on a range of multimodal and embodied resources for
the production of meaning. These can include textual forms and patterns, visuality, still
and moving images, sound, music, facial expressions, bodily actions, gestures, and perfor-
mances, as well as cultural discourses. They also involve processes of recontextualisation
in which discourse chunks are lifted from their original context and re-embedded in new
contexts (Bauman and Briggs 1990), and of resemiotisation whereby discourse content is
converted from one mode to another/others, from verbal text to (moving) image, for exam-
ple (Iedema 2003).
In the following, we will first discuss critical issues related to intersectionality. We will
touch upon some historical perspectives in relation to intersectionality theory, but will focus
on recent research on language, gender, and sexuality in social media practices. We will then
move on to introduce some current research and recommendations for practice, through
discussing a couple of examples from our recent research projects.

Critical issues: intersectionality as an approach to investigate the

complexity of identity work
The concept of intersectionality was introduced by the legal studies scholar Crenshaw
(1989; 1991) to address the specific positions of Black women whose experiences could
not be explained only through the category of gender or ethnicity, but through looking at
these two identity categories as mutually constitutive of each other. Since the 1980s, the
concept of intersectionality has become central in feminist theory and gender and sexual-
ity studies (see McCall 2005). It has also been adopted by language, gender, and sexuality
researchers (e.g. Levon 2015; Machin, Caldas-Coulthard and Milani 2016; Queen 2014),
especially in the field of queer linguistics (see Bucholtz and Hall 2004; Motschenbacher
2011). Intersectionality theory maintains that power relations, constituted in and through
sociocultural categories such as gender, sexuality, race, class, ethnicity, and age, do not exist
in isolation from each other either in individual people’s experiences or as social structures,
but co-construct one another in a complex manner. While this premise serves as the start-
ing point of intersectionality perspectives in research, there have been different ways to
categorise types of intersections, and various methodological approaches to examine inter-
sectionality in practice.
In the field of sociolinguistics, there are, to date, relatively few studies that utilise an
intersectional approach. However, in a sense, sociolinguistics has addressed the intersec-
tions of different identities from the beginning. In the classic studies, gender and social class
were examined as categories that affect each other and the speakers’ output. This necessity
to pay attention to several identity categories simultaneously has also been a key focus
of sociolinguistic studies of gender later on (see e.g. Eckert and McConnell-Ginet 1999;
Milroy and Milroy 1993). However, especially in variationist sociolinguistics, the different
identity categories have been treated as background variables, rather than as something that

545
Sirpa Leppänen & Sanna Tapionkaski

is situationally constructed through language and other semiotic means. A notable exception
is Levon (2015: 297–298, italics original) who argues that there are three main principles
in intersectionality research that are important for language scholars. These include that
(1) ‘lived experience is ultimately intersectional in nature’ (cf. Crenshaw 1989, 1991); (2)
‘intersections are dynamic, and emerge in specific social, historical and interactional config-
urations’; and (3) ‘these (dynamic) categories not only intersect but also mutually constitute
one another’. These basic tenets can lead to diverse approaches, however. Indeed, in terms
of methodological approaches, intersectionality theory has been criticised for obscurity and
lack of critical discussion of what it actually means to examine the complexity of social
identity construction (see Levon 2015). Dealing with the complexity, or all the possible
identity categories that may be significant in any given socio-historical context, is more
easily said than done.
In her review of (sociological) intersectionality studies, McCall (2005) outlines three
main approaches to intersectionality: (1) anticategorical complexity, (2) intracategorical
complexity, and (3) intercategorical complexity. The first approach ‘rejects categories’ and
aims at deconstructing them to capture the multiplicity, fluidity, and complexity of social
life (McCall 2005: 1773). The second one, intracategorical complexity, refers to approaches
such as Crenshaw’s (1991), which aims at questioning the boundaries of macro-categories
by examining the diversity inside a larger category (e.g. gender) that occurs when other
identity categories (e.g. ethnicity) are considered in relation to it (McCall 2005: 1774).
Finally, approaches to intercategorical complexity do not aim at questioning categories but
require ‘that scholars provisionally adopt existing analytical categories to document rela-
tionships of inequality among social groups and changing configurations of inequality along
multiple and conflicting dimensions’ (McCall 2005: 1773).
Our approach falls under the category of anticategorical complexity: we are interested in
the complex, often messy, and unexpected intersections of gender, sexuality, and other iden-
tities in social media and the ways in which they tie up with particular identity politics. In
this, we also follow Levon’s (2015: 303) suggestion that analyses of intersectional identity
performances should not be done by ‘simply adding these other categories into the empirical
mix but instead by centring our analyses on the social, historical, ideological, and linguistic
relationships between these categories and the different lived articulations of gender and
sexuality we study.’ What this means in practice is that although we start with certain cat-
egories – here, gender and sexuality – we do not approach our data with a checklist of other
potentially intersecting categories (e.g. ethnicity, religion, class) but pay attention to those
intersections and identities that emerge from the data. That is, we will empirically pay atten-
tion to the ‘lived articulations of gender and sexuality’ that may vary from one social media
context to another. In this, we will focus on a range of semiotic resources.
But what exactly are these ‘lived articulations’ and how can we examine them through
a discourse-analytical and ethnographic framework? We suggest that this can be done by
focusing on the ‘performativity’ of gendered, sexualised, and other identities that can, again,
be analysed through the indexicality of language and other semiotic resources. In other
words, we will examine how people mobilise different semiotic resources in their presenta-
tion of the self and/or others in social media, (cf. Halonen and Leppänen 2017; Lehtonen
2017).
In Butler’s (1990, 1993, 2004) theory, performativity refers to the embodied perfor-
mances of gender and sexuality that through repetition begin to look as if they are natu-
ral and self-evident. Apart from embodiment, gender and sexuality are performed through
positioning of the self or others in discourse where performativity refers to ‘a practice of

546
 Doing gender & sexuality intersectionally

improvisation within a scene of constraint’ (Butler 2004: 2). This echoes the understand-
ing of performance from the perspective of the Goffmanian root metaphor ‘life is a stage’
(Goffman 1959), according to which all human actions are seen as performances, or presen-
tations of the self, aimed at particular audiences, in particular spatio-temporal settings, and
based on particular cultural values, norms, and beliefs. In this respect, social media sites are
not unlike a stage in which the performers never know the whole audience but on which
they still need to perform in ways that make sense to and are interpretable by their audience.

Current research and recommendations for practice

In our discussion of the nexus of multimodal participatory cultural practices on social media
and intersectionality, we will highlight how, in order to capture the multi-sited and entan-
gled nature of how gender and sexuality are constructed in the lived realities of participants,
the analysis needs to rely on insights provided by a range of disciplines. We will illustrate
this with the help of examples from our previous studies. The discussion of these examples
here only involves certain key observations about them to illustrate the complexity of the
multi-semiotic practices that participants rely on in their social media practice – a full analy-
sis of each case would require more space than we have.
However, our analytic approach to the investigation of how participants on social media
perform their gendered and sexualised identities involves a number of analytic tasks (see
also Leppänen and Kytölä 2016). First, the researcher should carefully contextualise the
fan/activity culture in focus socioculturally. Second, to achieve a nuanced understanding
of its practices and their meanings to the participants, they need to observe the culture in
a systematic ethnographic way. Third, the study needs to involve a close analysis of how
social media participants mobilise particular linguistic and other semiotic resources and
how these choices index specific understandings of gender and sexuality. Finally, the study
also needs to consider how in their semiotic practice, the participants orient to norms of
gender and sexuality, as well as pay attention to the availability of and access to meaning-
making resources and the potential inequalities in this.

Doing young alternative masculinity intersectionally

Our first example comes from a Finnish discussion forum for ‘bronies’ (a blend of ‘brother’
and ‘pony’), that is, young adult male fans of the children’s animated television series ‘My
Little Pony: Friendship is Magic’ (2011–, from here on MLP). With the help of this example
we will illustrate, first, the unexpected social media contexts where gender and sexuality
are performed and, second, the complexity of intersecting identity performances, realised
through the use of various semiotic resources that index particular meanings and are conse-
quential for the actors’ sense of self and agency in this participatory online culture.
According to Hasbro Studios’ website (2015), the target audience of the MLP series –
based on Hasbro’s line of toys – are 4- to 7-year-old girls. Against marketing and cultural
expectations, a substantial part of the fans of the series have been young (adult) men, calling
themselves ‘bronies’ (see Lehtonen 2017). Bronies themselves explain their attraction to the
series by the ‘soft’ values that it celebrates – friendship and tolerance – and find their fan-
dom (fan community) very supportive; many of them have made long-term investments in
building up an active on- and offline community. The fandom has, however, attracted mixed
media attention because it challenges hegemonic gender and age discourses: men who admit
to liking media products targeted at girls are often met with ridicule (Jones 2015: 121), if not

547
Sirpa Leppänen & Sanna Tapionkaski

pathologisation. Bronies have inspired ‘antifandom’ practices both online and in traditional
media where they are considered disturbing, immature, or sexually deviant (Jones 2015).
In Finland, the community has been active since 2012 both on- and offline. In the Finnish
mainstream media, bronies have been treated as a group that deviates from gendered and
aged norms – although they are not represented as disturbing, they are viewed as strange,
exceptional, or funny (see Lehtonen 2017).
The example is a post in a thread where members of the forum introduce themselves.
The poster is an active member of the Finnish brony fandom who was also interviewed for
a larger research project on the Finnish bronies (see Lehtonen 2017). At a first look, the post
seems like a typical discussion forum post. A lot could be said by merely examining the
post as a representative of a certain genre on social media. It has the format of a forum post
signalling the title of the thread, the nickname and avatar of the author, the time-stamps, and
so forth. It is multimodal and involves text, emoji, images, hyperlinks, and an embedded
video. It follows the textual genre of the thread – titled ‘Introduce yourself’ – which is a
chain of introductions; these introduction threads are typical on discussion forums focusing
on shared interests (see Page 2012: 31–33). Each post presents the poster, his occupation
and hobbies, and his relationship to the MLP television series. However, we will argue that
– in terms of gender, sexuality, and other intersecting identity categories – it is impossible to
understand the meanings constructed in the post without a knowledge of the cultural context
of the brony fandom and ethnographic understanding of the shared practices in the Finnish
fandom.
Figure 36.1 shows a modified, anonymised image of the data excerpt.1
First, the language use in the text indexes identities some of which are also explicitly
stated in the post. The poster, for instance, says that he is in the military service and a
21-year-old student. His young adult male identity – opposed to the expected target audi-
ence of MLP, young girls – is also signalled by the fairly formal language use with some
swear words, as well as his way of initially distancing himself from the MLP series as a ‘lit-
tle girls’ thing’. In the written text, the unexpectedness of a young adult male’s interest in the
series is rationalised through various strategies. Being a brony goes against the expectations
tied to normative Finnish adult masculinity, especially in its institutionalised forms, such
as the compulsory military service, which is commonly described as the rite of passage to
manhood. However, the rationalisation is not only necessary in terms of gender and age to
explain away the ‘girlishness’ or ‘childishness’ of one’s interest, but also in terms of sexual-
ity. Although sexuality is nowhere explicitly mentioned, in other contexts bronies have been
suspected of representing queerness, non-normative sexualities, or, in the worst cases, pae-
dophilia. This makes sense in the framework of hegemonic, heterosexual masculinity where
masculinity itself is defined in opposition to women, children, and queers (e.g. Connell
2005). Bronies need to negotiate outsiders’ expectations and thus explain their unexpected
interest in a girls’ media franchise by refusing to follow the logic of hegemonic masculinity
and by offering alternative motivations for, and ways of doing, brony masculinity.
In our example, the rationalisation happens by relying on two recurring narrative pat-
terns that are circulated widely in the translocal brony community: a trauma narrative and a
conversion narrative (Lehtonen 2017). Typically, in the context of brony fandom, the trauma
narrative has to do with the poster’s past experiences of bullying and depression. Here the
poster refers to his ‘shitty past’ right at the beginning. While this kind of trauma narrative
occurs in various contexts, in the brony community, through the process of resemiotisation,
it is usually also used to explain the writer’s interest in a girls’ media franchise. That is,
in the bronies’ version of the trauma narrative the television series offers consolation and

548
 Doing gender & sexuality intersectionally

Figure 36.1 Modified image of a post from a Finnish discussion forum for ‘bronies’.

values, such as friendship and respect for others, that have helped the writer deal with his
past experiences. In the brony community, the narrative is so typical that the poster does
not have to explain it in detail – the mere mention of his depressive past at the beginning
indexes that he is like others with similar past narratives who have entered the community.
The same is the case for the second narrative pattern, the conversion narrative, borrowed
from religious contexts but typical also in other fan communities. At the end of the post, the
writer explains his ‘conversion’ from a hater into a fan in one paragraph. After watching two
episodes of the ‘little girls’ thing’ because some of his online friends were also doing this, he
was suddenly emotionally hooked, or found himself ‘singing along during the theme song’.

549
Sirpa Leppänen & Sanna Tapionkaski

Here the musical and emotional content of the series is marked as significant, although else-
where the same writer also offers more analytic explanations for why it is that this particular
girls’ series is so good (e.g. it has more complex screenwriting than other children’s series).
While the written text thus indexes an adult male rationalising his interest in a young
girls’ series, the other semiotic resources index his fan identity more directly. This fan iden-
tity partly draws on resources often used in online communication by young girls: emoji
based on an animated series (here ponies), an avatar that is an animated figure (here a pony),
simple lines in English borrowed from a television show (‘Hello everypony’). However,
other resources signal an adult fan identity: the poster’s signature that involves text explain-
ing that he has been participating in offline ‘pussikalja’ meet-ups (‘pussikalja’, literally ‘bag
beer’ is an informal expression that refers to the activity of buying a bag of beers from a
shop and drinking it outside) and includes a banner advertising an offline meet-up. Other
hyperlinks indicate his interest in the translocal brony fandom (a link to an MLP video
remix), as well as his identity as a fantasy RPG gamer (a link to the Elder Scrolls game fran-
chise). The poster’s pony avatar is a combination of (stereotypically) child and adult things:
an animated pony character wearing an army helmet with the acronym N.E.E.T. (not in
education, employment, or training). On the basis of an interview with the poster, we know
that the avatar is his ‘ponysona’, designed by himself that changes in relation to his life
events. We should note that the multimodal resources in the post have, in fact, accumulated
over time: it is not possible, for instance, that the poster could have advertised for the offline
meet on the date of the original post (other threads reveal that it had not been planned yet).
All the profile information also changes constantly; thus, the total number of posts by the
poster, listed under the avatar, is from the date of collecting the data (year 2016), not from
the original date of the post (2012). While this means that there is no access to the post as
it was at the original time of posting, the accumulated resources offer plenty of materials to
examine a fan identity performed through time, rather than merely in the moment of post-
ing. In sum, looking at the identity performances of the poster by taking into account the
ways in which the brony fan identity intersects with gender, sexuality, and age is not only
necessary in terms of understanding the complex, lived experience of this particular poster
but also enables the researcher to move beyond examining the text in terms of polar identity
categories (e.g. female–male, feminine–masculine, child–adult, heterosexual–asexual).
Finally, we suggest that an ethnographic understanding of the bronies’ fan practices is
needed to be able to read all of this as (mainly) sincere and not (merely) ironic. While there
are some indications in the language use, namely its sincere tone, as well as in the discursive
context – a thread of introductions by participants who wish to join the community – that the
post is not meant to be read as ironic, it could be posted by a skilful troll. There are, after all,
several memes, that is, images, videos, or pieces of texts copied and spread rapidly on the
internet that mock the bronies. Obviously, one could analyse the post as a text by a writer
whose sincerity cannot be decided: whether sincere or ironic, the gendered and sexualised
discourses and multi-semiotic choices in the post can be examined through textual analysis.
However, if one wants to analyse the social media practices as people’s lived experiences,
the textual analysis should be supported by ethnographic research. Online ethnography con-
ducted on the discussion forum could already help to contextualise the poster as a member
of the community: one would, for instance, find out that he has been very active on the
forum (more than 2,000 posts over the course of several years), offers support for others, is
on friendly terms with several other members, and also active on other online and offline
sites of the community (offline fan activities are often documented online, so the researcher
does not have to go ‘out’ to the field to be able to access these). All of this supports the

550
 Doing gender & sexuality intersectionally

reading of his identity performances as sincere. In the case of this participant, however,
online ethnography was also supported by an offline interview that shed further light on his
fan trajectory and its connections to his everyday life, such as his statement that his avatar’s
changing outfits, including the N.E.E.T. helmet, reflect what is actually going on in his life.

Doing migrant masculinity intersectionally

Our second set of examples with which we illustrate multimodal social media practices
that are centrally concerned with performing, sharing, and interrogating identities intersec-
tionally, are short, entertaining, often explicitly parodic, videos by young Finland-based
migrant men. These are disseminated via video sharing platforms, such as YouTube and
Vimeo, where they are also discussed and debated by diverse audiences. These videos are
part of a popular trend whereby social media serve as a popular ‘stage’ for lay and celebrity
performances as well as ‘a virtual lounge’ for audience reactions to and evaluations of these
performances (Marwick and boyd 2011). On video-sharing platforms, such as YouTube,
performances are disseminated for appreciative audiences who, in turn, take up what is
being performed in discussions and debates, but also in replications and recontextualisa-
tions of different kinds and degrees of appreciation and criticality (Häkkinen and Leppänen
2014). Platforms like YouTube are thus not simply media in which people watch videos,
listen to music, broadcast their own videos, but they also serve as a forum for participatory
cultures, cultural niches, and memes.
Multi-sited ethnographic observations have shown that for migrants, video-sharing social
media have become means of reaching audiences beyond their locality and immediate social
circles. For many of them, their mediatised performances have proved an important means
for establishing their identity and for generating more followers and fans (Leppänen and
Westinen 2018; Westinen 2017). Social media performances are also something that the
performers can themselves monitor, and thus they provide them with an opportunity to have
a voice that can have considerable reach and potential in terms of audience engagement.
Typically, the videos feature everyday happenings, such as driving a car, talking with
one’s mother, or having an argument with a girlfriend. They are always to some extent
scripted mono- or dialogic performances. Technically, they are shot with a smartphone or
a video camera, after which they are usually (moderately) edited, and sometimes comple-
mented with special effects, such as sounds, music, animation, and subtitles. The language
used in them is most often Finnish, although they can also include English, as well as snip-
pets of other languages. The varied language choices of the videos show, in fact, that they
are not aimed at particular migrant groups, but at more mixed audiences, including ethnic
Finns as well as other migrants who are Finnish- and English-speaking.
As multimodal productions, the videos call for detailed and multifaceted analysis that
can describe the ways in which the modes drawn on in them – varied language resources,
embodied performances, cinematic representation, music, for example – are coordinated
and integrated to create coherent narrative performances. As highlighted in interviews with
creators of these videos, for both themselves and their audiences the crafted multimodal nar-
ratives serve as entertainment and critical commentary: through humour and disparagement,
they depict ways in which migrants are treated by the host society, as well as how migrants
themselves strive to make sense of the host society
A very typical strategy that the video performances apply to achieve these effects is
double voicing (Bakhtin 1984). In concrete terms, this means that the narratives are pre-
sented from the perspective of young migrant men, pretending not to be young migrant

551
Sirpa Leppänen & Sanna Tapionkaski

men. For example, in one video,2 entitled ‘A migrant who thinks he is a Finn’, a young
Iraqi migrant, with the artist name Kilikali, gives a parodic performance of ethnic Finnish
masculinity. The video depicts Kilikali driving his car and giving an impassioned speech
in Finnish on how Finland – which has recently received unprecedented numbers of
migrants and refugees – is being invaded by foreigners who ‘steal our women’ and ‘fill
our country’ so that there is ‘no space left for us anymore’. In this way, his speech takes
up recognisable themes of the racist and populist discourses circulating in Finnish society
at the moment, and recontextualises them as part of his parodic-critical performance of
ethnic Finnish masculinity. In addition, his exaggerated (‘stylised’, Coupland 2007) way
of speech functions as an index of Finnishness: it flows effortlessly and idiomatically in
a fluent eastern-Finnish dialect – a dialect typically associated with a traditional lifestyle.
At the same time, the embodied style Kilikali has chosen for himself in the video (Figure
36.2) – the sunglasses hiding his eyes and the military-style jacket, for example – empha-
sise the fact that he is not an authentic Finn.
The video thus gives us a parodic performance of a stereotypical social persona, that of
a traditional heterosexual Finnish man. It borrows his voice, thus crossing over to a habitus
and style that are generally not taken to belong to people such as the migrant author of the
video. Thus, the video performance portrays to its audiences a double-voiced, intersectional
identity: migrant and non-migrant; non-Finnish and Finnish; Finnish-speaking and non-
Finnish speaking; non-racist and racist; critical and humorous.
Other examples that rely on a similar double-voiced strategy are videos that give us
(drag) performances of Finnish women, such as the one by an Angolan migrant man, who
has adopted Bianca Sossu as his artist name.
This video, ‘When you are trying to teach the language of Finland to your refugee boy-
friend’, features the character of Bianca Sossu engaged in dialogue with her refugee boy-
friend. In it, Bianca is presented as the Finnish L1 speaker instructing the boyfriend on
how to say in Finnish ‘I love you’ (‘Minä rakastan sinua’). Like the previous example,
this performance, too, mobilises a range of semiotic resources. These include, for example,
Bianca’s didactic use of vernacular Finnish – the character’s expertise in Finnish is used to

Figure 36.2 Kilikali in the video.

552
 Doing gender & sexuality intersectionally

Figure 36.3 Bianca’s performance in the video as a blond Finnish woman.

suggest that the performance is essentially about Finnishness. In addition, Bianca’s embod-
ied performance in the video as a blond Finnish woman (Figure 36.3) can be interpreted as
indexing a particular kind of young white Finnish woman, a bold and frivolous ‘pissis’ girl
(Halonen and Leppänen 2017), located at a particular intersection of gender, age, race, sexu-
ality, ethnicity, and nationality. Finally, in the same way as the previous example, the video
also includes snippets of recognisable populist and racist slogans (e.g. ‘in Finland we don’t
beat up women’; ‘you are going back to your country’). These are recontextualised as part
of this performance of gendered Finnishness, in order to drive home a humorous and criti-
cal commentary of the racist and xenophobic discourses currently mushrooming in public
debates and discourses in Finnish society.
In sum, both videos tell multimodal narratives of selves and their ethnic Finnish others
at the intersection of categories – race, gender, sexuality, nationality, and ethnicity, most
significantly. While they are humorous in nature, they are also deeply critical in how they
transgressively make visible and engage with particular norms, distinctions, and relations
in power related to the lived realities of migrant men: what and who they may and can be
in the often-hostile host society which projects onto them various forms of otherness, dan-
gerousness, and threat.

Future directions
As both intersectionality research in sociolinguistics and discourse studies and research
on multimodal gendered and sexualised identity practices on social media are fairly recent
fields of study, more research is called for in the future. Our own research has focused on
voluntary, participatory social media practices that often involve conscious norm-break-
ing both in terms of identity discourses and the use of linguistic and semiotic resources
(e.g. Halonen and Leppänen 2017; Lehtonen 2015, 2017; Leppänen 2008; Leppänen and
Häkkinen 2012). Future research could look at social media contexts where people are more
inclined to follow norms, rather than challenge them. A particularly interesting example
could be social media activities involving explicit relationship work, such as guidance and

553
Sirpa Leppänen & Sanna Tapionkaski

advice sites and online dating services (see e.g. Herring and Kapidzic 2015; Milani 2013;
Kohler Mortensen 2017). Moreover, attention could be paid to how intersectional iden-
tity performances are both carried out and reflexively made sense of in interactions among
social media participants. In particular, it is a timely topic to investigate the various forms
of disparagement and policing that are targeted at particular individuals and groups and that
can have serious consequences for those focused on. These can sometimes take elaborate
forms of multimodal crafting, recontextualisation, and resemiotisation, and become widely
influential and virally spreading memes. Examples of these include social media campaigns
focusing on public persona such as politicians, civil servants, and journalists in which their
credibility is attacked and ridiculed; as well as the abundant racist and populist memic
discourses about the other – refugees and migrants, for example – widely circulating and
multiplying in social media.

Notes
1 For privacy and copyright reasons the original screenshots of the brony sites cannot be used. All the
examples in the text are English translations of the Finnish originals.
2 The authors of the videos have given SL their permission to include links to their videos, to show
screenshots of them, and to refer to their verbatim discourse, in public presentations and publications
discussing their work.

Further reading
Leppänen, S., Westinen, E., and Kytölä, S. (eds.) (2017) Social Media Discourse, (Dis)identifications
and Diversities. New York, NY: Routledge.
The introductory chapter gives an overview of recent work in the sociolinguistic work on identity
work on social media as well as discusses key concepts and approaches in this. Many of the chapters
exemplify an intersectional take on identity as well as different approaches to multimodal analysis of
social media discourse.
Levon, E. (2015) ‘Integrating intersectionality in language, gender and sexuality research’. Language
and Linguistics Compass, 9(7), pp. 295–308.
Levon’s article gives a useful introduction to how intersectionality can be integrated in
sociolinguistic research on language, gender, and sexuality.
Machin, D., Caldas-Coulthard, C., and Milani, T. (2016) ‘Doing critical multimodality in research on
gender, language and discourse’. Gender and Language, 10(3), pp. 301–308.
The article discusses multimodal work in the area of gender, language, and discourse, and proposes
the kinds of multimodal approaches that are most appropriate for this task.

Related topics
Digital ethnography in the study of language, gender, and sexuality; using communities of practice
and ethnography to answer sociolinguistic questions; applying queer theory to language, gender, and
sexuality research in schools; poststructuralist research on language, gender, and sexuality; analysing
gendered discourses online.

References
‘A migrant who thinks he is a Finn’ (‘Mamu joka luulee olevansa suomalainen’). A Video. Available at:
https://www.youtube.com/watch?v=J0TFLDw1Ifk (Accessed: 12th November 2017).

554
 Doing gender & sexuality intersectionally

Androutsopoulos, J. (2008) ‘Potentials and limitations of discourse-centred online ethnography’.

Language@Internet, 5, article 8.
Bakhtin, M. (1984 [1963]) Problems of Dostoevsky’s Poetics, Emerson, C. (ed. and trans.). Minneapoli,
MN: University of Minnesota.
Baxter, J. (2003) Positioning Gender in Discourse: A Feminist Methodology. Basingstoke: Palgrave
Macmillan.
Bauman, R. and Briggs, C. L. (1990) ‘Poetics and performance as critical perspectives on language
and social life’. Annual Review of Anthropology, 19, pp. 59–88.
Bucholtz, M. and Hall, K. (2004) ‘Theorizing identity in language and sexuality research’. Language
in Society, 33(4), pp. 469–515.
Butler, J. (1990) Gender Trouble. New York, NY: Routledge.
Butler, J. (1993) Bodies That Matter: On the Discursive Limits of “Sex”. New York, NY: Routledge.
Butler, J. (2004) Undoing Gender. New York, NY: Routledge.
Connell, R. W. (2005) Masculinities. Cambridge: Polity Press.
Coupland, N. (2007) Style. Language Variation and Identity. Cambridge: Cambridge University Press.
Crenshaw, K. (1989) ‘Demarginalizing the intersection of race and sex: A black feminist critique of
antidiscrimination doctrine, feminist theory, and antiracist politics’. University of Chicago Legal
Forum, 14, pp. 538–554.
Crenshaw, K. (1991) ‘Mapping the margins: Intersectionality, identity politics, and violence against
women of color’. Stanford Law Review, 43(6), pp. 1241–1299.
Eckert, P. and McConnell-Ginet, S. (1999) ‘New generalizations and explanations in language and
gender research’. Language in Society, 28(2), pp. 185–201.
Foucault, M. (1978) The History of Sexuality, Vol. 1: An Introduction, Hurley, R. (ed.). London:
Penguin.
Gibson, J. J. (1986) The Ecological Approach to Visual Perception. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Goffman, E. (1959). The Presentation of Self in Everyday Life. New York, NY: Anchor Books.
Halonen, M. and Leppänen, S. (2017) ‘“Pissis stories”: The Self and the other as gendered, sexualized
and class-based performance on social media’. In: Leppänen, S., Westinen, E., and Kytölä, S.
(eds.) Social Media Discourse, (Dis)identifications and Diversities (Routledge Studies in
Sociolinguistics). New York, NY: Routledge, pp. 39–61.
Herring, S. C. (2013). ‘Discourse in Web 2.0: Familiar, reconfigured, and emergent’. In: Tannen, D.
and Tester, A. M. (eds.) Georgetown University Round Table on Languages and Linguistics 2011:
Discourse 2.0: Language and New Media. Washington, DC: Georgetown University Press, pp.
1–25.
Herring, S. C. and Kapidzic, S. (2015) ‘Teens, gender, and self-presentation in social media’.
In: Wright, J. D. (ed.) International Encyclopedia of Social and Behavioral Sciences, 2nd ed.
Amsterdam: Elsevier, pp. 146–152.
Hill Collins, P. and Bilge, S. (2016) Intersectionality. Cambridge: Polity Press.
Häkkinen, A. and Leppänen, S. (2014) ‘YouTube meme warriors. Mashup videos as satire and
interventional political critique’. eVarieng, p. 15.
Iedema, R. A. (2003) ‘Multimodality, resemiotization: Extending the analysis of discourse as multi-
semiotic practice’. Visual Communication, 28(1), pp. 29–57.
Jones, B. (2015) ‘My Little Pony, tolerance is magic: Gender policing and Brony anti-fandom’.
Journal of Popular Television, 3(1), pp. 119–125.
Kaplan, A. M. and Haenlein, M. (2010) ‘Users of the world, unite! The challenges and opportunities
of Social Media’. Business Horizons, 53(1), pp. 59–68.
Køhler Mortensen, K. (2017) ‘Flirting in online dating: Giving empirical grounds to flirtatious
implicitness’. Discourse Studies, 19(5), pp. 581–597.
Kress, G. and van Leeuwen, T. (2006 [1996]) Reading Images: The Grammar of Visual Design.
London: Routledge.

555
Sirpa Leppänen & Sanna Tapionkaski

Lehtonen, S. (2015) ‘Writing oneself into someone else’s story – Experiments with identity and
speculative life writing in Twilight fan fiction’. Fafnir – Nordic Journal of Science Fiction and
Fantasy Research, 2(2), pp. 7–18.
Lehtonen, S. (2017) ‘“I just don’t know what went wrong”: Neo-sincerity and doing gender and age
otherwise in a discussion forum for Finnish fans of my little pony’. In: Leppänen, S., Westinen, E.,
and Kytölä, S. (eds.) Social Media Discourse, (Dis)identifications and Diversities. New York, NY:
Routledge, pp. 287–309.
Leppänen, S. (2008) ‘Cybergirls in trouble? Fan fiction as a discursive space for interrogating gender
and sexuality’. In: Caldas-Coulthard, C. R. and Iedema, R. (eds.) Identity Trouble: Critical
Discourse and Contested Identities. Basingstoke: Palgrave Macmillan, pp. 156–179.
Leppänen, S. (2012) ‘Linguistic and discursive heteroglossia on the translocal internet: The case of
web writing’. In: Sebba, M., Mahootian, S., and Jonsson, C. (eds.) Language Mixing and Code-
Switching in Writing: Approaches to Mixed-language Written Discourse. London: Routledge, pp.
233–254.
Leppänen, S. and Häkkinen, A. (2012) ‘Buffalaxed superdiversity: Representations of the Other on
YouTube’. Diversities, 14(2), pp. 17–33.
Leppänen, S. and Kytölä, S. (2016) ‘Investigating multilingualism and multi-semioticity as
communicative resources in social media’. In: Martin-Jones, M. and Martin, D. (eds.) Researching
Multilingualism: Critical and Ethnographic Approaches. London: Routledge, pp. 155–171.
Leppänen, S. and Westinen, E. (2018) ‘Migrant rap in the periphery: Performing politics of belonging’.
AILA Review 2016: Meaning-Making in the Periphery.
Marwick, A. and boyd, d. (2011) ‘To see and be seen: Celebrity practice on Twitter’. Convergence:
The International Journal of Research into New Media Technologies, 17(2), pp. 139–158.
McCall, L. (2005) ‘The complexity of intersectionality’. Signs, 30, pp. 1771–1800.
Milani, T. (2013) ‘Are “queers” really “queer”? Language, identity and same-sex desire in a South
African online community’. Discourse & Society, 24(5), pp. 615–633.
Milroy, J. and Milroy, L. (1993) ‘Mechanisms of change in urban dialects: the role of class, social
network and gender’. International Journal of Applied Linguistics, 3(1), pp. 57–77.
Motschenbacher, H. (2011) ‘Taking Queer Linguistics further: sociolinguistics and critical
heteronormativity research’. International Journal of the Sociology of Language, 212, pp. 149–179.
Page, R. E. (2012) Stories and Social Media: Identities and Interaction. New York, NY: Routledge.
Queen, R. (2014) ‘Language and sexual identities’. In: Ehrlich, S., Meyerhoff, M., and Holmes, J.
(eds.) The Handbook of Language, Gender and Sexuality, 2nd ed. Oxford: Wiley-Blackwell, pp.
203–219.
Silverstein, M. (2006) ‘Pragmatic indexing’. In: Brown, K. (ed.) Encyclopedia of Language and
Linguistics, 2nd ed. Amsterdam: Elsevier, pp. 14–17.
Staehr, A. (2015) ‘Reflexivity in Facebook interaction–enregisterment across written and spoken
language practices’. Discourse, Context & Media, 8, pp. 30–45.
Westinen, E. (2017) ‘“Still alive, nigga”: Multisemiotic Constructions of self as the other in Finnish
rap music videos’. In: Leppänen, S., Westinen, E., and Kytölä, S. (eds.) Social Media Discourse,
(Dis)identifications and Diversities. New York, NY: Routledge, pp. 335–361.
‘When you are trying to teach the language of Finland to your refugee boyfriend’ (‘Kun yrität opettaa
sun pakolais poikaystävälle Suomen Language’). A Video. Available at: https://www.youtube.com
/watch?v=ac7hMbaonQ0&app=de (Accessed: 15th August 2017).

556
 Part VIII
Corpus linguistic approaches
 37
Lovely nurses, rude receptionists,
and patronising doctors
Determining the impact of gender
stereotyping on patient feedback

Paul Baker and Gavin Brookes (Part VIII leads)

Introduction
Part VIII of the Handbook demonstrates how techniques from corpus linguistics can be
useful with large amounts of linguistic data for studying the relationship between language,
gender, and sexuality. Corpus linguistics is a collection of methods that involve the use
of specialist computer programmes to analyse language use in large bodies of machine-
readable text (McEnery and Wilson 2001). Corpus approaches to linguistic analysis typi-
cally involve the use of quantitative computational techniques in order to identify frequent,
salient, or atypical patterns in texts. Corpus analysis tools also display the language data
visually, in ways that make it easier for human analysts to notice such patterns through
qualitative analyses. This first chapter in Part VIII provides a detailed introduction to corpus
linguistics, outlining its key principles, strengths, and limitations, and providing worked
examples of how established corpus techniques can be used to study gendered discourse. We
then review a selection of corpus-based studies of language, gender, and sexuality, before
concluding with a brief introduction to the other chapters in this section.

Corpus linguistics: defnitions and main principles

A corpus (Latin for body, plural corpora) is a large collection of naturally occurring texts
which have been sampled to be representative of a particular language or linguistic variety.
There are two main types of corpora: reference (or general) and specialised (or purpose-
built). Reference corpora, which are designed to represent a whole language (usually at a
particular point in time), often contain thousands of texts, amounting to millions or billions
of words. Specialised corpora, on the other hand, are designed to represent language in more
specific contexts (e.g. newspaper articles about a particular topic, published within a par-
ticular time frame) and are usually smaller than reference corpora. The concepts of sampling
and representativeness are vital to ensuring that corpora are representative of the language
or variety we are interested in studying. For example, if we aim to sample female speech in

559
Paul Baker & Gavin Brookes

England, we would need to ensure that the speech we sampled for the corpus represented
the speech of women from different age groups, geographical locations (e.g. North, South,
Midlands), social classes, ethnicities, and so forth. We would also have to consider the
extent to which our corpus represented speech in different situational contexts (i.e. where,
when, and with whom the conversations took place). Once these variables have been con-
sidered, we also need to ensure that the corpus is relatively balanced so that no demographic
group or other variable is over- or under-represented in the data.
Once selected or assembled, the corpus is analysed with the help of specialist computer
programmes (e.g. WordSmith Tools (Scott 2016), AntConc (Anthony 2014), CQPWeb
(Hardie 2012), Sketch Engine (Kilgariff et al. 2014)). To subject the corpus to computa-
tional analysis, the text files it contains must be stored in a machine-readable, digital format.
For texts that already exist in electronic form, such as emails, e-books, and online news
articles, this is a relatively straightforward task. However, texts originating in other formats
pose more of a challenge in this regard. Printed texts, such as printed books or handwritten
letters, need to be typed up or digitised using optical character recognition software. Spoken
texts, such as face-to-face conversations, need to be recorded and digitally transcribed prior
to analysis. When dealing with spoken data, the corpus builder must also decide whether
(and if so, how) to include paralinguistic features of the interaction, such as tone and pitch
of voice, body language, gesture, and facial expression. Once they have been assembled,
corpora can be annotated with additional information relating to the texts or the language
they contain. Annotation at the text level is likely to indicate the genre or mode of the texts,
or demographic metadata about the authors/speakers (e.g. speaker sex, age, social class,
etc.), while annotation at the linguistic level usually indicates a word’s grammatical class or
semantic field. Linguistic annotation is normally carried out automatically via taggers such
as Wmatrix (Rayson 2008).Importantly, issues around ethics and copyright also need to be
considered when building a corpus, particularly if the language collected involves private
forms of communication or we intend to make the corpus publically available.
The option of introducing corpus linguistics methods into the analysis of gender and
sexuality is appealing for many reasons. First, because corpora are characteristically larger
and more representative than datasets comprising just single or handfuls of texts, analysts
can usually claim that their findings are more generalizable with respect to the type of lan-
guage they are investigating. A related strength of corpus methods is that they are useful for
pinpointing what Baker (2006: 13) describes as the ‘incremental effect of discourse’, that is
to say, the propensity for discourses to be subtly established through linguistic patterns that
might feature sparingly in one or two texts, but become significant when considered as part
of a broader discourse type or collection of texts. Larger corpora are also more likely than
smaller datasets to represent a fuller range of discourse positions around a particular subject,
which can be particularly useful for apprehending so-called resistant or minority discourses.
A further advantage of corpus methods is that the specialist computer programmes that
underpin such methods can perform frequency counts and complex statistical calculations
faster and more reliably than the human mind, and so can reveal linguistic patterns that
might otherwise evade manual observation or run counter to intuition (Baker 2006: 10–14).
Corpus linguistics methods can also add a degree of objectivity to discourse analysis in that
they advocate a principle of methodological transparency in terms of two important princi-
ples: (i) no systematic bias in the selection of texts included in the corpus (i.e. do not exclude
a text because it does not fit a pre-existing argument or theory); and (ii) total account-
ability (all data gathered must be accounted for) (McEnery and Hardie 2012). Combined,
these principles of methodological transparency can help discourse analysts overcome

560
 Gender stereotyping and patient feedback

the accusation that certain texts or examples have been cherry-picked (Widdowson 2004)
because they support a preconceived argument.
Like any other methodology, however, corpus linguistics is not perfect. One limitation
concerns what corpora have the power to represent. The rendering of a text or a collection
of texts into a corpus is a transformative process, the product of which bears important dif-
ferences to the originals. Because corpora tend to contain transcriptions of written or spoken
language, it is difficult to systematically use corpus techniques to analyse, say, the gestures
in a spoken interaction, or the visuals in a newspaper text (Baker 2008). A related criticism
levelled at corpora is that they offer decontextualised versions of the texts they represent.
The corpus analyst must therefore carry out more work in order to become cognisant of the
role that the contexts of situation and culture play in the production and consumption of the
texts contained in the corpus. Another limitation of corpus approaches is that they tend to
focus on presence (i.e. those linguistic features that do occur in the texts under analysis), but
tell us less about what is absent (i.e. what is missing or obscured from the texts in the cor-
pus). Finally, although the computer programmes underpinning corpus linguistic analyses
are useful for pinpointing frequent and statistically interesting features of the data, they can-
not interpret those patterns or tell us why they are significant. Crucially, it is up to the human
analyst to dig deeper (and often wider – drawing on corpus-external sources) to make sense
of and explain the significance of those patterns. At this point, we reiterate Baker’s (2014)
argument that analysts should be wary of the danger in relying exclusively on corpora and
corpus techniques, particularly when carrying out social research. A corpus by itself does
not always yield explanations for language patterns. Only by considering other forms of
context – by reading individual texts in their entirety or drawing on our knowledge of the
world outside the corpus, in which those texts exist – can we fully account for our findings.
It should be borne in mind that many of the criticisms of corpus methods outlined above
can also be directed at other approaches applied to the study of language, gender, and sexu-
ality. These criticisms should not discourage the use of corpus methods in this field. Rather,
our discussion of them here is intended to make analysts aware of what corpus methods
can and cannot do well so that informed decisions can be made about when and how such
methods might be used most effectively, and work out what steps can be taken to meet any
methodological shortcomings. Analysts employing corpus methods should also take care if
making claims about objectivity. Our own political leanings may bias us by causing us to
disregard aspects of texts we disagree with, and even when dealing with non-political data,
human brains tend to process data inefficiently (such as paying more attention to emotion-
ally charged or recent information). Corpus analytical computer programmes are not subject
to such biases in the same way as human minds. However, both the designers and users of
these programmes are still subject to such biases. It is therefore important for analysts to
avoid uncritical over-reliance on corpus techniques, and be self-reflexive about the influ-
ence that their own positions and biases are likely to have had in shaping their findings (see
also: Baxter 2003; Mills and Mullany 2011).

Using corpus methods to analyse gender: a case study of patient

feedback
In order to demonstrate how some of the most commonly used techniques within corpus
linguistics can be applied to research on language, gender, and sexuality, this short section
examines a corpus of 29 million words of online patient feedback to the British National

561
Paul Baker & Gavin Brookes

Health Service, posted to the NHS Choices website between 2013 and 2015. The corpus
contains 228,113 separate comments relating to GP practices, hospitals, dentists, and a
range of other NHS providers. The research question which frames this section is: to what
extent is evaluation made about NHS practitioners gendered? To answer this question we
will use techniques from word lists, collocation, and concordancing.
Prior to the analysis, the corpus was loaded to an online interface called CQPweb (Hardie
2012). As a first step we examined a word list which simply consists of a table containing all
of the words in the corpus, displayed in order of frequency. The top ten words in the corpus
were the, to, I, and, a, have, of, for, is, and in. These words often appear as the most frequent
words in any corpus and are not the most helpful ones for us to answer our research ques-
tion. Instead, we are more interested in frequent words which indicate evaluation. The ver-
sion of the corpus we are working with has been part-of-speech tagged, meaning that each
word in it has been annotated with one of 148 fine-grained grammatical categories or codes
such as NN1 (singular common noun) or RT (adverb of time). This is a process that was car-
ried out automatically with a computer program called CLAWS (Garside and Smith 1997)
and can be useful in helping analysts to focus on words with certain properties. In our case
we want to consider words tagged as JJ (general adjective). We can specify that CQPweb
shows a wordlist just containing words tagged as JJ, and from that list we can pick out the
most frequent terms which indicate positive and negative evaluation. The most frequent
positive evaluative words (in order of frequency) are: good, excellent, helpful, friendly,
great, professional, caring, fantastic, lovely, and polite, while the ten most frequent nega-
tive words are: rude, bad, unhelpful, terrible, awful, appalling, unprofessional, disgusting,
abrupt, and useless.
Bearing these 20 adjectives in mind, we can examine whether certain words tend to be
gendered, or in other words, tend to be used to differently to describe men or women. There
are various ways that this can be approached, but as an illustrative example we decided to
examine the extent to which these words collocate with the gendered pronouns he and she.
A collocate is simply a word which occurs near or next to another word, usually more fre-
quently than would be expected by chance alone. In order to identify collocates we first need
to decide on which collocational measure to use (for this case study we will simply count the
cases of co-occurrence rather than using a more complex measure like mutual information
or log-likelihood). We also need to consider the collocational span; i.e. the maximum num-
ber of words apart that two words can appear together. Using different spans will produce
different results. A wide span (say, ten words to the left or right of a word), will probably
produce a lot of collocates, including some words where the relationship between the words
is not a direct one. On the other hand, too small a span (say, one word either side of a word)
will produce fewer collocates but is also likely to miss some important cases. Generally, a
span of between three and five words either side of the word we are focusing on is used. For
this study we will use a span of five words and remove unwanted cases by hand.
After carrying out some experiments with different gendered terms, we chose to examine
the evaluative adjectives which collocate with the pronouns he and she. These two terms
are highly frequent so are likely to give us a reasonable number of cases to consider. They
also tend to co-occur with the evaluative adjectives in constructions like ‘I have to praise the
new doctor he is very good’. However, we need to be careful about assuming that every case
of ‘good’ and ‘he’ collocating involves a positive evaluation of someone who works at the
NHS and is male. For example, consider the following excerpt of text: ‘he received equally
good care’. Here, the male person being discussed is a patient rather than a practitioner. For
this reason, it is necessary for us to examine concordances of each collocational pair and

562
 Gender stereotyping and patient feedback

adjust the frequencies in order to remove cases which are not relevant to the research ques-
tion. A concordance is a table showing all of the citations of a word, phrase, or collocational
pairing in a corpus, within the immediate context of the text they occur in. As concordances
often only show around 10 to 20 words at a time, it is sometimes necessary to expand con-
cordance lines to see more context in order to make an accurate interpretation. Table 37.1
shows a concordance of the first ten cases of the collocational pair of ‘he’ and ‘good’ in the
corpus.
Examining Table 37.1, we only categorised lines 1, 3, 5, and 6 as counting towards an
evaluation of a male practitioner as good. Line 8 refers to a doctor giving good advice,
whereas line 10 refers to the patient’s relationship with the doctor as being good. A case
could be made for counting these two cases as well, but for this analysis we are only going
to consider direct evaluations. Another analyst may make a different decision though, and
because of this we should note that it is more important for analysts to explain how a cat-
egorisation is made and be consistent, rather than trying to produce a ‘definitive’ categorisa-
tion scheme which satisfies everyone. Therefore, for this study, we counted cases of direct
evaluation e.g. ‘he is good’ as well as cases where a practitioner’s behaviour or manner was
evaluated e.g. ‘what she did is disgusting’.
Table 37.2 shows the top ten positive and negative evaluative adjectives and the number
of times they collocate with he or she in order to label an NHS practitioner.
What can we conclude from the numbers in Table 37.2? First, it looks as if positive
evaluation is more frequent than negative evaluation, and that women are evaluated more
often than men (both positively and negatively). This might suggest that women receive
more scrutiny than men in this context, although the higher numbers for women could be
due to a range of other reasons. For example, perhaps patients are more likely to encounter

Table 37.1 Concordances of the collocation of ‘he’ and ‘good’

1 recepton staff I have to praise the new good and thorough . However trying to get
doctor he is very an appointment is a nightmare
2 nowt rong with him : @ i rushed.him 2 good job i.did he had a really.really bad chest
walkin center n infection n if left
3 who my husband had had in the past as good , when I get te comment ‘for goodness
he was really sake’ i cant believe
4 a lot of love put him to bed . He had a good night sleep and if I can write you now
, it is
5 wont waste my time ! GP Practice Review good . he is always very helpful for patient .
My doctor is very He is the
6 allow for better access . good receiption good dr coz he freindly behavier and
my GP is a very reception is good also The GP
7 Sadly though this was to the adhd service good as he does n’t suffer from this condition
which was no . When I asked
8 feel I can trust this doctor , he has always good advice . He never rushes , which is a
given me godsend , he
9 now when I have a small baby . Not only good during appointments but he also follows
is he things up to check how we
10 and the staff are very courteous.The good and he consulted me about my
doctor relationship with me was very treatment.I recommend this practice
to to

563
Paul Baker & Gavin Brookes

Table 37.2 Frequency of positive and negative evaluative adjectives as collocates of ‘he’ and ‘she’ when
referring to NHS practitioners

Positive adjectives he she Negative adjectives he she

good 40 21 rude 91 184

helpful 53 86 bad 3 2
excellent 47 35 unhelpful 6 15
friendly 53 88 terrible 1 7
great 34 47 awful 4 10
professional 83 91 appalling 3 5
polite 30 37 unprofessional 5 7
fantastic 36 65 useless 0 0
caring 45 65 disgusting 1 6
lovely 35 136 abrupt 17 25
Total 456 671 Total 131 261

women than men in the NHS. Overall, there are more references to she (44,938 cases) than
he (37,513 cases) in the corpus. This does not necessarily mean that patients do encounter
women more, but it does imply that they are written about more in the comments. It is also
worth considering the proportion of positive and negative evaluations that men and women
receive. Women receive twice as many negative evaluations as men (261 vs. 131), although
only 1.47 times as many positive evaluations as men (671 vs. 456). However, if we compare
the ratio of positive to negative evaluations, for women it is 1 negative adjective to 2.57
positive adjectives, while for men it is 1 negative adjective to 3.48 positive adjectives. So
while women receive the most positive adjectives overall, compared to men, and compared
to the ratio of negative adjectives with which they also collocate, they do not seem to be
rated as well.
These results should not allow us to conclude that men are better at their jobs than women
in the NHS. Patient feedback is impressionistic, and when people orient to gender in their
comments they are likely to be (unconsciously) drawing upon a wide range of gendered
discourses (Sunderland 2004), resulting in expectations about what counts as ‘appropriate’
or ‘notable’ male and female behaviour. This trend is further illuminated when we consider
words which are used at least double the amount for one sex compared to the other. For
women, these are: lovely, rude, unhelpful, terrible, awful, and disgusting. We may want
to discount the latter four as their frequencies are all quite low, leaving us with lovely and
rude. For men, the only adjective that applies is good. However, we may also want to note
that only two adjectives are used to refer more to males than females (good and excellent),
so these two words are both salient. Therefore, a complementary finding from this current
study in terms of male and female professional behaviour, women are less likely than men
to be evaluated as excellent or good, although they are almost four times as likely as men
to be evaluated with the somewhat more patronising word, lovely (see also Baker 2010).
An important part of corpus-based research is to consider context in order to interpret
and explain our findings. For example, a (partial) explanation for the pattern that we have
found is that certain roles in the NHS involve different expectations about technical, emo-
tional, social, or communicative skills, and may thus attract certain types of evaluation. So
in the corpus the word nurse is more likely (than other NHS roles) to be labelled with the

564
 Gender stereotyping and patient feedback

adjectives lovely or caring, while the word receptionist is more likely to be described as
rude. In addition to this, some roles have a high gender imbalance. For example, a report
from the Nursing and Midwifery Council (2014) found that 90 per cent of nurses were
female, while Yar et al. (2006: 56) gave the proportion of female receptionists in the NHS
as being almost 99 per cent. There is a societal expectation that nurses and women will
be lovely and caring, whereas a receptionist role is paid less and is generally seen as less
prestigious than other NHS jobs (such as doctor or dentist), so perhaps people are less likely
to tolerate behaviour they view as rude from receptionists, as opposed to similar behaviour
from, say, doctors, who have a higher social status, and the ability to heal them.
We could also consider context within the corpus itself, for example, by looking in more
detail at some of the patterns around the collocates. Take the word excellent, which was
found to occur more with he than she. This word is particularly interesting because of the
most frequent ten positive evaluative adjectives it could be seen as the most positive. It is
therefore notable that it is used to describe males more than females (as 18 out of 20 of the
top evaluative words refer to females more often). A more detailed analysis of the concord-
ance lines around excellent found that it was most commonly used with he to refer to doc-
tors (20 cases), dentists (8), and consultants (7). However, when occurring alongside she,
excellent referred most often to female nurses (12 times), consultants (5 times), and doctors
(4 times).
In order to compare more directly how excellent is used in context for male and female
practitioners, we examined only cases of excellent doctors. While there was less data to
examine for women than men, we noticed that the praise around the male doctors was
generally longer and tended to be accompanied by specific reasons to warrant the label
of excellent. For the four excellent female doctors, three of them were simply labelled as
excellent but without any reason given. The fourth was described as ‘very friendly, caring
and professional’ – positive descriptions, but also somewhat generic and unspecific. On the
other hand, excellent male doctors were described in the following ways: ‘spoke clearly,
patiently and without talking down to me … included my wife in our conversation’, ‘com-
passionate without being patronising and was perceptive enough to see that I needed more
than the antibiotics’, ‘he explained everything extremely thoroughly and made me feel very
relaxed’, ‘he found a vein successfully and managed to get bloods and a line in’, and ‘has a
great deal of knowledge’. This sample of longer quotes about male doctors relates to their
attitude, communication skills, technical ability, and knowledge. Not only, then, do males
get referred to as excellent more than females in the NHS, but when they do, they also
appear to get longer amounts of (more specific) praise heaped upon them.
The analysis presented here is preliminary and incomplete in terms of exploring gen-
dered patterns across the whole corpus. A more comprehensive analysis would consider
patterns around the other evaluative adjectives in Table 37.2. For example, the word rude
is the most frequently used negative adjective in the corpus and is also used twice as much
for women as for men. It would therefore be interesting to see what types of behaviours
are described as rude for men and women and explore the extent to which these reflect
different expectations for men and women. We could also further consider the intersection
between behaviour and role. For example, one description of a ‘rude’ receptionist involved
someone’s daughter being ‘asked personal questions about the nature of her ailment by the
receptionist’. This most likely relates to the fact that receptionists often play a ‘medical
gatekeeping’ role which requires them to ask such questions to ensure that patients with the
most serious medical problems are treated first, as well as that patients receive the right type

565
Paul Baker & Gavin Brookes

of treatment from the most appropriate practitioner (Gallagher et al. 2001). Another exam-
ple from the corpus refers to a rude receptionist, with the only additional information about
her being that she told a patient that they could not have an appointment for seven days.
However, receptionists do not control the number of appointments that can be given out,
they simply have to administer timetabling systems that were created by other people. These
two cases of perceived rudeness then, could be more to do with the requirements of being
a receptionist (Hammond et al. 2013), and might be exacerbated by gendered expectations
involving appropriate female behaviour (e.g. women, especially those in relatively low-paid
jobs, should not engage in face-threatening behaviour) (see Mills 2003).
Having demonstrated some of the main techniques of analysis in corpus linguistics, the
following section surveys existing research which has applied such techniques to the study
of language, gender, and sexuality.

Corpus studies of language, gender, and sexuality

In the past, researchers investigating language, gender, and sexuality have tended to take
mainly qualitative perspectives on relatively small datasets, a trend which prompted Baker
(2008: 73–74) to describe corpus linguistics as ‘[sitting] somewhat uncomfortably on the
sidelines of language and gender research.’ This under-subscription is likely to have resulted
from various factors, including: unfamiliarity, lack of access to data and tools of analysis,
and a misperception that corpus linguistics is difficult or requires advanced knowledge of
computer programming or statistics (Baker 2014: 12). On a theoretical level, researchers
with an interest in language, gender, and sexuality have tended to align with postmodern
positions which reject the possibility of an ‘objective’ research method, and are likely to
view the use of quantification, for example to distinguish language associated with sex, gen-
der, or sexuality, as contributing to the essentialist reification of gendered linguistic norms.
Although discourse studies of gender and sexuality have generally been slow to adopt
corpus approaches in their analyses, there is nonetheless a growing body of research in this
area that has harnessed the possibilities of corpus linguistics to study larger collections of
data and support theory-driven insights with quantitative linguistic evidence. We will first
review a selection of studies of language and gender, before moving on to language and
sexuality. The earliest corpus-based studies of language and gender examined similarities
and differences across corpora of texts (spoken and written) produced by men and women.
Shalom (1997) studied similarities and differences between men’s and women’s personal
adverts. McEnery et al. (2000) compared the prevalence of swearing in speech produced
by men and women in the British National Corpus (BNC). This early interest in the simi-
larities and differences between male and female speech has maintained, with more recent
corpus-based studies exploring gendered language use in more specific contexts. For exam-
ple, Jiménez Catalan and Ojeda Alba (2008) compared the vocabularies of male and female
primary-school children learning English in their written English. More recently, Charteris-
Black and Seale (2010) used corpus methods to compare the ways that men and women
with cancer talk about their illness. The men in their study were found to draw on technical,
medico-scientific language to make sense of and communicate about their illness, while the
women’s accounts tended to focus on emotional disclosure and talk about relationships.
Corpus linguistics methods granted these researchers the opportunity to test gendered
language norms by observing linguistic trends across larger (even population level) groups,
rather than relying on small groups of speakers or researcher introspection. Because of
their commitment to analysing authentic language data (rather than researcher-invented

566
 Gender stereotyping and patient feedback

examples) corpus linguistic studies of language produced by men and women can help in
challenging difference models of gender and language, which erroneously post two gen-
dered ‘species’ of language user, but have little (or no) empirical grounding in terms of how
men and women actually use language. For example, Harrington (2008) combined quantita-
tive and qualitative corpus techniques to compare reported speech in language produced by
men and women. The analysis suggested that speakers’ use of reported dialogue was more
to do with social context than it was to do with gender as a social category.
Another popular strand of corpus-based research into language and gender focuses on
uncovering gendered discourses and ideologies in society. Stubbs (1996) examined gen-
dered expectations of boys and girls manifested in speeches given by the founder of the
Boy Scouts Association, Robert Baden-Powell. Baker (2008: 203–208) compared the dis-
course surrounding the terms bachelor and spinster in the BNC, finding that although the
two words are semantic equivalents (unmarried male and female, respectively), the word
spinster tended to accrue negative attributes, such as sexual frustration and unattractiveness,
while the word bachelor evidenced comparatively positive attributes for young men, such
as being eligible and fun-loving, with older bachelors represented as problematic, and their
status needing to be ‘explained’. The media – especially the print media – has provided
a popular data source for corpus-based studies of gendered representation. For example,
Taylor (2013) compared the use of the terms boy and girl in three corpora of British broad-
sheet newspapers, focusing on similarities as well as differences, while Baker and Levon
(2016) studied the representation of men in UK newspaper articles about masculinity. More
recently, Baker (2014) studied the representation of trans people in a corpus of British news-
paper articles. Other studies have focused on gendered representations in more specialised
corpora of media texts, for example, the Harry Potter book series (Hunt 2015) and pop song
lyrics (Kreyer 2015).
Another area of interest in corpus-based studies of gender is sexist language. In an early
study, Kjellmer (1986) compared the frequencies of male and female pronouns and the terms
man, men, woman and women in American and British English, reporting an overall bias
towards the male terms. Sigley and Holmes (2002) studied the frequency of gender-marked
suffixes, such as -ess and -ette, and reported that such terms declined in use over time. More
recently, Baker (2010) expanded this research by studying British English corpora dated
between 1931 to 2006, finding that while Ms continued to be very rare, the male equivalent,
Mr, was also in decline. Baker hypothesised that this trend might signal the diminishment
(and possible eventual disappearance) of the unequal gender title system.
The majority of corpus-based research into language and sexuality has explored the lin-
guistic construction of gay and lesbian identities, paying particular attention to how these
intersect with desire. For example, Morrish and Sauntson (2011) utilised corpus methods to
study how lesbian desires and identities were represented in a corpus of lesbian erotica from
the 1980s and 1990s. Using the corpus techniques of frequency and collocation, the authors
explored the ways in which ‘lesbian gender’, power, and desire were represented, (re-)pro-
duced, and enacted in their data, which they argued often challenged hegemonic discourses
of gender and sexuality. Similarly, Baker (2002, 2005) studied the changing constructions
of gay men’s identity and desire in a corpus of gay men’s personal adverts in the UK, while
Bogetić (2013) considered a similar corpus of adverts written by Serbian gay teenagers.
A growing area of interest in corpus-based studies of gender and sexuality concerns lan-
guage and homophobic discrimination. Baker (2004) used the keywords method to compare
oppositional stances in House of Lord debates about the equalisation of the age of sexual
consent for gay men. He found that, while the word homosexual was associated with acts,

567
Paul Baker & Gavin Brookes

the word gay was linked to identities. Furthermore, those who argued in favour of law
reform focused on a discourse of equality and tolerance, while those who were against law
reform constructed homosexuality by accessing discourses linking it to danger, ill-health,
crime, and unnatural behaviour. Similar work undertaken on political debates around equal-
ity for LGBT people have been carried out by Bachmann (2011), Love and Baker (2015),
and Findlay (2017). Finally, as part of a broader study into the discourse of white suprema-
cism, Brindle (2016) combined corpus linguistics with critical discourse analysis to examine
the language used by members of a white supremacist web forum to construct heterosexual,
white masculinities and represent gay men and other out-groups.

Section summary
In this section we briefly outline the remaining three chapters in Part VIII of the Handbook.
While all of the chapters in this part have drawn on online sources in order to collect corpus
data, the types of data, and the ways that they relate to gender and/or sexuality are very
different. First, Taylor’s chapter extends the discussion of collocation which began in this
chapter to consider how it can be used in order to imply evaluative meanings at a subcon-
scious level. Taylor considers a range of different statistical techniques for calculating col-
locations as well as considering how altering the span will impact on the type of collocates
retrieved. This is followed by a discussion of commonly used corpus tools for calculating
collocates: Sketch Engine and LancsBox and a survey of studies in language and gender
which have used collocates. Taylor’s case study involves an examination of the phenom-
enon of mock politeness in a 61-million-word corpus of posts taken from the online forum
Mumsnet. Examining terms that have been found to reference mock politeness (like sar-
castic, patronising and bitchy), Taylor uses collocational networks to show how particular
terms and evaluations are associated with different gendered identities. For example, sar-
castic is found to be associated with male identities while bitchy is most strongly associated
with female ones. Taylor’s analysis indicates that general understandings of social interac-
tions are filtered through gendered discourses, with different criteria applied for males and
females.
Hunt’s chapter focuses on the analysis of different types of gendered and sexed identities
in a corpus of South African newspaper articles. This corpus presents particular problems
for corpus analysts due to its multilingual nature and Hunt spends some time discussing how
to identify and search for terms relating to identities when working with multiple languages.
Then, using the corpus analysis tool AntConc, Hunt discusses how dispersion plots can be
used in order to gain an impression of usage of particular terms over time, before moving
on to show how collocation analysis should be ideally combined with analysis of concord-
ance lines in order to gain a deeper understanding of the relationship between two words,
focusing particularly on an analysis of words that are used to refer to people who are gay or
lesbian. The analysis, taking into account a comparison of the two terms in a collocational
network, reveals the extent of intersectionality in representations of gay people – with sexu-
ality and gender functioning as a matrix.
Finally, Potts and Formato examine a smaller corpus (around 27,000 words) of sentencing
remarks from judges where women were the victims of male-perpetrated homicide. When
using small amounts of corpus data, it can often be fruitful to carry out some form of annota-
tion in order to enable a more sophisticated analysis. In this case, Potts and Formato carry
out XML mark-up of the social actors in their corpus in order to indicate the gender, terms

568
 Gender stereotyping and patient feedback

of reference, and grammatical status of each of them. Their subsequent analysis focuses on
how naming strategies are used to construct solidarity or distance between judges and vic-
tims, and how the victims are often described in terms of their (disrupted) relationship with
other people, as sisters, mothers, wives, etc. Subsequent analysis reveals how such women
are described as collections of body parts that have been acted upon violently. Opportunities
to functionalise victims in terms of their careers are less frequent in the corpus, leading Potts
and Formato to discuss their findings in relationship to the extent that judges humanise and
individualise female victims of violence.
The four chapters in Part VIII thus collectively cover a range of commonly used cor-
pus procedures (collocation, concordancing, keywords, dispersion plots, collocational net-
works, annotation) and corpora from different contexts (healthcare feedback, online social
networks, news media, and legal language). Together they demonstrate the potential of cor-
pus linguistics to contribute towards language, gender, and sexuality research, by enabling
analysts to identify frequent or salient linguistic patterns in large sets of texts, which can
then be subjected to more detailed forms of qualitative interrogation. Corpus linguistics
approaches, in identifying patterns that we may not have noticed without the help of statisti-
cal tests and visual models, help to reduce researcher bias to a reasonably acceptable level
as well as giving us a toolbox to facilitate a speedier analysis of very large datasets. While
we do not wish to present corpus linguistics as outclassing existing analytical approaches,
we hope that the chapters in this part will encourage the consideration of this approach as
a complement or supplement to existing toolkits. We particularly note that much gender
and sexuality research within corpus linguistics has been carried out on general reference
corpora and/or newspaper corpora in the English language, so note the potential for the
approach to be extended to other languages and for a wider range of text types. Currently
underexplored types of data which could benefit from a corpus approach could include
transcripts of interviews, advertising, managerial and business texts, fiction (particularly
LGBT fiction, modern fiction, or children’s fiction), online news comments, and online sites
aimed at particular identity groups. Often such data types raise methodological issues in
terms of multimodality which corpus linguists have yet to fully resolve and which are likely
to require the development of new tools and techniques for the future. This is a fast-paced
and moving field and there are many opportunities for new researchers to make their mark.

References
Anthony, L. (2014) AntConc (Version 3.4.3) [Computer Software]. Tokyo, Japan: Waseda University.
Available at: http://www.laurenceanthony.net/
Bachmann, I. (2011) ‘Civil parternship – “Gay marriage in all but name”: A corpus-driven analysis of
discourses of same-sex relationships in the UK Parliament’. Corpora, 6(1), pp. 77–105.
Baker, P. (2002) ‘No fats, femmes or flamers: Changing constructions of identity and the object of
desire in gay men’s magazines’. In: Benwell, Β. (ed.) Masculinity and Men’s Lifestyle Magazines.
Oxford: Blackwell.
Baker, P. (2004) ‘“Unnatural acts”: Discourses of homosexuality within the House of Lords debates
on gay male law reform’. Journal of Sociolinguistics, 8(1), pp. 88–106.
Baker, P. (2005) Public Discourses of Gay Men. London: Routledge.
Baker, P. (2006) Using Corpora in Discourse Analysis. London: Continuum.
Baker, P. (2008) ‘“Eligible” bachelors and “frustrated” spinsters: Corpus linguistics, gender and
language’. In: Sunderland, J., Harrington, K., and Stauntson, H. (eds.) Gender and Language
Research Methodologies. Basingstoke: Palgrave Macmillan, pp. 73–84.

569
Paul Baker & Gavin Brookes

Baker, P. (2010) ‘Will Ms ever be as frequent as Mr? A corpus-based comparison of gendered terms
across four diachronic corpora of British English’. Gender and Language, 4(1), pp. 125–149.
Baker, P. (2014) Using Corpora to Analyze Gender. London: Bloomsbury.
Baker, P. and Levon, E. (2016) ‘“That’s what I call a man”: representations of racialised and classed
masculinities in the UK print media’. Gender and Language, 10(1), pp. 106–139.
Baxter, J. (2003) Positioning Gender in Discourse: A Feminist Methodology. Palgrave Macmillan:
Basingstoke.
Bogetić, K. (2013) ‘Normal straight gays: Lexical collocations and ideologies of masculinity in
personal ads of Serbian gay teenagers’. Gender and Language, 7(3), pp. 333–367.
Brindle, A. (2016) The Language of Hate: A Corpus Linguistic Analysis of White Supremacist
Language. London: Routledge.
Charteris-Black, J. and Seale, C. (2010) Gender and the Language of Illness. Basingstoke: Palgrave
Macmillan.
Findlay, K. (2017) ‘Unnatural acts lead to unconsummated marriages: Discourses of marriage within
the House of Lords debate on same-sex marriage’. Journal of Language and Sexuality, 6(1), pp.
30–60.
Gallagher, M., Pearson, P., Drinkwater, C., and Guy, J. (2001) ‘Managing patient demand: A qualitative
study of appointment making in general practice’. British Journal of General Practice, 51, pp.
280–285.
Garside, R. and Smith, N. (1997) ‘A hybrid grammatical tagger: CLAWS4’. In: Garside, R., Leech, G.,
and McEnery A. (eds.) Corpus Annotation: Linguistic Information from Computer Text Corpora.
London: Longman, pp. 102–121.
Hammond, J., Gravenhorst, K., Funnell, E., Beatty, S., Hibbert, D., Lamb, J., Burroughs, H.,
Kovandžić, M., Gabbay, M., Dowrick, C., Gask, L., Waheed, W., and Chew-Graham, C. (2013)
‘Slaying the dragon myth: An ethnographic study of receptionists in UK general practice’. British
Journal of General Practice, 63(608), e177–e184.
Hardie, A. (2012) ‘CQPweb – Combining power, flexibility and usability in a corpus analysis tool’.
International Journal of Corpus Linguistics, 17(3), pp. 380–409.
Harrington, K. (2008) ‘Perpetuating difference? Corpus linguistics and the gendering of reported
dialogue’. In: Sunderland, J., Harrington K., and Stauntson, H. (eds.) Gender and Language
Research Methodologies. Basingstoke: Palgrave Macmillan, pp. 85–102.
Hunt, S. (2015) ‘Representations of gender and agency in the Harry Potter series’. In: Baker, P.
and McEnery, T. (eds.) Corpora and Discourse Studies: Integrating Discourse and Corpora.
Basingstoke: Palgrave Macmillan, pp. 266–284.
Kilgarriff, A., Baisa, V., Bušta, K., Jakubíček, M., Kovář, V., Michelfeit, J., Rychlý, P., and Suchome,
V. (2014) ‘The sketch engine: Ten years on’. Lexicography, 1(1), pp. 7–36.
Kjellmer, G. (1986) ‘“The lesser man”: Observations on the role of women in modern English writings’.
In: Aarts, J. and Meijs, W. (eds.) Corpus Linguistics II. Amsterdam: Rodopi, pp. 163–176.
Kreyer, R. (2015) ‘“Funky fresh dressed to impress”: A corpus-linguistic view on gender roles in pop
songs’. International Journal of Corpus Linguistics, 20(2), pp. 174–204.
Jiménez Catalán, R. M. and Ojeda Alba, J. (2008) ‘The English vocabulary of girls and boys:
Similarities or differences? Evidence from a quantitative study’, in Harrington, K., Litosseleti, L.,
Sauntson, H. and Sunderland, J. (eds) Gender and Language Research Methodologies. London:
Palgrave Macmillan, pp. 120–132.
Love, R. and Baker, P. (2015) ‘The hate that dare not speak its name?’. Journal of Language Aggression
and Conflict, 3(1), pp. 57–86.
McEnery, T., Baker, P., and Hardie, A. (2000) ‘Swearing and abuse in modern British English’. In:
Lewandowska-Tomaszczyk, B. and Melia, J. (eds.) PALC 99 Practical Applications in Language
Corpora. Frankfurt: Peter Lang, pp. 37–48.
McEnery, T. and Hardie, A. (2012) Corpus Linguistics: Method, Theory and Practice. Cambridge:
Cambridge University Press.

570
 Gender stereotyping and patient feedback

McEnery, T. and Wilson, A. (2001) Corpus Linguistics: An Introduction. Edinburgh: Edinburgh

University Press.
Mills, S. (2003). Gender and Politeness. Cambridge: Cambridge University Press.
Mills, S. and Mullany, L. (2011) Language, Gender and Feminism: Theory, Methodology and
Practice. London: Routledge.
Morrish, L. and Sauntson, H. (2011) ‘Discourse and identity in a corpus of lesbian erotica’. Journal of
Lesbian Studies, 15(1), pp. 122–139.
Nursing and Midwifery Council (2014) Equality and Diversity Annual Report. Available at: https
://www.nmc.org.uk/globalassets/sitedocuments/annual_reports_and_accounts/th-equality-and-d
iversity-annual-report---english-january-2015.pdf
Rayson, P. (2008) ‘From key words to key semantic domains’. International Journal of Corpus
Linguistics, 13(4), pp. 519–549.
Scott, M. (2016) WordSmith Tools Version 7. Stroud: Lexical Analysis Software.
Shalom, C. (1997) ‘That great supermarket of desire: Attributes of the desired other in personal
advertisements’. In: Harvey, K. and Shalom, C. (eds.) Language and Desire. London: Routledge,
pp. 186–203.
Sigley, R. and Holmes, J. (2002) ‘Looking at girls in corpora of English’. Journal of English
Linguistics, 30(2), pp. 138–157.
Stubbs, M. (1996) Text and Corpus Analysis: Computer-Assisted Studies of Language and Culture.
Oxford: Blackwell.
Sunderland, J. (2004) Gendered Discourses. Basingstoke: Palgrave Macmillan.
Taylor, C. (2013) ‘Searching for similarity using corpus-assisted discourse studies’. Corpora, 8(1),
pp. 81–113.
Widdowson, H. G. (2004) Text, Context, Pretext: Critical Issues in Discourse Analysis. Oxford:
Blackwell.
Yar, M., Dix, D., and Bajekal, M. (2006) Socio-Demographic Characteristics of the Healthcare
Workforce in England and Wales – Results from the 2001 Census. Newport: Office for National
Statistics.

571
 38
Investigating gendered language
through collocation
The case of mock politeness

Charlotte Taylor

Introduction/defnitions
Overview
This chapter makes use of the notion of ‘collocation’, a key concept in corpus linguistic
work. Collocation refers to the tendency of certain words or phrases to occur together with
other words and phrases. It is an important aspect of language because it gives us a way
of understanding the associations that particular words or phrases may carry for language
users. While collocation comes from corpus linguistics work, and as such pertains to a more
quantitative approach to language studies, it has been extensively used in the subfield which
combines corpus linguistics and (critical) discourse analysis and sits astride the traditional
qualitative/quantitative divide.
Collocation analysis gives a way in to understanding how words and phrases are used, and
the associations they trigger, which is essential to identifying the discourses that surround
the representation of groups – a topic which is often of interest in research on language,
gender, and sexuality. I would argue that collocation can also be an important preparatory
stage in variation studies in this area. That is to say, before we can ask who does ‘x’ (most
frequently), we need to know the full semantic and pragmatic profile of ‘x’ to be sure that
we are measuring what we intend. In the case of mock politeness, which is the focus of the
case-study in this chapter, previous research regarding gender has tended to report that men
are more likely than women to perform sarcasm and more likely to perform patronising
behaviours in mixed-sex interactions. In this chapter, I aim to step back from these binary
comparisons to question whether the terms sarcastic and patronising are themselves gen-
dered. For instance, is there a tendency to use sarcastic to describe behaviour by a male
speaker when a different label would have been applied to the same behaviour if it had been
performed by a female speaker?
The chapter starts by defining what is meant by collocation and how it can be inves-
tigated in studies of language, gender, and sexuality, including some guidance for good
practice in the area. The case study is then presented in which collocation is employed to
investigate sarcastic and patronising. The chapter ends with indications of future directions

572
 Collocation and mock politeness

and recommendations for further reading for scholars interested in using collocation analy-
sis to investigate the relationship between language, gender, and sexuality.

What is collocation?
Collocation is a fundamental notion within corpus linguistics, and is perhaps best summed
up by Firth (1957: 11) who famously stated that ‘you shall judge a word by the company
it keeps’. The Firth quotation is important because it sums up not just how we understand
collocation (the relationship between words), but why we are interested in collocation. That
is how knowing which words tend to go together can tell us more about the contextual
meanings (including evaluative potential) of the item we are particularly interested in. The
role of corpus linguistics here is in allowing us to look at a greater number of instances than
would be feasible by manual analysis, and in giving us information about the significance
of collocation.
There are two different ways of thinking about the significance of collocation: the first
relates to the identification of collocation and the second relates to why it is of interest to
those of us studying language. Starting with the first, here we are trying to identify which
collocates or pairings of words we should consider to be significant. To address this point we
might want to go back to defining collocation. According to Stubbs (2001: 29), a collocate is

a word-form or lemma which co-occurs with a node [the word the researcher is inter-
ested in] in a corpus. Usually it is frequent co-occurrences which are of interest, and cor-
pus linguistics is based on the assumption that events which are frequent are significant.

However, frequency alone is insufficient because, as Biber, Conrad, and Reppen (1998:
265) remind us, ‘more common words are more likely to occur in a collocate pair simply
by chance’. Thus it would be overly simplistic to look at frequency alone, and potentially
uninformative for our research purposes. For this reason, in the practical sections we will
look at which measurements can be used to identify collocates.
To take the second aspect of significance in relation to collocation, we need to consider
more fundamentally why word pairings should be of interest to a researcher interested in
investigating the relationship between gender and sexuality and language. Here, the reason
that collocation is seen as meaningful is because it gives us a way in to understanding the
evaluative meanings that are potentially bundled up with any lexical item.
In the most obvious cases, the evaluative meaning actually is the central meaning, as for
instance in the term bitch where it is used as a term of abuse for a woman. In this context,
we cannot separate out the negative evaluation from the thing to which the word refers; the
evaluation is entirely intrinsic to the word. The unfavourable meanings, or negative conno-
tations, are absolutely apparent to the fluent speaker and not in any sense peripheral or hid-
den. Similarly, to take a well-known gender pair, the connotations of spinster and bachelor
are quite apparent to a fluent speaker; the difference between the two is not just that one is
used for an unmarried man and one for an unmarried woman, but that bachelor evaluates
that unmarried status more favourably than spinster.
Then, there are lexical items which are less obviously evaluative in function. For instance,
Cameron (2003) discusses the term openly gay. At first glance, we might not see anything
so obviously evaluative here but, as Cameron urges, we only have to think about what else
we might describe someone as being openly. Using a corpus, we can check those intuitions

573
Charlotte Taylor

and, according to the general corpus EnTenTen13 (described below), the words which have
the strongest collocation with openly are, in order of strength: gay, hostile, homosexual,
lesbian, racist, contemptuous, bisexual, critical, defiant, anti-semitic. Apart from the words
that refer to sexuality and perhaps ‘defiant’, we might note that these are mainly attributes
that are unlikely to be used to describe someone the speaker admires (hostile, contemptuous,
critical) and describe highly offensive views (racist, anti-semitic). With this information to
hand, we might reconsider whether openly gay is a neutral expression. This tendency for a
lexical item to regularly occur with items that have favourable or unfavourable connotations
has been discussed in corpus linguistics as ‘semantic prosody’ or ‘evaluative prosody’. It
is often referred to as the ‘aura’ that a word carries (following Louw 1993) and often we
only realise that a word has a particular prosody when the expected pattern is broken. For
instance, does openly friendly sound usual or familiar?
In addition to these evaluative meanings, there may be other connotations that are held as
part of our conscious or unconscious knowledge of a lexical item, and these are the aspects
now likely to be described as ‘discourse prosodies’. To take an older example, Sinclair
(2004: 38) discusses how the phrase my place has a prosody of ‘informality’ and ‘invitation’
as in ‘Would you like come to back to my place for a while?’. The uptake of collocation and
connotation in studies of language, gender, and sexuality is discussed in more detail below.

Using collocation in studies of language, gender, and sexuality

In a survey of work that employed corpus linguistics approaches in order to study the rela-
tionship between gender and sexuality and language, I identified 47 articles published in
international journals between 2006 and 2016. Of these, 21 focused on variation in lan-
guage used by people classified as belonging to different gender or sexuality groups and 22
focused on the representation of people classified as belonging to different gender or sexual-
ity groups. Collocation was used as a tool only in the latter group of representation-based
studies and was employed in 11 of those papers, so nearly one quarter of the studies overall.
If we consider why collocation tends to be used most frequently in representation studies, it
is likely to be because of its close relationship to discourse and therefore to ideology.
As outlined in the previous section, the study of collocation provides us with an entry
point to our data in terms of understanding what evaluations and other types of connotation
may be packaged up with certain lexical items in given contexts. We may consider the con-
tribution of collocation analysis as affecting two principal areas within discourse studies. In
the first, as Bogetić (2013: 334) puts it, ‘collocation analysis offers a productive means for
understanding ideology, as lexical co-occurrence may shed new light on complex webs of
identities, discourses and social representations in a community’. Indeed it is because of this
notion of ‘webs’ that the concept of the collocational network is particularly useful. We may
not be consciously aware of how particular ideas are persuasively grouped in discourses
and so the analysis of collocates gives us a means for drawing out these connections and
non-obvious meanings. The second important use of collocates is that they ‘can be useful in
revealing how meaning is acquired through repeated uses of language, as certain concepts
become inextricably linked over time’ (Baker 2014: 13). If we consider discourse to be
cumulative, then looking at the accumulated associations around particular lexical items can
help make this process more evident.
In the papers surveyed for the literature review, 6 of the 11 focused explicitly on the
gendered noun pairs man/woman and girl/boy and all but one focused on gendered nouns.
This focus on gender pairs may represent a strength of the approach because analysing

574
 Collocation and mock politeness

the collocation patterns allows us to discover new information about how such seemingly
straightforward pairs are used in different domains. However, it also shows where more
research may be required in collocation studies of language, gender, and sexuality because
it reveals that the research to date tends to operate along binary gender lines. What this
suggests is that the knowledge which is valued is difference between two genders. More
recent research into these gendered nouns, such as Baker and Levon (2016), digs deeper by
examining the different language surrounding racialised and classed pre-modifiers of ‘man’
and this intersectional approach reflects more accurately the wider movement in gender and
sexuality studies.

Good practice in collocation analysis

In terms of good practice, those principles that apply to collocation analysis are the same as
those that apply to most research and certainly most corpus linguistic work.

Transparency
When reporting collocation analysis, provide sufficient information for the reader to under-
stand exactly how you manipulated your data. This would include details of which word
forms you searched for, what software you used, the statistical measure of strength of col-
location, the cut-off points implemented, and the span examined for collocates (how many
to the left and how many to the right of the node). Ideally, you should aim to produce enough
information for a reader to repeat the analysis, thus fulfilling the goal of ‘replicability’.
Although space is often an issue, it is usually possible to include this information at least
as a footnote. The reason this is so important is that there are different ways of calculating
collocation and these will produce different results (see discussion in Baker 2014; McEnery
and Hardie 2012). To take an example which is relevant to the case study we discuss in the
following section, if we look at the collocates of sarcastic in EnTenTen13 (a very large
corpus of texts gathered online), the Sketch Engine software (Kilgarriff et al. 2014) offers us
four different measures for calculating collocates and the results are displayed in Table 38.1.
The number given to the right-hand side of each column tells us how often these two words
occur together.

Table 38.1 Comparison of collocate ranking according to different measures

Frequency t-score Mutual information logDice

and 12,655 And 12,655 insultsude 14 witty 684

a 11,499 a 11,499 quotesarcastic 11 snarky 277
the 9,216 I 7,848 jerksarcastic 10 Revive 233
I 7,848 the 9,216 sarkastisches 5 remark 597
to 6,974 to 6,974 bithcy 14 cynical 469
of 6,236 being 5,070 fringeheads 13 remarks 928
being 5,070 of 6,236 raucher 3 humor 1,183
is 4,568 was 4,252 fringehead 8 snide 151
in 4,363 is 4,568 Fringehead 7 wit 430
was 4.252 be 3,835 Marcot’s 8 Dont 323

575
Charlotte Taylor

In the first column we see that the ten most frequent collocates are mostly function words
and this is not particularly surprising because these are the most frequent types of words
in the corpus overall: in fact seven of the ten most frequent collocates are also among the
ten most frequent words in the corpus. The following three columns use different statisti-
cal measures to calculate the likelihood of two items occurring together by chance and we
can see great variation in the kinds of words that are identified. The t-score measure in the
second column privileges high frequency of co-occurrence and indeed the ranking is very
similar to the simple frequency ranking shown in the first column (nine of the top ten words
are shared). By contrast, the words identified as significant collocates using the mutual
information calculation (the third column) appear to be mainly usernames. They are very
low frequency overall and are reported because the calculation foregrounds items that occur
very infrequently and nearly always occur together with the node (sarcastic). In the fourth
column we see that the logDice calculation provides items that are neither particularly high
nor particularly low in overall frequency and so looks the most likely to be productive in
better understanding the company kept by sarcastic. So we can see that the choice of
statistic will depend on the purpose and ‘the sort of words that the researcher is interested
in obtaining’ (Baker 2006: 102) and needs to be reported because the results can vary so
greatly. The same would apply to other factors which will affect the words that are calcu-
lated to be collocates such as the span and cut-off points that are chosen by the researcher
and/or set as default in the software.

Total accountability
When discussing collocates, it is good practice to show all the collocates that were gathered
using the chosen settings and criteria. This is important in order to fulfil the requirements
of ‘total accountability’, a term coined in Leech (1992), which encapsulates the principle
that we account for all findings and do not simply select those which are favourable to our
hypothesis. Restrictions on space often mean that we have to be selective about what we
discuss, but it is usually possible to use the appendices to list all collocates so the reader can
get a fuller picture.

Generalisations
Collocates are register specific and so we should be careful not to assume that the collocates
found in one corpus will be found elsewhere. For instance, in a study of newspaper language
I found that collocates of girl and boy which remained constant over a 12-year time period
related to violence (for girl only: abduct, burn, death, lure; for boy only: wound). Within
the context of newspaper discourse, these offered interesting routes for investigation but it
is also the case that they are a result of the news values of our press which tends to report
negative events.

Context
When we examine collocates, we need to make sure we do not look at these lexical items
in isolation. This is particularly the case when categorising collocates because that process
of categorisation is a process of interpretation of meaning. Therefore, a key part of colloca-
tion analysis involves going back to the text (often via the concordance line) to see how the
terms were used in that specific set of data. For instance, in my study of girl and boy in the
press sometimes the two shared the same collocate but closer investigation showed that one

576
 Collocation and mock politeness

was the object and the other was the subject of the same verb, which requires a different
interpretation.

Background
The last suggestion for good practice is to remember to look outside the corpus at the wider
field of language, gender, and sexuality studies. The rationale for filling a gap cannot simply
be that no one has done a collocation study of ‘x’ previously; there needs to be a theoreti-
cally grounded reason as to why that is a worthy topic of investigation. More fundamentally,
the researcher needs to be aware of how the theoretical framework in the field has developed
and can support their research. As mentioned above, it is perhaps the case that collocation
studies lag behind shifts in focus in the broader field at present.

Case study: gendered language and mock politeness

Background
In the case study, I explore two issues relating to the topic of mock politeness and gender.
Mock politeness is defined as occurring when there is an im/politeness mismatch leading
to an implicature of impoliteness (see Taylor 2016 for a more detailed discussion). Thus, it
encompasses utterances such as those in Examples (1) and (2) which come from the corpus
of forum interactions used in the case study:

(1) Lift came, doors opened, we stepped forward to get in and were almost knocked down
by a couple with their own pushchair. I’m afraid I was rather sarcastic and exclaimed
“Don’t mind the queue”;
(2) People that carry on like you Alba, are often described as twattish, or a bit of a tit. hth.
[hope that helps]

In the first, the impoliteness mismatch comes from the apparently polite move of ‘Don’t
mind the queue’ and the context in which the targets had already entered the lift ahead of
the speaker. The speaker intends that the incompatibility of what is said and the context lead
to an implicature that what is meant is a reproach for non-observance of a perceived social
norm (the impolite move). In the second, the mismatch is made more explicit as it occurs
at the textual level where the polite move is given at the end of the utterance (‘hth’) follow-
ing on from an impolite move in which the speaker associates the target with unfavourable
characteristics.
The term ‘mock politeness’ is probably not one that you use in ordinary conversation,
and as such we may consider it a ‘second-order’ label, which is to say it is a label for an
academic concept (as summarised in the definition above). In previous work (Taylor 2016),
I have found that the following are all ‘first-order’ or lay labels which were used to describe
mock polite behaviours in a corpus of British English forum interactions: patronising, bit-
ing, make fun, condescending, cutting, caustic, mock, bitchy, tease, passive aggressive, put
down, overly polite, sarcastic. This distinction between first- and second-order uses (see
for instance Watts et al. 1992), between the lay and academic constructs and terms, is an
important one, particularly when trying to elicit data, because any researcher needs to be
confident that they are employing terms with which the participants are familiar and which
may be used to describe the full range of behaviours under study and all kinds of people
who perform those behaviours. So, for instance, if a researcher wanted to collect accounts

577
Charlotte Taylor

of mock politeness, they would not ask interviewees ‘Can you tell me about a time someone
was bitchy to you’ unless they wanted to focus on stories of female mock politeness. In this
case the gendering of the lexical item is obvious, in other cases, the researcher might need
to study the collocates to check for such bias.
In the following sections I briefly describe the data used here and then introduce two
areas that overlap significantly with mock politeness and which have consistently been asso-
ciated with male behaviour: sarcasm and patronising behaviour.

Data
The dataset used in this study comes from an online forum which was selected because it
allows access to ‘everyday’ or ‘conversational’ comments on mock politeness, while retain-
ing much of the context. The forum, mumsnet.com, is UK based and predominantly popu-
lated by people presenting as women (an imbalance which clearly has implications for any
discussion of gender). By way of illustration of the size, as of January 2015, Mumsnet claims
to have over 14 million visits per month (Mumsnet 2015). The 61-million-token corpus was
compiled from the forum using the free software BootCat (Baroni and Bernardini 2004),
which gathers text from entire webpages using seeds (search words).1 The EnTenTen13 cor-
pus, which is available through Sketch Engine, was also used.2 This is an English language
corpus of online texts and contains approximately 19 billion words (Jakubíček et al. 2013).

Tools
The latest tool for visualising connections between words is Lancsbox (Brezina et al. 2015,
applied in Baker and McEnery 2015) which shows the way that collocates link to the node
and also to one another.3 The importance of visualising the collocational network is that it
allows us to see the company that a word is keeping and, crucially, it places that company in
context. As Brezina et al. (2015: 141) state, ‘[c]ollocates of words do not occur in isolation,
but are part of a complex network of semantic relationships which ultimately reveals their
meaning and the semantic structure of a text or corpus’. Furthermore, because the networks
can be displayed simultaneously, it is also possible that we may be able to identify what is
absent (Duguid and Partington 2018) by noting which items collocate with some nodes and
not with others.
The Sketch Engine thesaurus (Rychlý and Kilgarriff 2007) allows us to see which words
share similar collocates. It works by identifying collocates for a search word and then in
the second stage identifies other words which share those collocates. So, for instance, in
the previous study (Taylor 2013) of boy and girl in a corpus of newspaper texts, the Sketch
Engine thesaurus identified the word with the most similar collocates to boy as girl and vice
versa (more revealingly, the second word in the girl list was woman, while the second word
in the boy list was child).

Sarcasm
Starting with research into sarcasm, to date attention from a language, gender, and sexuality
perspective has primarily focused on variation in use as correlated with gender. The issue
of frequency of use has received most attention and the consensus has been that men use
sarcasm more than women. The most common measurement has involved self-assessment,
for instance, Dress et al (2008: 83, my italics) asked participants the following questions:

578
 Collocation and mock politeness

1) What is the likelihood that you would use sarcasm with someone you just met?
2) How sarcastic do you think you are?
3) What is the likelihood that you would use sarcasm when insulting someone?
4) What is the likelihood that you would use sarcasm with your best friend?

The majority of studies using this method found men self-reported as being sarcastic
more often than women (e.g. Bowes and Katz 2011; Dress et al. 2008; Milanowicz 2013;
Rockwell and Theriot 2001). This method assumes that the participants are both self-aware
and truthful and, perhaps not surprisingly, two of these studies (Bowes and Katz 2011;
Dress et al. 2008) found that although the male participants reported using sarcasm more
than the female participants, they did not do so in elicitation tests. Furthermore, the choice
of metalanguage suggests a problematic blurring between first- and second-order uses. The
use of the word ‘sarcastic/sarcasm’ when interacting with participants means that they will
not answer with reference to the researcher’s second-order concept of sarcasm, that is, the
scientific construct, but the kinds of contexts in which they personally would describe a
behaviour as ‘sarcastic’, that is, the first-order understanding. However, we know that lay
and academic uses of ‘sarcasm’ are not the same (e.g. Creusere 1999) and lay uses will be
influenced by sociolinguistic variables, including whether these terms are gendered.
In terms of expectations of gendered performance, previous research again points towards
an association of sarcasm with male behaviour. In experimental conditions, Colston and Lee
(2004) reported that speakers of sarcastic utterances were more likely to be assumed to be
male. Furthermore, Katz, Piasecka, and Toplak (2001) found that the perceived gender of
the producer of a sarcastic utterance affected processing, with reading times for texts featur-
ing male producers of sarcasm being lower than for female producers. This was interpreted
as occurring because ‘sarcasm is more likely to be associated with males than females,
comprehension of noncanonical usage is delayed as people attempt to integrate the text they
are reading with their stored “knowledge” (stereotypes) of men and women’ (Katz et al.
2004: 187). Indeed, what is not clear, and what this case study aims to address, is the extent
to which these gender effects are the results of stereotypes or actual gendered tendencies.
For instance, Katz et al. (2004: 187, my italics) report that ‘when the gender of the speaker
is manipulated in a textoid [a short text], the same comment is rated as more sarcastic when
made by a male than when made by a female’. This suggests that participants are drawing
on stereotypes in associating sarcasm with male speakers.
To investigate the potential gender associations with sarcastic we start by visualising the
collocates. Figure 38.1 visually displays the metapragmatic labels which I had previously

Figure 38.1 Gendered collocates of adjectival mock politeness labels.

579
Charlotte Taylor

found to indicate mock politeness and all their collocates which were gendered terms refer-
ring to people. In the LancsBox visualisation, the length of the line linking the search word
and any given collocate reflects the strength of collocation between those items. The items
in the central positions (bitchy, patronising etc.) are the nodes which are entered manually,
and the items radiating out are the collocates.4
Two metapragmatic labels which emerge as gendered are bitchy on the right (the only
one collocating with mil [mother-in-law], girls, mum, dd [dear daughter], sister) and patron-
ising (the only one collocating with man, ds [dear son], his, bloke, he’s). However, these two
items also share a large number of collocates referring to both male and female participants
(she, her, mums, she’s, women and dh [dear husband], men). From this measure, sarcastic
and passive aggressive collocate with just he and she which suggests a more neutral set
in terms of gendering. Obviously at this point, what we do not know is how the gendered
items relate to the node, that is – who is being described as mock polite and to whom they
are being mock polite? This could be tackled by analysing collocates using a tool like Word
Sketch which allows us to draw on grammatical (part of speech) information.
In this case study, the behaviours described by the metapragmatic labels were all ana-
lysed to identify the gender of the performer, that is, the person who was described as being
‘sarcastic’ etc. As Figure 38.2 shows, there were gender preferences for the different labels.5
The most gendered metapragmatic label of those examined was ‘bitchy’, which showed
a semantic preference for describing female behaviour while the item most strongly associ-
ated with male behaviour in terms of statistical significance (measured using log-likelihood)
was ‘sarcastic’; these uses are illustrated in Examples (3) and (4) with key terms in bold.

(3) i ended up telling a couple of bitchy customers [I was pregnant], because I was lying
down on the floor because I felt sick as shit, and this random woman came in and snot-
tily said ‘oh! having a lie-down are we?’ ‘yes, I replied, I’m pregnant and feel sick’.
(4) DH [dear husband] is happy for me to be at home BUT he moans at me if the house isn’t
tidy or I get behind. He is sarcastic and says things like ‘I know you’re really busy’ or
‘if you could spare the time’…

100%
90%
80%
70%
60%
50%
40%
30% female
20% male
10%
0%

Figure 38.2 Distribution of male/female performance of behaviours.

580
 Collocation and mock politeness

Figure 38.3 Evaluative collocates of adjectival mock politeness labels.

From Figure 38.2 we see that both bitchy and sarcastic carry gendered associations in terms
of who they tend to describe. So when studies of gender ask participants to self-report the
extent to which they are sarcastic, they are asking participants to apply a gendered label to
their own behaviour. Understanding that sarcastic is actually gendered helps explain why
men self-report as sarcastic more frequently than women even though there is no evidence
of differences in actual practice: that is to say they self-report as sarcastic because they asso-
ciate the term with a male behaviour (what remains unknown at this point is how a woman
who performs the second-order construct of sarcasm would self-describe).
The collocation analysis also reveals differing patterns of evaluation for the labels inves-
tigated here, as shown in Figure 38.3. Bitchy and sarcastic and patronising share negative
collocates (rude, mean). Additionally, bitchy collocates with a set of strongly unfavourable
evaluations (e.g. nasty, awful) while those for sarcastic are less strong (e.g. dry, odd). Thus,
it appears that the two most gendered labels (sarcastic for male behaviour and bitchy for
female behaviour) also carry very different evaluative prosodies, indicating the potential for
male and female participants to be judged differently for similar mock polite behaviours.
To further explore the evaluative connotations, Sketch Engine Thesaurus was used to
investigate which terms share similar lexical environments in a much larger web corpus
(EntenTen13). The results are visualised in Figure 38.4, which presents those items with
the most shared collocates in the biggest font (the variation in shading is simply for ease of
reading).
From Figure 38.4 we can observe that terms with similar collocates include those previ-
ously identified as mock politeness labels (e.g. condescending, ironic) alongside a mixture
of apparently favourable (e.g. humorous, witty) and unfavourable (e.g. obnoxious, dismiss-
ive) labels, in addition to some which would need context to disambiguate (e.g. irreverent).
What emerges then is a picture in which the label sarcastic is more likely to be used to
describe a man’s behaviour and that the evaluation of the lexical item is mixed, with both
favourable and unfavourable interpretations co-existing.

Patronising
As a second-order label, patronising is usually understood to encompass behaviours which
might also be labelled as condescending, another word that is under-investigated in terms of
gender connotations. As Culpeper (2011: 95) notes, ‘[b]eing patronized involves a kind of
‘double whammy’: your face is devalued in some way, but it is also devalued in a particular
relational context that does not licence the “patroniser” to do so’. This latter point about
‘licence’ invokes another key aspect: power. Thus, it is perhaps not altogether surprising

581
Charlotte Taylor

bitchy judgmental
humorous quirky
argumentative
mean-spirited

condescending
childish
cocky
insensitive
hateful

cynical
manipulative

obnoxious wry funny

ironic
grumpy
witty
sardonicinsulting
edgy
defiant contemptuous cheeky

snarkydismissive
angry
opinionated clever

irreverent
selfish
provocative
silly
arrogant amusing
disrespectful
sly
vulgar brach

goofyrude
comical snide cruel
hilarious satirical
nasty
smug
impatient
sassy
confrontational playful mischievous
bitter
stubborn
tongue-in-cheek caustic rebllious
spiteful

Figure 38.4 Sketch Engine Thesaurus output for ‘sarcastic’.

that gender has often been a dominant focus of study in research into patronising behaviour
from social psychology (alongside intergenerational interactions), in particular, so-called
‘benevolent sexism’ (Glick and Fiske 1997). In the conceptualisation of patronising and
condescending behaviours in these areas, mismatch is given a central role because it is
assumed that the patronising speaker is under-estimating the competence of the hearer.
This potentially generates an im/politeness mismatch between the ostensibly helpful utter-
ance and the simultaneous devaluing of the target by that under-estimation of competence.
Another area where patronising behaviour has been researched in relation to gender is work
on intimate relations. For instance, Buss (1989) identified 147 sources of upset (essentially
impoliteness as it is conceptualised here) that men perform on women and vice versa. One
of these factors was labelled as ‘condescending’ and this was more frequently complained
about by women with regard to men’s behaviour than vice versa.
To investigate the potential gendering of the term patronising itself, we can return to
Figure 38.1 in which we saw that the collocates of patronising included more male par-
ticipants than other mock politeness labels (man, ds [dear son], his, bloke, he’s, he, dh
[dear husband], men) but also collocated with female participants (girl, she, her, mums,
she’s, women), leaving the picture unclear. As previously, we may look to the larger corpus
of EnTenTen13 to see what terms are used in similar environments through the Sketch
Thesaurus function. The findings are reported in Figure 38.5 which shows there is a distinct
negative evaluative prosody and the presence of lexical items with clear gender associations
(paternalistic, misogynistic, sexist).
However, the data reported in Figure 38.2 did not show that this label was more likely
to be used to discuss male behaviour in the context of the forum. One explanation for this
which could be explored in further work is that patronising is not actually gendered in the
same way as sarcastic but that it is so strongly associated with behaviours performed by
those in positions of power that in many contexts it is biased towards male participants
because they are more likely to hold such positions in societies.

582
 Collocation and mock politeness

contemptuousthoughtless
conceited
derogatory
self-indulgent inane
haughty
insincere paternalisticdismissive sarcastic
self-important preachy overbearing
self-confided

condescending
whiney
pretentious callous crass derisive

self-serving
judgmentalsmug
unhelpul
snotty
pedantic narrow-minded
judgementalsnobbish self-righteous
insultingsupercilious
smarmy snidepompousglibself-centered
self-accorded bigoted

egotisticalsexist mean-spiritedsneering
pushy inconsiderate
colum

disingenuous tritemisogynistic flippant sanctimonious

presumptuous prejudiced snarkyspiteful self-congratulatory
manipulative disdainful
belligerent

Figure 38.5 Sketch Engine Thesaurus output for ‘patronising’.

Summary
In this short case study, I hope to have shown how we can use collocation analysis as an
entry point to our data and not only to start to unravel complex webs of connotations in rela-
tion to representation of gender but also as a preparatory stage for variation-focused studies
of language and gender.
What collocation analysis can offer language, gender, and sexuality studies is an empiri-
cal basis for discussions of connotations and the two tools used here, LancsBox and the
Sketch Engine Thesaurus, go further by offering visualisation techniques for displaying
collocation patterns. The use of such techniques may help to make the integration of col-
location analysis more accessible and something that can supplement language, gender,
and sexuality research coming from a broad range of methodological backgrounds (such
as those discussed and illustrated in this volume), in addition to forming the centre point in
research coming from corpus linguistics.

Future directions
Given the adaptability of the method, the range of topics ripe for investigation using col-
location research in language, gender, and sexuality studies is almost as wide as the range of
topics in the subject area itself. Labels can be interrogated for non-obvious meanings across
a wide spectrum of questions. This may constitute the main scope of the research, as in some
of the examples below which centre on the collocates of gendered nouns to understand how
men and women are represented in public discourses. Alternatively, the collocation analysis
may be integrated as a way of offering another ‘way in’ to the data, as a form of triangula-
tion alongside frequency analysis or non-corpus methods. Finally, the collocation analysis
may be a preparatory stage to sociolinguistic variation investigations, as discussed above.
In terms of which future directions are likely to be particularly beneficial for the
field, there is scope for research that takes a non-binary and/or intersectional approach in

583
Charlotte Taylor

collocation analysis of language, gender, and sexuality. From a methodological perspective,

it would be highly interesting to see variation studies which look beyond frequency and
integrate collocation analysis.
At a larger scale, important avenues for future research within language, gender, and sex-
uality include studies of how lexical items accumulate and shed connotations over time, and
the rise of diachronic corpora will facilitate this. Another significant avenue is to achieve
better understanding of the extent to which people are aware of, or can be made aware of,
the prosodies surrounding particular lexical items and this will be significant for creating
impact from collocation studies. Finally, as a broad long-term goal, understanding how the
networks of evaluations and connotations relate and intertwine in representing and con-
structing gender and sexuality is an important area for future attention.

Notes
1 See Taylor (2016) for the full corpus description.
2 Sketch Engine is currently free to universities in EU member states.
3 Freely available from http://corpora.lancs.ac.uk/lancsbox/
4 Calculated using a span of 5L/R, logdice, minimum collocation frequency of 5.
5 Only those instances where the behaviour of a third person was being described were counted in this
stage because this mean that the gender was more likely to be specified and to avoid bias from the
fact that the first- and second-person references would be disproportionately female in this particular
corpus.

Further reading
Baker, P. (2014) Using Corpora to Analyze Gender. London and New York, NY: Bloomsbury.
This is a comprehensive overview of how corpus linguistics may be used in investigating the
relationship between language and gender. Chapters 5 and 6 are particularly relevant for collocational
analyses.
Baker, J. P. and Levon, E. (2016) ‘“That’s what I call a man”: Representations of racialised and classed
masculinities in the UK print media’. Gender and Language, 10(1), pp. 106–139.
This paper shows how collocation analysis may be integrated with other approaches to the study
of language, gender, and sexuality.
Bogetić, K. (2013) ‘Normal straight gays: Lexical collocations and ideologies of masculinity in
personal ads of Serbian gay teenagers’. Gender & Language, 7(3), pp. 333–367.
This is an example of work in which collocation is absolutely central to the investigation. It is
worth noting that collocation in this paper is calculated as simple frequency.
Brezina, V., McEnery, T., and Wattam, S. (2015) ‘Collocations in context: A new perspective on
collocation networks’. International Journal of Corpus Linguistics, 20(2), pp. 139–173.
This offers an overview of the importance of collocation networks as well as an introduction to the
LancsBox tool.
Jaworska, S. and Hunt, S. (2017) ‘Differentiations and intersections: A corpus-assisted discourse study
of gender representations in the British press before, during and after the London Olympics 2012’.
Gender and Language, 11(3), pp. 336–364.
This is an example of how collocation analysis may be combined with frequency to investigate
constructions of gender. It also addresses issues of intersectionality.

Related topics
The South African news media and representations of sexuality; XML mark-up for nomination,
collocation, and frequency analysis of language of the law; analysing gendered discourses online;

584
 Collocation and mock politeness

determining the impact of gender stereotyping on patient feedback; affect in language, gender, and
sexuality research.

References
Baker, P. (2006) Using Corpora in Discourse Analysis. London: Continuum.
Baker, P. and McEnery, T. (2015) ‘Who benefits when discourse gets democratised? Analysing a
Twitter corpus around the British benefits street debate’. In: Baker, P. and McEnery, T. (eds.)
Corpora and Discourse Studies. Basingstoke: Palgrave Macmillan, pp. 244–265.
Baroni, M. and Bernardini, S. (2004) ‘BootCaT: Bootstrapping corpora and terms from the web’.
Proceedings of LREC. Available at: http://sslmit.unibo.it/~baroni/publications/lrec2004/bootcat
_lrec_(2004)pdf
Biber, D., Conrad, S., and Reppen, R. (1998) Corpus Linguistics: Investigating Language Structure
and Use. Cambridge: Cambridge University Press.
Bogetić, K. (2013) ‘Normal straight gays: Lexical collocations and ideologies of masculinity in
personal ads of Serbian gay teenagers’. Gender & Language, 7(3), pp. 333–367.
Bowes, A. and Katz, A. (2011) ‘When sarcasm stings’. Discourse Processes, 48(4), pp. 215–236.
Brezina, V., McEnery, T., and Wattam, S. (2015) ‘Collocations in context: A new perspective on
collocation networks’. International Journal of Corpus Linguistics, 20(2), pp. 139–173.
Buss, D. M. (1989) ‘Conflict between the sexes: Strategic interference and the evocation of anger and
upset’. Journal of Personality and Social Psychology, 56(5), p. 735.
Cameron, D. (2003) ‘Narrow church?’. Critical Quarterly, 45(4), pp. 109–112.
Colston, H. L. and Lee, S. Y. (2004) ‘Gender differences in verbal irony use’. Metaphor and Symbol,
19(4), pp. 289–306.
Creusere, M. A. (1999) ‘Theories of adults’ understanding and use of irony and sarcasm: Applications
to and evidence from research with children’. Developmental Review, 19(2), pp. 213–262.
Culpeper, J. (2011) Impoliteness: Using Language to Cause Offence. Cambridge: Cambridge
University Press.
Dress, M. L., Kreuz, R. J., Link, K. E., and Caucci, G. M. (2008) ‘Regional variation in the use of
sarcasm’. Journal of Language and Social Psychology, 27(1), p. 71.
Duguid, A. and Partington, A. (2018) ‘Using corpus linguistics to investigate absence/s: You don’t
know what you’re missing. Or do you?’. In: Taylor, C. and Marchi, A. (eds.) Corpus Approaches
to Discourse: A Critical Review. Abingdon: Routledge, pp. 38–59.
Firth, J. R. (1957) Papers in Linguistics, 1934–1951. Oxford: Oxford University Press.
Glick, P. and Fiske, S. T. (1997) ‘Hostile and benevolent sexism’. Psychology of Women Quarterly,
21(1), pp. 119–135.
Jakubíček, M., Kilgarriff, A., Kovář, V., Rychlý, P., and Suchomel, V. (2013, July). ‘The TenTen
corpus family’. Paper presented at Corpus Linguistics Conference 2013, Lancaster.
Katz, A., Piasecka, I., and Toplak, M. (2001) ‘Comprehending the sarcastic comments of males and
females’. Poster presented at the 42nd Annual Meeting of the Psychonomic Society, Orlando, FL.
Katz, A. N., Blasko, D. G., and Kazmerski, V. A. (2004) ‘Saying what you don’t mean: Social
influences on sarcastic language processing’. Current Directions in Psychological Science, 13,
pp. 186–189.
Kilgarriff, A., Baisa, V., Bušta, J., Jakubícek, M., Kovář, V., Michelfeit, J., Rychlý, P., and Suchomel,
V. (2014) ‘The sketch engine: Ten years on’. Lexicography, 1(1), pp. 7–36.
Leech, G. (1992) ‘Corpora and theories of linguistic performance’. In: Svartvik, J. (ed.) Directions in
Corpus Linguistics: Proceedings of Nobel Symposium 82. Berlin: Mouton de Gruyter, pp. 105–122.
Louw, B. (1993) ‘Irony in the text or insincerity in the writer? The diagnostic potential of semantic
prosodies’. In: Baker, M., Thompson, G., and Tognini-Bonelli, E. (eds.) Text and Technology. In
Honour of John Sinclair. Amsterdam: John Benjamins, pp. 240–251.
McEnery, T. and Hardie, A. (2012) Corpus Linguistics: Method, Theory and Practice. Cambridge:
CUP.

585
Charlotte Taylor

Milanowicz, A. (2013) ‘Irony as a means of perception through communication channels. Emotions,

attitude and IQ related to irony across gender’. Psychology of Language and Communication, 17,
pp. 115–132.
Mumsnet (2015) ‘About us’. http://www.mumsnet.com/info/aboutus (page last updated on 29 January
2015, Accessed: 12th May 2015).
Rockwell, P. and Theriot, E. M. (2001) ‘Culture, gender, and gender mix in encoders of sarcasm: A
self-assessment analysis’. Communication Research Reports, 18(1), pp. 44–52.
Rychly, P. and Kilgarriff, A. (2007) ‘An efficient algorithm for building a distributional thesaurus’.
Proceedings of ACL, Prague, Czech Republic. Available at: http://www.kilgarriff.co.uk/Publica
tions/2007-RychlyKilg-ACL-thesauruses.pdf (Accessed: 12th May 2015).
Sinclair, J. M. (2004) Trust the Text: Language, Corpus and Discourse. London and New York, NY:
Routledge.
Stubbs, M. (2001) Words and Phrases. Corpus Studies of Lexical Semantics. Oxford, Blackwell.
Taylor, C. (2016) Mock Politeness in English and Italian: A Corpus-Assisted Metalanguage Analysis.
Amsterdam: John Benjamins.
Watts, R., Ide, S., and Ehrlich, K. (1992) ‘Introduction’. In: Watts, R., Ide, S., and Ehrlich, K. (eds.)
Politeness in Language: Study in its History, Theory and Practice. Berlin: Mouton de Gruyter, pp.
1–17.

586
 39
The South African news media and
representations of sexuality
Sally Hunt

Introduction
In this chapter I demonstrate the application of a combination of corpus linguistics and
critical discourse analysis to a corpus of the most widely read newspaper in South Africa:
the Daily Sun. I analyse the data in terms of the representation of sexuality, in the context
of the newspaper’s inclusion of words from languages other than English. I discuss some of
the challenges of working with this multilingual data set, and make some comments on the
benefits of using corpus methods in critical discourse analysis (CDA) to analyse the repre-
sentation of sexuality in media texts.

Sexuality in South Africa

South Africa’s constitution is noteworthy for its strong stance on equality in terms of sex-
uality, amongst other aspects of identity (The Constitution of South Africa 1996: n.p.).
However, the statistics on rape and gender-based violence in the country and the practice of
‘corrective rape‘, that is, rape to ‘cure‘ lesbians, suggest that the lived reality of many South
Africans is far removed from the hopeful statements of the constitution (Boonzaier and
Zway 2015). It has been claimed, by public figures from presidents to pastors, that homo-
sexuality is ‘un-African’, which not only reproduces ‘a logic of essentialised racial identi-
ties’, quite common in South Africa (Epprecht 2012, cited in Bhana 2015: 135), but also
bolsters heteronormativity. Indeed, as recently as 2008, between 82 and 85 per cent of South
Africans considered sex between people of the same sex to be ‘always wrong’ (Roberts and
Reddy 2008). This chapter reports on research into the representation of sexualities in the
most widely read newspaper in the country, in an attempt to explore the attitudes reflected
and perpetuated in the local media.

The representation of identity in the media

In this context, the media provide a useful reflection of the views of ordinary citizens, the
imagined ‘ideal readers’, and the dominant commonsense beliefs about sexuality, especially
non-heterosexual sexualities. The significance of media constructions of sexuality also lies in
587
Sally Hunt

their potential to influence ideological ebbs and flows, to thwart efforts to move towards equal-
ity through omission or negative coverage, and/or to support and thereby normalise them.
The media form one of several categories of text producers who possess symbolic capital
(Bourdieu 1984) and constitute the symbolic elite. Their ideas and actions and texts affect
more people and therefore potentially have a disproportionate impact on the general public:

[j]ournalists, writers, professors, and other symbolic elites thus have a primary role in
setting the agenda, and hence have considerable influence in defining the terms and the
margins of consent and dissent for public debate, in formulating the problems people
speak and think about, and especially in controlling the changing systems of norms and
values by which … events are evaluated.
(Van Dijk 1993: 47)

In the context of this chapter, how sexuality is represented in the South African media both
limits and defines the ways in which this aspect of identity is conceived in the public eye,
and frames and shapes debates concerning the position of various sexualities in society.

Current contributions and research

Research into the representation of various aspects of identity in the media increasingly
uses corpus methods to support a critical discourse analytic approach. In 2010, Johnson and
Milani edited a book on ideological representation in the media. Two subsequent journal
special issues have also stimulated work in this area: in 2011 Jaworska and Larivee edited an
edition of the Journal of Pragmatics, ‘assessing the bias’ towards women in the media, and
Milani drew together research on ‘Language, gender and sexuality in South Africa’ in SPIL
Plus 46 (2015) with a focus on a range of media from T-shirts to advice columns. In other
contexts, Caldas-Coulthard and Moon (2010) examined gendered patterns of representation
in the British press, while Santaemilia and Maruenda (2014) analysed the representation of
women and gender-based violence in the Spanish news media. In the South African context,
Hunt and Hubbard (2015) contrast the representations of sex work(ers) in two local news-
papers. Baker and Levon (2016) consider the intersectional representations of race, class,
and masculinity in the UK print media, and Bartley and Hidlgo-Tenorio (2015) examine the
Irish press in terms of their representation of homosexuality. Many of these studies incor-
porate corpus linguistics methods to analyse corpora of media texts. These, and many other
contributions, indicate an academic concern with social identity that often rests on personal
and political convictions regarding the perpetuation of discriminatory ideas and practices
via the mass media. This chapter, in showing how corpus linguistics can be used to inter-
rogate large bodies of texts for patterns of ideological representation, reveals how various
non-heterosexual sexualities are represented in the most widely read South African daily
newspaper, and thereby adds to the efforts of critical analysts to reduce inequality.

The case study

In this case study, I answer the following questions:

• What methodological adaptations to common practice in the combined use of corpus

linguistics and critical discourse analysis are needed to analyse sexuality in this context
and in this corpus in particular?

588
 Sexuality in the South African news media

• How is sexuality represented in this corpus of news reporting? Are any identities related
to sexuality included (lesbian, gay, bisexual, transsexual, queer, intersex, asexual and
other) and, if so, how (positively or negatively? using which terms?)?
• What do these patterns of representation reflect about prominent views on sexuality in
South Africa?

Methodology
To draw out patterns of representation for each of these aspects, I make use of an increas-
ingly popular combination of methods: CDA and corpus linguistics. The first of these,
CDA, has been described as ‘discourse analysis with attitude’ (Van Dijk 2001: 96), which
quite aptly captures the overtly political nature of this kind of research. CDA examines the
language choices made by a speaker or a writer and interprets the world view so implied.
CDA offers emancipatory potential and analysts strive to reveal any inequality in the
representations of relatively powerless groups in society. The second, corpus linguistics,
is the analysis of large bodies of text (‘corpora’) using quantitative methods of various
kinds. Available software offers a range of ways of processing the data to reveal patterns
in usage, which in turn reveal ideological emphases and assessments. How these two
methods were adapted to the specific data will be expanded upon in the description of the
analysis below.
The data for the case study come from the most widely read national newspaper in South
Africa, the Daily Sun, which has an estimated daily readership of around 5 million people
(Media 24 2017), just under 10 per cent of the population (Census 2015), and 14.2 per
cent of adults (SAARF 2017). The tabloid is published in English, although the readership
claims many different languages as mother tongues. This last feature is part of what makes
the publication so interesting to study, as the multilingual readership means that words from
languages other than English are used.
The linguistic context in South Africa is diverse, both in terms of the language families
represented and individual repertoires, with multilingualism being the norm. The country
has approximately 30 languages, from several language families. English is widely used as
a lingua franca, with approximately 40 per cent of South Africans reporting English as a first
or second language (Posel and Zeller 2016). In this context, the use of English in the Daily
Sun means it can reach a sizeable proportion of the population while the variety of English
used reflects the mother tongues of the readership in terms of the borrowings evident in the
articles.
The frequency of words originating from African languages is higher in the Daily Sun
corpus than in some other corpora of South African English. In the corpus of 132,603 tokens
and 12,289 unique word forms, the words from Indigenous languages total just under 250
types, reflected in 2,724 tokens, giving words from African languages just over a 2 per cent
share. This is significantly higher than that in SAE11, a general corpus of South African
English built on the Brown-LOB sampling frame (Hunt and Bowker 2015), which has 0.2
per cent African lexis calculated on the same basis. The varied linguistic repertoire of the
Daily Sun readership may account for the ten-fold usage of African words in the media
corpus, when compared to a corpus of South African English from multiple, typically for-
mal, genres. It could be argued that the Daily Sun corpus reflects less formal everyday
usage amongst multilingual South Africans and reveals how they construe sexual identities.
Corpus analysis of terms from African languages referring to sexuality allows the explora-
tion of how these are represented as part of African identity, or as outside of it.

589
Sally Hunt

The Daily Sun data were sourced from NewsBank, a limited access online news reposi-
tory. I searched the repository using a list of 27 nouns referring to sexual identity across the
entire 11 years of publication available online. In descending order of frequency, the most
productive terms found in the publication were: gay, lesbian, gays, lesbians, homosexuality,
homosexual, homosexuals, moffie (gay man), stabane, isitabane, setabane (all: gay person).
Some of the terms indexing sexuality also denote gender, such as moffie (Afrikaans, a male
homosexual) and lesbian, but most do not, at least not overtly. I included in the list of search
terms both standard and slang terms, and those regarded as offensive, although formal terms
are more likely to be found in a genre like news reporting. Of these, the more formal terms
gay*, lesbian* or homosexual*’1 were by far the most productive, accounting for 349 arti-
cles over 11 years, versus 36 from all the other terms combined. Each report containing at
least one of the search terms was downloaded and converted to txt format, stripped of all its
metadata, such as date of publication, author, and the section of the newspaper, which was
saved separately. The output was a topic specific corpus (Gabrielatos 2007), which are often
used in critical discourse studies to focus on the representation of a minority within general
reporting (ibid.).
Corpora are seen as more reliable if the criteria used to build a corpus are external to the
data e.g. the total output of a newspaper. Analysing the usage of particular terms when those
same terms were used to select articles to build the corpus could reduce objectivity. This
potential for circularity is observed by McEnery, Xiao, and Tono (2006: 14):

A corpus is typically designed to study linguistic distributions. If the distribution of

linguistic features is predetermined when the corpus is designed, there is no point in
analysing such a corpus to discover naturally occurring linguistic feature distributions.
The corpus has been skewed by design.

This would be true of this research if the corpus had indeed been built on the basis of a
particular linguistic feature, such as a grammatical construction, or to measure the overall
frequency of particular words in the context of the newspaper’s entire output. But in this
case, with data from a limited access repository like Newsbank, downloading all articles
from 11 years of publication is neither allowed nor feasible, and so, following Baker and
McEnery (2005) in their analysis of the discourses surrounding refugees and asylum seekers
in the UK press, the articles were chosen simply as a subset of the articles in the newspaper,
and that subset is defined by the subject matter. Collecting all the articles which explicitly
reference sexuality as an identity label is, I would argue, a legitimate way to explore the
discourses of (homo)sexuality in the Daily Sun, as long as one is aware of its limitations.
Having collected the articles, I used corpus methods to reveal patterns suggestive of
particular ideological stances, as I will explain below.

Features and application of corpus linguistic analysis

Most corpus linguistics software has certain core features that exploit the advantages of
computer-based language processing: speed and accuracy, as well as the ability to deal with
enormous amounts of data, relative to what a human could process reliably. Corpus tools
may be used to identify patterns in the features common in traditional micro-level CDA,
such as over-wording, positive and negative expressive values, and experiential represen-
tation (e.g. Fairclough 2001). In this case study I use AntConc 3.4.3w (Anthony 2014), a
freeware concordancing programme, and focus on its facility to produce concordance lines,

590
 Sexuality in the South African news media

as well as the computation of collocates, words which co-occur to a statistically significant

extent, and show how they are useful for this critical analysis.

Concordance lines
The concordance facility is particularly useful for the analysis of ideological aspects of
a corpus and enables the researcher to extract every example of a particular word in the
corpus in the linguistic context in which it occurs, and then sort these in ways which
reveal patterns of representation. The main aim of this study is to see which discourses
about sexuality are dominant in the South African news media. Consequently, examin-
ing the context in which various terms that index these aspects of identity occur helps
to identify common semantic prosodies, in other words, the ways in which ‘speakers
establish and maintain connotational or evaluative harmony within a stretch of discourse
by co-selecting items of a consistent evaluative/attitudinal force’ (Morley and Partington
2009: 143–144, emphasis in original). This means that language users typically maintain a
consistent attitude to their subject matter in a text by choosing words which carry roughly
the same degree of positivity or negativity. These patterns of connotation and evaluation
can be traced by searching for the linguistic context of the terms in focus, such as gay* or
moffie* in this study. Through the examination of the concordance lines generated in this
way, patterns of prosody emerge from the data and reveal the discourses underlying the
language, the commonsense assumptions that the publication builds on, and perpetuates.
Ultimately, these evaluative patterns of meaning can, through reiteration, prime the reader
to view identity in certain ways (Hoey 2005).
What this concordance function relies upon is entering a search term which the software
finds and then lists each occurrence, or token. The multilingual list of terms discussed above
was used to interrogate the representation of sexuality in the corpus. It is significant that
heterosexuality is not referenced explicitly, apart from assumptions about husbands having
wives and vice versa, which supports heteronormativity and emphasises the marked, even
deviant, status of various other forms of sexuality. A corpus that includes multiple languages
proves challenging to search, and the morphological differences between English and the
local African languages add to this.
Given the origins of corpus linguistics and its software, there is an obvious lexical
emphasis, which Morley and Partington (2009: 143) call the ‘curse of the concordance
node’. Most software has been developed for languages which use suffixes rather than pre-
fixes on verbs and nouns, and works particularly well with languages like English due to the
fact that analytic languages typically locate units of meaning in separate words. However,
this approach to analysis was complicated by the nature of my data, and required some
adaptation. The main issue is that English and Southern Bantu African languages, such as
those spoken in South Africa, vary in terms of how the plural is constructed, and, given
my emphasis on identity terms, which are typically nouns, this is important. So whereas in
an English monolingual corpus a search for lesbian and its plural lesbians could easily be
achieved simultaneously using a wildcard: lesbian*, this is not quite as straightforward in
this case, with multiple borrowings from multiple African languages. First, a different plural
prefix is used according to the noun class of the word, and there are seven to ten plural forms
per language. Second, the set of prefixes varies from language to language. In addition, the
data show that the borrowing of these terms by English is still in progress in that two meth-
ods of forming the plural are used in parallel in this corpus: (1) applying the morphological
rules of an African language and (2) using English suffixes. For example, the isiZulu noun

591
Sally Hunt

isitabane (‘gay’), or more commonly just stabane, and its plural izitabane, are equivalent
to the SePedi/Northern Sotho word setabane, and its plural ditabane, although sometimes
the English suffix method of forming the plural is used instead of the original plural prefix:
stabanes.
The solution for this study is to create a list including all the singular forms, plural forms,
and alternate spellings and subsume them under one lemma, or headword. For example, all
the forms for stabane listed above, plus additional terms from other languages with equiva-
lent meanings, are grouped together under one ‘label’, or lemma. I adopted a two-pronged
approach to locate the types to include in this list. First, I used wildcards to search the data
for the stems of each known sexuality term. To continue with the stabane example, entering
*tabane as the search term returns both singular and plural forms, from a variety of lan-
guages. Second, I searched using common singular and plural prefixes, such as izi* or aba*,
to check other borrowed terms for potential reference to sexuality, reasoning that as this is
a topic-specific corpus, there was a reasonable chance of finding additional search terms in
this way. For this project, an existing lemma list compiled from the British National Corpus
(Anthony n.d.) was adapted, adding all the newly found forms under the appropriate lemma,
as in the following example:

gay->[gay] [gays] [gaye] [gayer] [gayest] [stabane] [isitabane] [izitabane] [stabanes]

[setabane] [ditabane]

The adapted lemma list was loaded into the software so that subsequent searches supplied
statistics and data for all the equivalent terms together.
A further decision was under which lemma to place the alternate forms. In the case of sta-
bane, the word originally meant ‘intersex’, suggesting that referents possessed both female
and male physical characteristics, but it is currently used to refer to gay people (Swarr
2012). This shows how language change can play havoc with the complex relationships
between related terms concerning sexuality, and also illustrates how complicated it can be
to map types to lemmas, which implies an equivalence in translation that is not always justi-
fied. The decision to classify stabane and related terms under the lemma GAY, as opposed
to HOMOSEXUAL, was ultimately based on the data, on translations by users of the term
in the corpus, such as the following:

‘People shout “stabane”, gay and so on at me’

The English version was chosen as the lemma in each case as these are the most frequent
tokens for each concept in this corpus, and because the newspaper is written in English.
Despite extensive searches for terms, the frequencies for most terms from African languages
were low, with the English terms used just over 99 per cent of the time, which made the
lemmatisation of the data even more important.

Collocates
A second useful feature of most corpus linguistics software is the collocation tool, which
calculates which words co-occur with the search term to a statistically significant extent.
The reasoning behind using this in a critical discourse analysis is that words derive at least
some of their meaning and their associations from the words around them, and especially
from those words with which they frequently co-occur (Baker 2006). The tool allows the

592
 Sexuality in the South African news media

researcher to explore not only the ways in which words group together, but, by looking at
their context and frequency, also the significance and meaning of patterns of collocation
(Bambrook, Mason, and Krishnamurthy 2013). For example, if the words killer or killers
occur more frequently in the environment of the word lesbian than do other words, and than
they do without the word lesbian, as is the case in this corpus, then the reader of the text
will gradually be ‘primed’ to associate lesbians with killer(s), rather than with more posi-
tive words (Hoey 2005). There is a variety of statistical measures of this relationship, but in
this study the mutual information (MI) measure of collocational strength was used, as this
favours open-class words such as nouns and verbs, which are usually the emphasis in this
kind of research. I investigated words which scored at least 3.84 on the MI measure, as this
indicates at least a 95 per cent confidence level, although most of the collocates scored much
higher than this. AntConc searched for collocate candidates in the words appearing from
four to the left of the search term to four to the right, where the mutual ‘colouring’ of mean-
ing between words is strongest, with a minimum frequency of three. Identifying the words
which most frequently and most significantly co-occur with the terms for various sexual
identities in this corpus reveals the discourses about sexuality which have been incorporated
into the texts as common-sense assumptions.
The inclusion of corpus linguistics in a critical analysis does not involve mimicking the
steps and elements of a traditional critical discourse analysis; instead the researcher uses
the software tools to explore patterns which are likely to reveal the ideological flavour of
the data, such as the central preoccupations of the texts via frequency, the concepts which
tend to be associated with each other through the analysis of collocates, and their positive or
negative values via concordance lines. The following section explores how both concord-
ance lines and collocates have been utilised in the analysis of the Daily Sun data.

The analysis of the corpus

The relatively small number of news reports reflecting non-traditional sexualities, even in
this topic-specific corpus, is potentially problematic for analysis. In 11 years of reporting,
GAY, the most frequent of the sexuality lemmas, is only found 1,057 times, or 79.7 per
10,000 words, in the specialised corpus. While this might seem quite common, it represents
an average of just 96 tokens a year in a daily publication, and one article alone can account
for several tokens. Despite the small number of examples to work with, there are clear pat-
terns in terms of concordance lines and collocation patterns to interpret. In the discussion
which follows I will explore the dominant semantic domains associated with each of the
most frequent sexualities mentioned, supported with examples from the concordance lines
and evidence from the collocation scores.
The terms lesbian, gay, bisexual, transgender, and intersex tend to collocate quite
strongly with each other because of the tendency for these terms to occur as a string of five
or six words, in the same order each time. Gay co-occurs with lesbian relatively frequently
(92 times out of the total frequency of lesbian of 316, or just under a third of tokens), but it
also occurs frequently elsewhere, which accounts for the fact that the MI score is not espe-
cially high at 5.4. Transgender and bisexual both score over eight, because although they
only occur 23 times each in the span around lesbian, they seldom occur elsewhere. This sup-
ports the interpretation that these identities are often seen to belong together as a collection
of ‘non-traditional’ genders and sexual orientations, and it could be argued that this serves to
‘other’ them. In the case of the fairly formulaic n-gram lesbian gay bisexual (transgender)
(queer) (intersex) (where bracketed items are not always included), while the commonly

593
Sally Hunt

used string has expanded over time in the interests of inclusivity, it also creates the impres-
sion of being a ‘catch-all’ category for those who do not adhere to dominant expectations
concerning sexuality or gender. In addition, the use of ‘one label for all’ serves to obscure
differences between the component groups, when there are considerable differences, and
differences of opinion and politics, between people claiming these identities.

HOMOSEXUAL
This lemma, which includes the type homosexuality, appears 158 times and seems to be
neutral as to gender, in contrast with either GAY or LESBIAN, and in fact tends not to refer
to individuals but more commonly to modify sexual acts, or laws. The strongest collocates
of the lemma (MI ≥ 6.1) are listed in Table 39.1, presented from highest to lowest MI.
The vast majority of collocates fall into two broad semantic domains: the legal position
on homosexuality and gay relationships in various countries (such as, jailed, rights, illegal,
law, country), and moral stances on homosexuality, especially those of religious leaders and
communities, chiefly from the Christian church (prejudice, discrimination, anti, god).
The concordance lines confirm these twin emphases with the three most frequent nouns
following homosexual being act/s, marriage, and right/s, and the most common predicates
of homosexuality are illegal, unnatural, foreign, taboo, (not) holy, and punishable, suggest-
ing strong disapproval. However, the publication lacks a consistent stance while addressing
these two aspects. The news reports vary in terms of whether they are against homosexuality

Table 39.1 Twenty strongest collocates for HOMOSEXUAL by MI:

4L to 4R, minimum frequency 3

Collocate MI score

outlawed 11.45
illegal 11.07
acts 11.00
punishable 10.77
engaging 10.77
jailed 10.31
prejudice 9.65
activists 8.77
discrimination 8.31
orientation 8.27
uganda 8.15
rights 8.08
anti 8.07
marriages 7.93
law 7.79
against 7.65
god 7.56
country 7.33
should 7.08
outlawed 6.43
illegal 6.27
acts 6.12

594
 Sexuality in the South African news media

Table 39.2 Selected concordance lines for HOMOSEXUAL

marriages. Dube said his church is against homosexual marriages because they
won’t tolerate homophobes (people who treat homosexuals badly) in Mzansi. Gigaba was
warned that the punishment for engaging in homosexual acts is death by stoning, burning
God made us who we are God did make homosexuals and they are here to stay!
Mugabe has for decades spoken out against homosexuality and has again threatened
Zulu Royal House must learn to accept that homosexuality exists in our country. Who are

or in favour of legalisation (in countries where it is illegal). There are, for instance, religious
pieces advocating tolerance and understanding, and commentaries that oppose legalisa-
tion, and vice versa. There are also some carefully neutral reports, particularly concerning
the legal situation in other countries. In several cases, only the headline reveals any kind
of evaluative dimension through the use of exclamation marks. The concordance lines in
Table 39.2 reflect the diversity of views expressed in the data.

GAY
While GAY is often assumed to refer generically to both men and women, it is clearly used
in this corpus to refer mostly to men. Of the 150 human reference tokens that pattern with
GAY, exactly half (75) refer to males, as in gay man (25) or gay men (29), and there are 21
instances of he is gay. While the rest are unspecified as to sex, such as gay couple (45), it is
reasonable to assume that at least some of these are male.
Lesbian, of course, refers only to women, meaning that in the much repeated list ‘lesbian,
gay, bisexual …’ gay once again refers exclusively to men. One of the strongest collocates
of the lemma GAY is the identity label lesbians (MI 6.6), further supporting the idea that
the word gay is used in these contexts to refer to men only. This throws into question who
is referred to in gay rights (29) and other supposedly gender neutral phrases such as gay
community (20) and gay marriage (20); whether GAY is really used as a true generic. One
explanation for this could be that the term is following the lead of man, used to refer to
males only but also often claimed as a generic. It could be argued that in both cases the
restricted use undermines the potential for general reference. Other strong collocates of
GAY include further identity terms, such as trans (7.2), intersex (7.2), transgender (7), and
bisexual (MI 7), although the first three in this list refer to sex, rather than to sexuality, sug-
gesting a tendency in the reporting to associate these identities with each other, as ‘others’.
The collocates of a search term often cluster into semantic domains, with groups of
words relating to particular areas of experience, allowing the researcher to identify par-
ticular discourses and ideological stances in the data. In this data, tokens referring to gay
pride parades and parties are common (waving, marching, banner, thousands, parade, flag,
proudly, with MI scores ranging from 7.2 to 6), and in context these reflect, for the most
part, a positive view of gay people. But there are also those which reflect the condemnation
of gay lifestyles expressed by parts of society. Bashing (MI 6.9), penalty, and outlaw (both
MI 6.8) refer to the impact of social disapproval, while disgrace (MI 6.2) and pastors (MI 6)
reflect the fact that anti-gay sentiments are often couched in religious terms. Okay (MI 7.2),
while apparently a positive term, appears most frequently in phrases like (being) gay is not
ok(ay) (frequency 12), which underlines the importance of checking the usage of collocates
in the concordance, and avoiding assumptions that their denotational meaning is being used
without qualification. The diversity with respect to positive and negative evaluative terms

595
Sally Hunt

in the collocates for GAY reiterates the diversity in stance in the reporting on matters of
sexuality.
The isiZulu word isitabane, or more commonly just stabane, and the SePedi word seta-
bane, mentioned above, all originally meant intersex (Swarr 2012), but are now used as
synonyms for the English gay, and so are included in the lemma GAY. These terms are
frequently used to insult (ibid.), rather than simply to refer neutrally to people or sexuality,
as the representative concordance lines in Table 39.3 show.
GAY also patterns with other words to form compounds such as anti-gay (27). Just under
half of the instances of anti-gay (11) refer to legislation in other countries (Table 39.4).
This is supported by the strongest collocates with anti-gay: laws (MI 9.3), bill (MI 9.0),
and Uganda (MI 8.3). What is said about these laws is significant: they are described as
wrong and unfriendly, and there are calls for them to be slammed and condemned. Because
they are only used once each with anti-gay, these words with strong negative sentiment do
not show up in the list of collocates, but nonetheless build up a prosody of opposition to the
anti-gay laws being discussed. Examining the text surrounding important terms for indi-
vidual tokens which together suggest a semantic domain, with or without evaluative stance,
is a productive strategy in this kind of analysis.
Apart from the references to laws, which constitute 11 of the 27 anti-gay concordance
lines, 9 uses refer to activism and negative behaviour directed at gay people, or the people
engaged in that behaviour (Table 39.5).
Here anti-gay behaviour is referred to as a hate campaign, as a movement which is bub-
bling up, involving gestures and insults by protesters and activists on radio and online.

Table 39.3 Selected concordance lines for *tabane

of the road laughed at him and said Isitabane esimbi kanje, meaning he’s an ugly gay
eyes he finds hard to resist. SePedi confusion “Setabane” refers to a gay person, rather than a
Senyaka come off stage angrily calling Ayanda a stabane and slapping him. We didn’t know what
verbal abuse has been increasing. “People shout “stabane”, gay and so on at me. “I should

Table 39.4 Selected concordance lines for anti-gay: legal aspect

to cut aid to governments that pass anti-gay laws. In recent months, Nigeria outlawed
after the East African nation strengthened its anti-gay legislation last week. A foreign ministry
a bold and unfriendly step in its anti-gay laws. The stance has seen the
government should have condemned the anti-gay bill recently signed by Ugandan President

Table 39.5 Selected concordance lines for anti-gay: anti-gay ‘activism’

waiting for, he said. In Nairobi, Kenya, anti-gay activists have protested against legalisation
law, said Heterosexual Pride Day is not anti-gay, but a protest against the privileges
for comment. Police say Morales yelled anti-gay insults before he shot Carson in
Joe allegedly received death threats from anti-gay listeners. However, the station has rallied
transsexuals in Roman Catholic Lithuania. Anti-gay protesters make gestures at police during

596
 Sexuality in the South African news media

This representation is difficult to characterise, as it seems to use both positive and negative
associations. While most of these words are negative, the inclusion of terms associated
with the anti-apartheid struggle, like movement, protesters, and activists, links the reported
behaviour to a just cause, one that is historically important and still salient in South African
society. This link is therefore at odds with the condemnation of the legislation criminalising
homosexuality discussed above.
On the other hand, gay patterns strongly with elements of the human rights discourse
mentioned above. There are 29 instances of gay rights, of which 10 are gay rights activist(s)
and 20 instances of gay community. Verbs that pattern with gay rights as their object include
fight for, recognise, promote, and respect (four), but gay rights activist/s are shown to be
tortured and killed, objects of murdering and killing. Once again, there is an ambivalence
reflected; this time in society itself: a strong movement for gay rights, but the potential for
tremendous personal danger for activists.

LESBIAN
Danger for gay people is particularly apparent for lesbians. LESBIAN shows similarities
with GAY in that the strongest collocates (all MI ≥ 7.2) include other identity terms like
transgender, bisexuals, gays, and black, as well as words which prove to be used positively
in the concordance lines, such as waving, happily, marched, lovers, openly, and equality.
But again an opposing semantic cluster emerges in contrast to the positive collocates of
social agency and happiness: 15 collocates refer to harm or violence, especially murder,
with MI scores greater than or equal to 7.1. These include, killing*, killers, raping, killer,
faced, murdered, raped, killed, buried, crimes, and fear.
The most frequent 4-gram including LESBIAN is for being a lesbian, occurring 14
times, which is preceded by killed (6), murdered (3), hack(ed) to death, died, beat (her),
taunted, and insulted. Not only does this reflect the extremely harsh treatment of lesbians
but also the attribution of responsibility that is tied to the victim’s identity. The identity of
the perpetrator as a violent homophobe is not given anywhere in the articles as a reason for
their behaviour, but rather the sexuality of the woman.
The contrast with the patterns found for GAY is striking. While there is social approbation
and bashing associated with the GAY collocates, the majority of the negative words associated
with LESBIAN are very much more violent, with seven of the strongest collocates referring to
the murder of the woman concerned. This, together with the fact that GAY is more often used
as a modifier (as in gay men or gay rights) than is LESBIAN (412 times versus 124), suggests
that GAY is used mostly to refer to a group or a movement, especially in its English form,
whereas LESBIAN typically refers to the lived experience of individual gay women.
For lesbians in South Africa, particularly young Black lesbians, there is also an ever-
present fear of rape (Boonzaier and Zway 2015). The strongest collocate for rape in the data
is corrective (MI 11.2), which refers to the belief that gay women can, and should, be ‘cured’
of their homosexuality by rape. This is reflected in the data, where the practice is negatively
evaluated, and the judgemental label ‘corrective’ is sometimes flagged as problematic with
scare quotes or another distancing mechanism, such as being preceded by the disclaimer
‘so-called’ (Table 39.6).
This tendency to represent lesbians in terms of the physical violence they may suffer as
a result of extreme moral disapproval, tends to obscure other, more positive and agentive,
elements of their lives and, as Boonzaier and Zway (2015) argue, thrusts gay women into
complete and permanent victimhood.

597
Sally Hunt

Table 39.6 Selected concordance lines for corrective rape

an organisation which helps lesbians fight corrective rape and hate crimes. Now she is more
harassment or criminal violence – gay bashing, corrective rape , and murder – directed at gay or
lesbians still live in fear of so-called corrective rape , mindless violence, and even murder
Atteridgeville, west of Pretoria, was a victim of corrective rape . The Department of Justice condemn
world has joined the battle against so-called corrective rape . This practice is seen by many in
the need to address the practice known as corrective rape , which is a violent and criminal

MOFFIE
Lemmas which are not particularly frequent represent a problem statistically, although
they can nonetheless contribute to the analysis of the data. The Afrikaans lemma MOFFIE,
which refers to an effeminate gay man, is just such a case. While it is not frequent enough to
generate statistically significant collocates, the concordance lines reveal some commonality
with other terms: in 80 per cent of the tokens, moffie is used as an insulting term of address,
like stabane, but this is often combined with violence, as is frequent in the representation
of lesbians. The linkage between gender and power is further reflected in a ‘moffie’s’ loss of
hegemonic masculine power when a man reportedly says he ‘wouldn’t be ordered around
by a moffie’.
These connections draw out the intersectional nature of identity, especially in relation to
gender, and its representation in this corpus. It is difficult to separate notions of sexuality
from elements of gender and the data reflect this. They also reflect the subordinate status
of females, and those associated with ‘femaleness’ by virtue of the kind of sexuality they
perform.

Implications of the case study

Corpus linguistics has proved to be a productive method to explore the discourses of sexual-
ity in the Daily Sun. Incorporating terms from a variety of languages in a process of lemma-
tisation means that the ideological construction of sexuality can be revealed irrespective of
the languages used. The data reveal ideologically interesting patterns in the representation
of attitudes towards sexuality in South African media.
Simple frequency counts can be revealing, even in a specialised corpus. The tendency
for the frequencies of terms like stabane to be significantly lower than gay, for example,
suggests that English terms are being borrowed into African languages, rather than the
incorporation of the indigenous terms into English. This is possibly in parallel with the
incorporation into South African life of identities seen as Western in origin themselves, like
homosexuality. It may also be due to the fact that the more formal gay is used in ways that
are more neutral in evaluation, and without the strong negative value of moffie or stabane,
thus making it less problematic to use in the press than most of the borrowed words for
homosexuals, which tend to be insulting slang forms of address, rather than labels.
Statistical measures like those for collocation are very useful in CDA to suggest direc-
tions for finding patterns of meaning, such as the violent terms associated with lesbian*,
but analysts should always consult concordance lines, especially in small corpora, to find
semantic prosodies and to ensure that the co-occuring terms are being used with their
apparent meaning, such as the example of ok discussed above. Combining collocates with

598
 Sexuality in the South African news media

concordance lines reveals intersectionality in representation in this corpus, in the degree to

which the depiction of gay people in South Africa is gendered. First, gay itself tends to refer
to males, and so does not function as a true generic. Second, the collocations and concord-
ances lines that pattern with lesbian and moffie, and gay, and homosexual, show starkly dif-
ferent discourses about how these individuals are viewed, based largely on gender: moffie, a
term to refer to more effeminate men, is associated with violent treatment in similar fashion
to that which is meted out to women in general, and gay women in particular.
Analysis of the HOMOSEXUAL lemma in its concordance lines reveals the divergent
discourses in the corpus, which in turn reflect the multiplicity of worldviews in the country.
Words reflecting daily violence and hatred exist alongside excited fandom for openly gay
performers; pleas for tolerance and acceptance are countered by warnings of eternal damna-
tion. It seems that it is entirely possible for a country with the highest rape statistics in the
world to also have a remarkably progressive constitution. It is polarised and it is contradic-
tory, and it is reflected in the language of the Daily Sun.

Future directions
The case study discussed in this chapter has revealed some of the methodological chal-
lenges of working with a relatively small, multilingual corpus, collected from a limited
access database, and I have described the strategies I employed to try to extract ideologi-
cal patterns in the linguistic representation of non-heterosexual sexuality in the Daily Sun.
Examining a larger corpus which included the entire output of a publication would test the
success of these strategies. It would also be interesting to use a corpus with a higher propor-
tion of tokens from languages other than the main one, especially if this allowed for lemmas
that were not dominated by types from one language, as GAY was by gay in this research.
Finally, the methods described in this chapter could usefully be adapted to explore the pat-
terns in usage in a range of identity labels, such as those relating to gender or (dis)ability.
Despite the difficulties mentioned, I was nonetheless able to harness the available corpus
tools to reveal complex and ideologically significant patterns of representation in the data.

Note
1 An asterisk * in a cited word form indicates a ‘wildcard’, a placeholder for one or more letters that
may occur in that position. For example, lesbian* indicates a set of words which includes lesbian,
lesbians, and lesbianism.

Further reading
Baker, P. (2012) ‘Acceptable bias?: Using corpus linguistics methods with critical discourse analysis‘.
Critical Discourse Studies, 9(3), pp. 247–256.
This article not only sets out how corpus linguistic methods may be used to reduce the subjectivity
of critical discourse analysis, but also thoughtfully critiques this combination and the pitfalls of
interpretation.
Edwards, M. and Milani, T. (2014) ‘The everyday life of sexual politics: A feminist critical discourse
analysis of herbalist pamphlets in Johannesburg‘. Southern African Linguistics and Applied
Language Studies, 32(4), pp. 461–481.
Edwards and Milani demonstrate the combined use of CDA and corpus linguistics to analyse the
representation of gender and sexuality in pamphlets advertising the services of herbalists in South
Africa, a very different multilingual genre.

599
Sally Hunt

Schmidt, T. and Wörner, K. (eds.) (2012) Multilingual Corpora and Multilingual Corpus Analysis.
Amsterdam: John Benjamins.
Multilingual corpora are the focus of this book, which approaches their analysis mainly from a
language contact point of view, such as learner corpora and translation and interpreting.

Related topics
Non-binary approaches to gender and sexuality; sexuality as non-binary: a variationist perspective;
leadership language of Middle Eastern women; XML mark-up for nomination, collocation and
frequency analysis of language of the law; multimodal constructions of feminism

References
Anthony, L. (2014) AntConc (Version 3.4.3) [Computer Software]. Tokyo, Japan: Waseda University.
Available at: http://www.laurenceanthony.net/
Anthony, L. (n.d.) AntBNC. http://www.laurenceanthony.net/software/antconc/
Baker, P. (2006) Using Corpora in Discourse Analysis. London and New York, NY: Continuum.
Baker, P. and Levon, E. (2016) ‘“That’s what I call a man”: Representations of racialised and classed
masculinities in the UK print media’. Gender and Language, 10(1), pp. 106–139.
Baker, P. and McEnery, A. (2005) ‘A corpus-based approach to discourses of refugees and asylum
seekers in UN and newspaper texts’, Journal of Language and Politics, 4(2), pp. 197–226.
Barnbrook, G., Mason, O., and Krishnamurthy, R. (2013) Collocation: Applications and Implications.
UK: Springer.
Bartley, L. and Hidalgo-Tenorio, E. (2015) ‘Constructing perceptions of sexual orientation: A corpus-
based critical discourse analysis of transitivity in the Irish press’. Estudios Irlandeses, 10, pp. 14–34.
Bhana, D. (2015) Under Pressure: The Regulation of Sexualities in South African Secondary Schools.
South Africa: Modjaji Books.
Boonzaier, F. and Zway, M. (2015) ‘Young lesbian and bisexual women resisting discrimination and
negotiating safety: A photovoice study’. African Safety Promotion Journal, 13(1), pp. 7–29.
Bourdieu, P. (1984) Distinction: A Social Critique of the Judgement of Taste. Cambridge, MA:
Harvard University Press.
Caldas-Coulthard, C. R. and Moon, R. (2010) ‘“Curvy, hunky, kinky”: Using corpora as tools for
critical analysis’. Discourse & Society, 21(2), pp. 99–133.
Ethnologue (n.d.) Available at: https://www.ethnologue.com/country/ZA/languages (Accessed: 19th
May 2017).
Fairclough, N. (2001) Language and Power. London: Longman.
Gabrielatos, C. (2007) ‘Selecting query terms to build a specialised corpus from a restricted-access
database’. ICAME Journal, 31, pp. 5–43.
Hoey, M. (2005) Lexical Priming: A New Theory of Words and Language. New York, NY: Routledge.
Hunt, S. and Bowker, R. (2015) ‘New South African English – from braai to chisa nyama’. Corpus
Linguistics 2015, Lancaster, UK 21–24 July 2015.
Hunt, S. and Hubbard, B. (2015) ‘Sex work in the South African media: Representations of “deviant”
sexuality’. SPIL Plus, 46, pp. 19–43.
Jaworska, S. and Larrivée, P. (2011) ‘Women, power and the media. Assessing the bias’. Journal of
Pragmatics, 43, pp. 2477–2479.
Johnson, S. and Milani, T. (eds.) (2010) Language Ideologies and Media Discourse: Texts, Practices,
Politics. London: Continuum.
McEnery, T., Xiao, R., and Tono, Y. (2006) Corpus-based Language Studies: An Advanced Resource
Book. UK: Taylor & Francis.
Media 24 (2017) Available at: http://www.media24.com/newspapers/daily-sun (Accessed: 9th May
2017).

600
 Sexuality in the South African news media

Milani, T. (ed.) (2015) ‘Language, gender and sexuality in South Africa’. SPIL Plus, 46, pp. i–v.
Morley, J. and Partington, A. (2009) ‘A few frequently asked questions about semantic – or evaluative
– prosody’, International Journal of Corpus Linguistics, 14(2), pp. 139–158.
News24 (2014) Available at: http://www.news24.com/SouthAfrica/News/Daily-Sun-is-SAs-best-read
-newspaper-20141001 (Accessed: 19th May 2017).
Posel, D. and Zeller, J. (2016) ‘Language shift or increased bilingualism in South Africa: Evidence
from census data’. Journal of Multilingual and Multicultural Development, 37(4), pp. 357–370.
Roberts, B. and Reddy, V. (2008) ‘Pride and prejudice: Public attitudes towards homosexuality’.
HSRC Review, 6, pp. 9–11.
Robins, S. (2008) ‘Sexual politics and the Zuma rape trial’. Journal of Southern African Studies,
34(2), pp. 411–427.
SA Census (2015) Available at: http://www.saarf.co.za/amps-readership/2014/AMPS%20JUN%2014-
%20Readership%20Summary%20incl.%20Non-pay.pdf (Accessed: 19th May 2017).
Santaemilia, J. and Maruenda, S. (2014) ‘The linguistic representation of gender violence in (written)
media discourse: The term “woman” in Spanish contemporary newspapers’. Journal of Language
Aggression and Conflict, 2(2), pp. 249–273.
Swarr, A. L. (2012) Sex in Transition: Remaking Gender and Race in South Africa. New York, NY:
SUNY Press.
The Constitution of South Africa (1996) Available at: http://www.gov.za/DOCUMENTS/CONSTIT
UTION/constitution-republic-south-africa-1996-1 (Accessed: 18th May 2017).
Van Dijk, T. (2001) ‘Multidisciplinary CDA: A plea for diversity’. In: Wodak, R. and Meyer, M. (eds.)
Methods of Critical Discourse Analysis. London: SAGE, pp. 95–120.
Van Dijk, T. A. (1993) Elite Discourse and Racism. Newbury Park: Sage.
Hunt, S. and, R. (2015)

601
 40
Women victims of men who
murder
XML mark-up for nomination, collocation,
and frequency analysis of language of the law

Amanda Potts and Federica Formato

Introduction
Corpus approaches are increasingly being adopted in a number of fields concerned with
the study of language, gender, and sexuality. Tools and methods from corpus linguistics
are praised for decreasing analysis time, reducing subjectivity, heightening generalisability,
and providing potential for triangulation when combined with methods from a wide range
of (interdisciplinary) fields (Baker 2006). However, corpus approaches have been criticised
for ‘counting what is easy to count’ (Stubbs and Gerbig 1993: 8) and divorcing data from
context critical to interpretation (Baker 2010). Most problematically for new adopters, there
are no instruction booklets or proscribed steps for ‘good’ corpus linguistics (and many
warnings that it is easy to do ‘bad’ corpus linguistics). Indeed, there are very few agreed
guidelines about exactly what a corpus is, except that it must be a machine-readable text
of some description. Students often ask, ‘How big is a corpus? Do you measure: in words,
utterances, authors, or texts?’ This is slightly easier to answer than, ‘How long is a piece of
string?’ but only just.
Some scholars believe that corpus approaches will not be appropriate or even fully func-
tional with their data, due to restricted size. Many of these scholars are working with ‘oppor-
tunistic corpora’, which ‘represent nothing more nor less than the data that it was possible to
gather for a specific task’ (McEnery and Hardie 2012: 11). Some datasets are extraordinarily
difficult to scale up: for instance, corpora of dead or dying languages. Others are very time-
consuming to collect and/or process, meaning that any gain in size is accompanied by an
extreme expenditure of labour, as is the case with spoken corpora. Finally, some corpora
are simply restricted or entirely closed: the works of Shakespeare or Dickens, for example,
are no longer growing; whereas certain texts (e.g. US presidential inaugural addresses) are
released at set intervals, and while it is possible to collect all of these at a given point, the
resulting corpus will not number in the millions of words anytime soon.
In this chapter, we make use of one such opportunistic corpus (a collection of sentenc-
ing remarks from England and Wales) to explore legal representations of female victims of

602
 XML mark-up for analysing law language

homicide perpetrated by men. The case study presented here follows on from pioneering
studies undertaken with explicit emphasis on femininist standpointism, which seeks to free
women’s experiences within the legal system (as both offenders and victims) from ‘the need
to be consistently positioned in relation to male knowledges and understanding’ (Moore
2008: 53). Feminist criminological work has ‘followed the same pattern as that for offend-
ers, with the primary task firstly focused on making women’s victimization visible, and
secondly on correcting distorted characterizations of them’ (Silvestri and Crowther-Dowey
2016: 22). This case study does draw upon the dualism of women as victims of male vio-
lence, but in exploring the language that supports the legal narrative emerging from these
cases, we seek to uncover more nuances in the constructions of female victims.
With this aim, we have collected 14 judicial sentencing remarks from England and
Wales, comprised of 27,931 words. These remarks are drawn from cases where women are
the victims of homicide perpetrated by men, which is a type of violence particularly associ-
ated with gendered meanings. We hope to demonstrate how corpus linguistic techniques –
including frequency, collocation, and concordance analysis, but specifically XML mark-up
(a way of inserting searchable data about discursive features into the corpus) – might be
used to assist in the analysis of social actor representation (van Leeuwen 2008). By explor-
ing nomination and agency in particular, we show how corpus linguistics tools might be
used to highlight interesting findings, even in quite small data sets.

XML mark-up
As critical discourse analysts, we are interested in the ways that social actors are represented
in powerful texts, such as legal discourse. Methods of referring to social actors within our
sentencing remarks corpus (and most others) are extremely diverse. To allow for precise
analyses of the language, references to social actors should be both disambiguated (e.g.
establishing whether she refers to a victim or to a witness) and linked (e.g. indicating that
Rebecca Godden, Becky, she, and her refer to the same social actor). This opens up the
opportunity for search-and-recall on a large array of search terms that are unrelated at the
word level but are nearly synonymous at the discourse level.
To allow for the use of corpus techniques such as frequency, concordance, and colloca-
tion, we have incorporated XML mark-up (Hardie 2014) to enrich the sentencing remarks
corpus with initial linguistic analysis. XML mark-up is a way to store extra information
about lexical items by wrapping them in tags. For instance, if we are interested in analysing
social actors in texts, we can begin to ‘tag’ references to them. As Ms Symonds is one such
reference, we can add the following tag around the name:

• <actor>Ms Symonds</actor> was sitting in the corner of the sofa

Now the computer will know where actor begins (marked with open angle brackets) and
ends (signalled with closing angle brackets, indicated by a forward slash). However, Ms
Symonds carries more information to the human reader than merely the fact that this is a
social actor. For instance:

a) Ms indicates that this social actor is likely a woman;

b) In this clause, the social actor is in the nominative position, as the agent or ‘doer’;
c) The term of reference here is a given name, rather than a pronoun or a common noun.

603
Amanda Potts & Federica Formato

Therefore, we may wish to be more nuanced in our tagging, adding more information about
the item of interest. ‘Elements are the main building blocks of XML’ and the ‘structure is tri-
partite consisting of element, attribute, and attribute value’ (Rühlemann, Bagoutdinov, and
O’Donnell 2015: 9). In our example, the element is an actor. Attributes of interest include
gender (a, above), grammatical case (b, above), and name type (c, above). The values for these
attributes are ‘woman’, ‘nominative’, and ‘given’. This translates into the following tag:

• <actor gender=“f” case=“nom” name=“given”>Ms Symonds</actor> was sitting in the

corner of the sofa

Adding such tags to all items of interest means that the computer will be able to see (and
search for) all actors at once. It will also be able to search for more fine distinctions (e.g.
all actors whose names are given) or combinations (e.g. all female actors in the nominative
position). Notice that these tags seem perfectly suited to our research interest. XML does not
have predefined tags and it is extensible, which means that users can define new elements,
attributes, or values at any time, and the computer will interpret them without issue.
This is not to say that racing into a corpus and assigning ad hoc, snowballing collections
of tags will be helpful for research. Like all methods, a measured, scientific approach to this
one is the most beneficial. Edwards (2001) wrote at length about general design principles
for representing spoken language in a written/spatial medium (i.e. transcription or alterna-
tive visualisations). These principles remain salient when applying XML mark-up to enrich
textual data, and therefore, we reproduce and adapt them below.

Computer tractability
As we are taking a corpus linguistic approach, working with computerised texts, the first and
foremost principle is that of computer tractability. For us, ‘the single most important design
principle is that similar instances be encoded in predictably similar ways’ (Edwards 2001:
324, emphasis in original). Human readers can tell, for instance, that tho and though are vari-
ant encodings of the same word; likewise, once social actors have been introduced in a text,
practised readers will follow cohesion strategies and recognise that variants (i.e. Georgina and
Ms Symonds) refer to the same person. However, unless special provisions are made, com-
puters will treat tho, though, Georgina, and Ms Symonds as entirely unrelated entities. Why
not just search for these separately? First, compiling all possibilities may be difficult or even
impossible. Perhaps more importantly, searching for – and considering – one variant may lead
to unrepresentative and misleading results. Adding tags which link all variants is one way of
allowing the computer to read a text like a human would … only much more quickly.

Category design
Now that we have decided to mark-up a set of interest within our corpus, we must decide which
categories are meaningful to our study. In our case study, we are adding XML mark-up under
one element (actor) with three attributes (gender, case, name type), each with a set of possible
values. To be useful, elements, attributes, and values must satisfy the following criteria:

1) They must be systematically discriminable. That is […] it must be clear whether or not
a given category applies. Category membership can be based on either defining charac-
teristics or similarity to prototypical exemplars.

604
 XML mark-up for analysing law language

2) They must be exhaustive. That is, for each relevant aspect or event in the data, there must
be a category which fits (even if, in hopefully rare cases, it is only “miscellaneous”).
3) They must be usefully contrastive. That is, they must be focused on distinctions of
importance to the research question.
(Edwards 2001: 323)

This set of criteria will look familiar to anyone who has undertaken a study with an ele-
ment of categorisation or qualitative description. Put simply: in order for resulting discus-
sions to be meaningful, we must have a research question which necessitates one or more
distinctions between categories and we must create enough categories that each instance
has a good, clear fit (and no more than that). It is important to be very systematic in design,
as categories ‘within a contrast set usually cannot be interpreted without knowledge of the
number and type of other categories in that set’ (Edwards 2001: 323). When doing discourse
analysis, being exhaustive in category design also comes with a built-in reward: very rare or
particularly interesting cases will have attributes and values of their own, and will be identi-
fied and detailed in any discussion. Therefore, low frequency is no longer a ‘hiding place’
of noteworthy results, and we are no longer guilty of counting what is easy to count (Stubbs
and Gerbig 1993: 78).

Visual display
In discussing how to translate information for computers, we have not lost track of the fact
that humans must also read the data and these readers have expectations of written texts.
In the case of XML mark-up, we must first make things a bit worse before we make them
a bit better.
Figure 40.1 is an example of a sentencing remark, as provided in PDF on the Courts and
Tribunals Judiciary website (2017); it is the first paragraph of R -v- Lawrence.
This is very easy to read, but perhaps quite difficult to analyse. Researchers interested in
naming strategies might have to mentally collate the various patterns happening throughout
the text. They might highlight (in various colours) or create tables or figures (in another
programme) to tally features of interest.
In Figure 40.2, the same paragraph appears with XML mark-up. Admittedly, this is more
difficult to read than the original. We are not used to so much ‘meaningless’ punctuation and
data that does not conform to expected sentence structure.
However, let’s search for all instances of a set (women homicide victims) in a corpus pro-
gramme, Sketch Engine. Figure 40.3 shows the output for a search for <actor gender=“f”/>.
Arguably, this is the easiest format to read. All of the items of interest are centred, with
some context on either side. We do not need to read the entire paragraph to pick out exam-
ples of our category.

You have been found guilty by the jury of murder. On 6 October 2012 – 2 days before the
decree nisi for your divorce – you murdered your wife, Sally, by deliberately crashing your
car in which she was a passenger into a tree.

Figure 40.1 Example text without any additional mark-up.

605
Amanda Potts & Federica Formato

Figure 40.2 Example text with XML mark-up.

nisi for your divorce - you murdered your wife , Sally, by deliberately crashing your car
for your divorce - you murdered your wife, Sally , by deliberately crashing your car in which
deliberately crashing your car in which she was a passenger into a tree. The way in

Figure 40.3 Example output in a Sketch Engine concordance window.

We can see, at a glance, how the social actors are referred to, and with the use of more
sophisticated search terms, we can restrict the results to certain values. The data and the
presentation remain separate.
In our case, XML is written by humans to be read by machines. Once it is added, it does
not need to be inspected directly; it is ‘hidden’ behind the data that humans find most eas-
ily understood, but is effortlessly accessible to do some computational heavy lifting, when
required. To demonstrate how this works in a bit more detail, we now move to the case study
of sentencing remarks.

Case study: women victims of homicide perpetrated by men

Male-perpetrated violence against women is contextualised within prevailing beliefs about
and attitudes towards gender (Dobash and Dobash 1998), making discursive constructions of
women homicide victims a salient topic for this volume. Previous work in this area has ana-
lysed media representations of female victims (e.g. Monckton-Smith 2012) and explored the
relationship of the victim to the perpetrator (Coates and Wade 2004; Ehrlich 2008; Monckton-
Smith 2012). However, a common scholarly focus is scrutiny of the normalisation of imbal-
anced gender(ed) roles, which cannot be disregarded when investigating women as victims
of violence. Conceptualisations of masculinities (of the offenders) and femininities (of the
victims) can be made relevant and visible as part of a gendered legal structure in which roles
and identities are interpreted and presented in language used by the judiciary. We interpret this
language as ‘ideological’, or encoding widespread beliefs and values in relation to gender(ed)
identities concerning the victims involved as well as a cultural male-oriented status quo.
Due to relative scarcity of forensic data, much previous linguistic research on the rep-
resentation of female victims and male perpetrators of intimate violence has focused on
media discourse. These studies found that women victims are often described in relation to
their partners (Meyers 1997) while men are exonerated from criminal agency, depicted as
having lost control or being under the influence of substances during the acts (Monckton-
Smith 2012). From a corpus linguistics and stylistic perspective, Tabbert (2016) investigates
the construction of offenders and perpetrators (not uniquely in relation to gender) in a cor-
pus of newspaper articles from The Guardian. Her findings reveal interesting insights into
how women are described, particularly highlighting their relationships (as wives, mothers,

606
 XML mark-up for analysing law language

daughters) and their age or other perceived vulnerabilities. We will revisit many of these
themes in our own case study, below.

Data collection and description

All sentencing remarks have been collected from the Courts and Tribunals Judiciary web-
site,1 where they are freely available for public viewing and download. The corpus is com-
prised of all available sentencing remarks meeting three criteria:

• Offender(s): must be adult men, only;

• Victim(s): must be adult women, only;
• Offence(s): must include murder or manslaughter.

This resulted in a corpus of 14 texts, published between 11th May 2012 and 23rd March
2017, totalling 27,931 words. A brief overview of the particulars of each case can be found
in Table 40.1.
It must be noted that the very small size of this corpus does not accurately reflect the
number of homicide cases in England and Wales. In the year ending March 2015, 530 homi-
cides were recorded, with 558 suspects. Court proceedings had concluded for 285 suspects,
of which 257 (90%) were male. Among these male offenders, 48% were indicted for a homi-
cide offence and were convicted of murder, with a further 33% convicted of manslaughter.

Table 40.1 Overview of cases included in sentencing remarks corpus

Offender(s) Offence(s) Victim(s) Relationship(s) to Words

killer

Steven Beards Rape, murder Susan Whiting Other known 2,236

Martin Bell Manslaughter Gemma Simpson Other known 4,187
Christopher Murder Sian O’Callaghan Stranger 1,665
Halliwell (1)
Christopher Murder Rebecca Godden Other known 1,987
Halliwell (2)
Michael Lane Murder, arson, theft Shana Grice Ex-partner 2,190
Iain Lawrence Murder Sally Lawrence Partner 1,251
John Lowe Murder Christine Lee & Lucy Ex-partner 1,567
Lee (her daughter)
Tony McCluskie Murder Gemma McCluskie Other family 844
Peter Morgan Murder Georgina Symonds Other known 3,666
David Oakes Murder Christine Chambers & Partner and other 1,296
Shania Chambers family
(her daughter)
Ahmad Otak Murder Kimberley Frank & Other known 2,876
Samantha Sykes
Justin Robertson & Murder (contracted by Pennie Davis Other family 1,007
Ben Carr Carr, carried out by
Roberston)
Devendra Singh Murder Charlotte Smith Partner 1,675
Ian Stewart Murder Helen Bailey Partner 1,484

607
Amanda Potts & Federica Formato

In this year, 186 victims (36%) were female. There are some interesting gender differences
in the relationships between victims and principal suspects: ‘female victims (68%) were
more likely than male victims (52%) to have been acquainted with the principal suspect’,
and ‘women were far more likely than men to be killed by partners/ex-partners (44% of
female victims compared with 6% of male victims)’ (Office for National Statistics 2016).
This pattern certainly holds true in our data, with only one offender (Halliwell 1) unknown
to his victim, and 5 out of 14 (36%) women killed by current or previous partners.
The limited size of the data set, therefore, does not represent a rarity of criminological
relationship but, instead, reflects the small number of sentencing remarks published online.
It is not required by law that these are made public, and usually sentencing remarks are
released only for those cases attracting particular public or media attention. Being cog-
nisant of the possible reach of their remarks, there is an argument that judges’ language is
influenced by the potential for publicity. However, we believe that this awareness, com-
bined with increased access, makes these texts all the more interesting as sites of ideologi-
cal work.

Methods
Previously, we stressed the importance of principled design of XML elements, attributes,
and values. This is due, in large part, to the fact that findings are difficult to interpret without
knowing the full range (and population) of the total set of possibilities. In our case study,
the main element is reference to social actors. In keeping with our research interest in the
construction of women victims of male-perpetrated homicide, every reference to a victim has
been manually marked-up for three features: sex/gender,2 (grammatical) case, and naming
convention. For the first tag, we only consider victims tagged as ‘female’, or WHV (Women
Homicide Victims). The full range of possible attribute values for case and name for WHV
appears below.

Mark-up of case
• nom: nominative; the WHV is the subject or ‘doer’ in the clause;
• acc: accusative; the WHV is the object or ‘receiver’ in the clause;
• gen: genitive/possessive; the WHV is the possessor of a noun in the clause;
• dat: dative, ablative, and locative; the WHV is the indirect recipient/beneficiary/loca-
tion of an action;
• voc: vocative; when an WHV is directly addressed by name, but this naming is not
embedded in a clause containing a verb phrase;
• ref: reflexive; the WHV is the subject and object of the action.

Mark-up of naming convention

• pro: designates pronoun usage;
• giv: use of a given name, which may include title, (reduced) forename, and/or surname;
• cat: reference to social actors through categorising naming strategies, foregrounding
some type of identification or functionalisation.

Manual XML mark-up (as opposed to a computational approach) was the most effective
method for a number of reasons:

608
 XML mark-up for analysing law language

1. High lexical variation: in the given name category alone, there are 48 naming strategies
for WHV;
2. Need to disambiguate: while it would have been easy to automatically mark-up pro-
nouns, for instance, only some of these refer to WHV, whereas others refer to female
witnesses and other figures;
3. Lack of a lexicon: categorising naming strategies cannot be adequately predicted and
manual mark-up allows for the emergence of organic findings.

When using search-and-recall methods, the size of the overall corpus is of less relevance
than the frequency of salient features contained within the data. Therefore, despite the rela-
tively small size of the data set, there are a number of interesting results which are indicative
of the genre. Incorporating XML mark-up allows us to instantaneously search for naming
strategies, cross-tabulated by gender or case. This, in turn, affords greater flexibility in the
application of frequency and concordance analysis, as we shall see below.

Analysis
In this section, we present demonstrative findings to illustrate how corpus methods may be
useful in shedding light on discursive choices made by the judges when constructing WHV.
Analyses were undertaken in Sketch Engine, as we believe that this is the corpus tool with
the most intuitive XML integration. Please note: when we show illustrative data, references
to social actors will be in boldface and remark names will be provided in square brackets.

Naming strategies
We begin with an overview of the full frequency of naming strategies. Assigning XML
attributes and values (i.e. name=“pro”, “nom”, or “cat”) allows an immediate window into
the overall picture of naming in the corpus (see Table 40.2).
Pronominal references are the most common method of referring to women victims in
our corpus of sentencing remarks. However, these references are extremely homogenous:
each of the 650 instances are in the third person. While this is largely to be expected, it is
noteworthy that in no cases have the voices of the victims themselves been incorporated to
assist in reconstruction or final re-narrativisation of the crime for the public record.
Whilst the use of pronouns is far more frequent than other naming strategies, frequency
data also shows that WHV are also commonly referred to (or ‘nominated’) by combinations
of their given names. In and of itself, this fact is unsurprising due to the non-deictic nature
of these references in this genre. However, as our subsequent analysis will demonstrate, the
distance and solidarity relationships intimated by certain naming strategies provide some

Table 40.2 Overview of naming strategy categories with frequencies and

percentages

Category of naming strategy Frequency Percentage

Pronominal 650 62.0%

Nominative 295 28.1%
Categorising 104 9.9%
Total 1049

609
Amanda Potts & Federica Formato

interesting and novel insights into the gendered identity constructions of the female defend-
ants under consideration.
Scholars have investigated degrees of power and solidarity in spoken interaction as
constructed through nomination strategies (Hook 1984; Leech 1999). The trend towards
discursive absence of a major category of social actors and the overall lack of dialogue in
sentencing remarks makes them somewhat different from other forms of spoken interaction
(though, of course, sentencing remarks are written to be spoken). However, the use of given
name strategies is still central in constructing identities for WHV. Preference of one form
over another has the potential to provide insights into solidarity/distance relations between
the narrator and the referent – in this case, judges and victims). Therefore, we have adapted
frameworks from Hook (1984) and Leech (1999) to accommodate the shift in genre and
register, and to account for all items in the corpus (see Figure 40.4).
Given name strategies comprise approximately 13% of all references in our corpus of
sentencing remarks. Even within this value, types are extremely diverse. In Table 40.3, we
provide an overview of given name types, frequencies, and percentages of nomination.
The table shows that the most used strategy is forename only. In relation to the scale
in Figure 40.4, this strategy can be seen to construct solidarity between the judges and
the WHV. This is likely due to an effort on the part of the judges to humanise the victims
through one of the most intimate attributes of social identity one can have, i.e. their personal
name.

(1) Spitting blood on Kimberley’s body was an act of bestiality. [Otak]

Initials (e.g. GM)

Greater solidarity

Greater distance

Surname only (e.g. Chambers*)

Title + surname (e.g. Ms Symonds)
Forename + surname (e.g. Gemma Simpson)
Forename only (e.g. Susan)
Forename: diminutives (e.g. Becky)

Figure 40.4 Framework developed to operationalise categories of solidarity and distance

between the judge and the victims. *Provided examples are from the corpus,
excepting ‘Surname only’, which does not appear in the sentencing remarks
analysed

Table 40.3 Type, frequency, and percentage of given name references

Reference type Frequency % Nomination

Forename only 170 57.6%

Forename + surname 90 30.5%
Title + surname 30 10.2%
Forename: diminutive 3 1.0%
Initials 2 0.7%
Total 295 100%

610
 XML mark-up for analysing law language

In Example 1, the judge empathetically constructs proximity with the victim by using
forename only and commenting on the brutality of the actions committed by the perpetrator.
This is a strategy which might be effective in a number of courtroom scenarios: Rosulek
(2008) found that defendants’ lawyers used forename only for their clients (accused of using
violence against children) and forename + surname for the victims. Formato (2019) and
Tabbert (2016) found similar results, where forenames were employed to construct soli-
darity with victims of violence, while surnames (a distancing form) were mostly used for
offenders. Frequent use of this strategy by judges aligns them with victims and against per-
petrators in a way which might not be obvious to uncritical readers.
The other forms of nomination listed above (i.e. forename + surname and title + sur-
name) are also interesting as expressions of judges’ solidarity with – or distance from –
the victims. We found that forename + surname is a neutral, institutional naming strategy.
Conversely, title + surname is used to construct traditional gender roles: Miss and Mrs
signal different (positive) ideological interpretations of the WHV’s age, relationship status,
and subsequent positions within our largely patriarchal, heteronormative, society. However,
title + surname can also serve to other a person, as in the case of Ms, which is used for one
victim who was soon-to-be-divorced, and one who was a sex worker.
Among the least used, diminutives are an interesting strategy to signal strong solidarity
between the judge and the victim. Diminutive use often occurs alongside voices and per-
spectives of family members, bringing WHV close to the reader and (re)constructing these
adult women victims as childlike and particularly vulnerable (see Example 2).

(2) Now both their children are gone and, in essence, both of Charlie’s parents, to whom
she meant so much have had their lives devastated and made empty and, as they say
in a joint statement, words cannot express how it feels to lose her in such a brutal way.
[Singh]

This section has shown how naming strategies can be used to construct solidarity (or dis-
tance) between the judges and the victims in our corpus. A notable result here is the use of
forename only and diminutive, which construct a sense of familiarity as well as vulnerability
in the sentencing remarks of WHV.
The main subject of this chapter so far has been ‘nomination’, or reference to social
actors’ unique identities in discourse. According to van Leeuwen (2008: 42–44), naming
strategies also fall under the following categories:

• Identification:
• Classification: representation based on membership of certain major social catego-
ries such as gender, age, wealth, e.g. teenager;
• Relational: representation based on relationships, such as personal, kinship, or
work relations, e.g. mother;
• Physical: representation based on physical characteristics, e.g. blonde;
• Functionalisation: references in terms of an activity, such as a role or occupation, e.g.
killer.

As above, we have run a search for all XML tags in which name=“cat”, capturing the
full frequency and range of this reference type. We coded these further within van Leeuwen
(2008)’s framework; results can be found in Table 40.4.

611
Amanda Potts & Federica Formato

Table 40.4 Frequency of categorising strategies

Reference type Sub-type Freq.

Identification relational 57
classification 25
physical 0
Functionalisation 12

The most frequent categorisation is relational identification. WHV are sisters (freq. 15),
daughters (12), mothers (11), wives (7), friends (7), children (2), partners (2), and fiancées
(1). This pattern reflects the fact that most of the WHV in our corpus (and in general) are
related to their killers (see Table 40.1 and Example 3 below), and echoes previous find-
ings showing that family members often provide statements incorporated into sentencing
remarks (see Example 4).

(3) … you have been convicted after a trial of murdering your fiancée Helen Bailey who
was excitedly making arrangements for your wedding while you were planning how to
kill her, hide her body and explain her disappearance as a case of an anxious woman
running off because she could no longer cope with the more stressful aspects of her life.
[Stewart]
(4) [Shana Grice’s mother] also said: ‘We have lost our beautiful, kind and thoughtful
daughter; we miss her giggles and laughter, the jokes we shared and having her to hold
and share our future lives together as a family.’ [Lane]

Even considering these two explanations, however, it is interesting to note that all of the
top relational identifications (sisters, daughters, mothers, wives, fiancées) are gendered;
only 12 out of 57 items in this subcategory (friends, children, partners) are gender-neutral.
When WHV are being constructed in law, they have still come as women, and their gendered
identities as relative to others is still foregrounded.
Certainly, these results are intriguing. However, we have already stated that one perceived
shortcoming of corpus methods is ‘counting what is easy to count’ (Stubbs and Gerbig 1993:
8), and we are aware of the pitfalls of focusing on high-level results only. Therefore, we will
make a more detailed exploration of the least populous subcategory: functionalisation. To
make sense of the forms, we grouped the naming strategies into two subcategories based on
the functions of these people:

• Legal role (freq. 9): victim (4), missing person (2), victims (2), vulnerable victim (1);
• Professional role (freq. 3): unnamed prostitute (1), breadwinner (1) gifted author (1).

The role of victims within the courtroom and legal process is three times as frequent
as their roles before their deaths. Interestingly, narrativisation of events leading up to the
crimes already position them as vulnerable victims. In Example 5, we see the judge refer-
ring to Sian O’Callaghan as a vulnerable victim (instead of, perhaps, ‘a vulnerable person’)
before the crime. This can be seen in relation to her physical condition at the moment of
the abduction (‘under the influence of drink’) yet more broadly within an accepted frame
for which walking alone at night is a recognised danger for young women (Mallicoat 2015;
Silvestri and Crowther-Dowey 2016).

612
 XML mark-up for analysing law language

(5) You abused your position as a taxi driver, in a car clearly marked as a taxi, and as some-
one Sian thought she could trust; her abduction was clearly premeditated; as a young
woman walking alone late at night and under the influence of drink she was a vulner-
able victim. [Halliwell 1]

To conclude, this naming strategy seems to present victims in terms of their functions
within the judicial case rather than in relation to their lived experiences before the crimes
that claimed their lives. Outside of their primary identities as victims, WHV are positioned
predominantly as family members, rather than people who had rich, multi-faceted lives as
friends, professionals, and community members.

Case
In this section, we discuss grammatical position (namely: nominative, accusative, genitive,
dative, vocative, and reflexive) of the WHV in our sentencing remarks, with the aim of gain-
ing further insights into how these women are constructed. In this section, we briefly discuss
results, a quantitative overview of which can be found in Table 40.5.
WHV most frequently occupy the genitive position in our corpus, followed by nomina-
tive and accusative. Though it appears that WHV are marginally more ‘active’ than ‘pas-
sive’, we must consider the dative case together with accusative, as this similarly constructs
WHV as secondary ‘recipients’ of objects and actions. Taking these in combination, WHV
are ‘possessors’ in 37.1% of instances, ‘done to’ 31.9%, and ‘doers’ 28.8%, limiting overall
opportunities for active agency (i.e. nominative case) quite dramatically. In the sample anal-
ysis below, we focus on the most frequent case to demonstrate how XML mark-up might
be combined with collocation analysis, or investigation of frequent co-occurrence of words.
The most frequent grammatical position occupied by WHV is the genitive, or posses-
sive. In order to investigate what the women possess and how this can be examined, we
conducted collocation analysis in the range of +1 to +3 (or one to three words to the right
of the naming strategy) to capture the most frequent form: her x or [name]’s x. In doing
collocation analysis, researchers must make a number of decisions, from which measure to
use (one based on confidence, strength of association, or a mixture of both), to what cut-offs
will apply (for instance, whether words must co-occur together a certain number of times or
reach a certain statistical threshold to be counted). We applied some commonly used cut-off
scores3 and grouped the resulting collocates by semantic category in Table 40.6.
Though WHV are frequent possessors, what they do possess is without nuance. They
are described as collections of body parts that have been acted (violently) upon, often in a

Table 40.5 Frequency and percentages of (grammatical) case

Case Frequency % Occurrence

Genitive 389 37.1%

Nominative 302 28.8%
Accusative 245 23.4%
Dative 89 8.5%
Vocative 12 1.1%
Reflexive 2 0.2%
Total 1,039 100%

613
Amanda Potts & Federica Formato

Table 40.6 Collocates of WHV in the genitive case, categorised semantically and ranked in descending
order of frequency of collocation

Semantic category Collocates Frequency

Body parts body, head, neck, throat, remains, legs 64

Relationships family, mother, daughter, father, dog, friend, sister, friends 54
Physical objects phone, mobile, car, money 21
Life death, life 21
Places house, home, bedroom 18

familiar setting (their house, home, bedroom – all forensic sites which also carry a particu-
larly gendered coding).
The second-most populous category (relationships) echoes previous findings from the
categorisation section of this study and from Formato (2019) and Tabbert (2016), in that
women’s identities are dominated by their familial positioning. Here, we see that they
possess far more relationships than physical objects (even in combination with places).
These collocates describe relationships that WHV had to other victims, to their killers, or to
remaining family members.
Of course, WHV also possess their lives, until the point of their deaths, the cause of
the criminal proceedings. Intuitively, possessing life could be perceived as humanising and
empowering these women, based on their achievements before death. However, concord-
ance analysis shows that victims’ lives can also be conceptualised (and negatively evalu-
ated) in relation to the notion of the ‘ideal victim’ (see Example 6).

(6) Sadly her young life was troubled and blighted by her drug abuse which forced her into
prostitution to fund her addiction. [Halliwell 2]

Extremely infrequently are WHV constructed as possessing physical property (e.g. phone,
car). When considered together with findings from the analysis of nomination and categorisa-
tion strategies, sentencing remarks (which form the final official narrative of a case and often
serve as emotional closure for families) construct victims simplistically. While judges do
establish solidarity through nomination, WHV are still reduced to a collection of body parts
and (disrupted) relationships, with little indication of their identities outside of the crime.

Discussion
Narratives of crime which: (1) refer to adult female victims through forename only and
diminutives; (2) focus on facets of self-regarding relationships to others while obscuring
nuances of individualism; and (3) reduce frequency and scope of agency, have been found
to associate WHV with the private, domestic sphere. It is this very patriarchal structure
within the home that manifests itself in extreme forms of male violence against women,
and as such, further research on larger corpora of forensic data may prove interesting and
insightful.

614
 XML mark-up for analysing law language

Conclusion
Methodologically speaking, we believe that XML mark-up has a number of strengths. To
revisit the points enumerated in the first section, this technique bolsters the strengths of
other corpus linguistic methods thus:

• It has the potential to reduce the influence of subjectivity and cognitive bias on analyses
by necessitating that the researcher confronts every instance of a certain element within
their data and to define and apply appropriate attributes and values, rather than ‘eye-
balling’ or searching for high-frequency, obvious, or (subjectively) interesting features;
• It decreases analysis time after the initial period of mark-up by optimising searching,
putting the onus on the computer;
• It heightens generalisability, as all attributes and values within a certain category/ele-
ment are known and accounted for. These can then be adopted or adapted in future
work, as XML is infinitely extensible;
• And as it is just a way of embedding more data about the text, it provides potential for
triangulation when combined with other methods. We have included some feminist
legal analysis and (critical) discourse analysis here, alongside analysis of case and nam-
ing, to demonstrate how many ‘ways in’ a small amount of data can provide.

XML mark-up also brings us some way towards addressing some of the perceived limita-
tions of corpus approaches:

• We are not simply ‘counting what is easy to count’ (Stubbs and Gerbig 1993: 8), but
rather, accounting for entire ranges and arrays of meaning-making in texts;
• By reading the data to do manual mark-up, we as researchers remain highly cognisant
of the context and full content of our corpora.
• Finally, by linking items which are related at the meaning level, we are able to effec-
tively ‘stretch’ small datasets, opening up the possibility to perform more sophisticated
techniques (such as collocation) on opportunistic corpora.

There are, of course, further limitations to this method. The initial process of XML mark-
up may be considered time-consuming; as an indication, mark-up on our corpus of approxi-
mately 28,000 words took about 24 hours to do and another 16 hours to cross-check and
correct. Human coders make errors and it is best practice (though not always possible) to
have two researchers on hand to help reach inter-rater agreement. Alternative methods or
future directions may include the assistance of – or replacement by – computational taggers.

Future directions
The methodology presented here can be employed to investigate comparative data from
several perspectives and approaches. For instance, scholars may wish to make use of XML
mark-up to conduct comparative analysis with corpora of sentencing remarks for women
who kill (e.g. Potts and Weare 2018) to explore alternative constructions of women in dif-
ferent positions in the legal system. Alternatively, scholars may wish to compare construc-
tions of men and women in a given corpus, or to investigate agency and naming strategies
in subgenres (e.g. witness statements, police reports, closing remarks), in languages other
than English (e.g. Formato 2019), or different genres entirely (e.g. doctor–patient interac-
tion, media discourse).
615
Amanda Potts & Federica Formato

We hope that we have demonstrated the usefulness of XML mark-up, particularly for
small, opportunistic corpora. Most corpus tools (i.e. WordSmith Tools, AntConc, Wmatrix,
and Sketch Engine) are XML compatible, offering the ability to query the underlying mark-
up, while keeping it hidden from view. In this illustrative case study, we have only made use
of one level of XML elements, though these are easily nested. For instance, most of the judges
in our corpus are men. If we wanted to restrict results to include only texts created by women
judges (or to distinguish between sentencing remarks from England and Wales, or to restrict
by year), we might nest additional levels of mark-up (called ‘child elements’), as such:

<text year=“2015” place=“Wales” judge=“f”>

…
<actor … > … </actor>
…
</text>

In this way, researchers may combine queries for text-external data and text-internal
data, or to search within layers of text-internal data. Once more: a great strength of XML
mark-up is its unlimited flexibility and extensibility.
For researchers who believe that tagging might be of use to their research but are not
willing or able to undertake manual mark-up, a number of alternative avenues are cur-
rently available. Part-of-speech (POS) tagging is an automated process that may help
researchers to analyse, for instance, differences in frequency of certain grammatical
types (e.g. interjections in women’s talk). This is available in English through CLAWS4
using stand-alone software, a free internet interface, or Wmatrix.5 Uploading corpora into
Sketch Engine6 also allows for POS tagging in a great variety of languages. Semantic tag-
ging is another automated method of adding information about word and phrase relation-
ships to the data.

Notes
1 https://www.judiciary.gov.uk/
2 To our knowledge, all victims in this corpus are cis-gender; they are referred to as both female(s) and
wom(a/e)n throughout the sentencing remarks. We acknowledge the problematic nature of conflat-
ing sex and gender in this way, though we do use female/woman interchangeably here, reflecting the
original data.
3 Mutual Information ≥ 3 and Log Likelihood ≥ 6.67 (for further discussion, see: Hunston, 2002;
Rayson, Berridge, and Francis, 2004).
4 http://ucrel.lancs.ac.uk/claws/
5 http://ucrel.lancs.ac.uk/wmatrix/
6 http://sketchengine.co.uk/

Further reading
Johnson, A. (2015) ‘Haunting evidence: Quoting the prisoner in 19th century old bailey trial discourse.
The defences of Cooper (1842) and McNaughten (1843)’. In: Arendholz, J., Bublitz, W., and
Kirner-Ludwig, M. (eds.) The Pragmatics of Quoting Now and Then. London: De Gruyter Mouton,
pp. 369–400.
Using a 15-trial, 133,000-word subsection of the Old Bailey Corpus, Johnson analyses patterns of
quoting in legal language. Though this work does not make use of mark-up, it is a corpus linguistic
study suggesting further avenues for query which would prove fruitful when combined with XML.

616
 XML mark-up for analysing law language

Lawson, R. and Lutzky, U. (2016) ‘Not getting a word in edgeways? Language, gender, and identity
in a British comedy panel show’. Discourse, Context and Media, 13, pp. 143–153.
This is a sociolinguistic analysis of features such as talkativeness and interruption, carried out on a
corpus of televised media and making use of XML.
Potts, A. and Weare, S. (2018) ‘Mother, monster, Mrs, I: A critical evaluation of gendered naming
strategies in English sentencing remarks of women who kill’. International Journal for the
Semiotics of the Law, 31, pp. 21–52.
For readers who are interested in reading a more complete, detailed study of this sort, this paper
uses the precise methodology outlined in this chapter (XML mark-up on a small, opportunistic corpus)
to investigate naming strategies of women who kill.
Rühlemann, C. and Gee, M. (2018) ‘Conversation analysis and the XML method’. Gesprächsforschung
[Discourse and Conversation Analysis], 18, pp. 274–296.
This work provides helpful, practical suggestions for implementing XML in corpora.

Related topics
Semiotic representations of women criminals; determining the impact of gender stereotyping on patient
feedback; investigating gendered language through collocation; the impact of language and gender
studies: public engagement and wider communication; text trajectories and gendered inequalities in
institutions.

References
Baker, P. (2006) Using Corpora in Discourse Analysis. London: Continuum.
Baker, P. (2010) Sociolinguistics and Corpus Linguistics. Edinburgh: Edinburgh University Press.
Coates, L. and Wade, A. (2004) ‘Telling it like it isn’t: Obscuring perpetrator responsibility for violent
crime’. Discourse & Society, 15, pp. 499–526.
Dobash, R. E. and Dobash, R. P. (eds.) (1998) Rethinking Violence against Women. London: SAGE.
Edwards, J. (2001) ‘The transcription of discourse’. In: Schiffrin, D., Tannen, D., and Hamilton, H.
(eds.) Handbook of Discourse Analysis. Oxford: Blackwell, pp. 321–348.
Ehrlich, S. (2008) ‘Sexual assault trials, discursive identities and institutional change’. In: Dolón, R.
and Todolí, S. (eds.) Analysing Identities in Discourse. Amsterdam: Benjamins, pp. 159–177.
Formato, F. (2019) Gender, Discourse and Ideology in Italian. Basingstoke: Palgrave.
Hardie, A. (2014) ‘Modest XML for corpora: Not a standard, but a suggestion’. ICAME Journal, 38,
pp. 73–103.
Hook, D. D. (1984) ‘First names and titles as solidarity and power semantics in English’. IRAL -
International Review of Applied Linguistics in Language Teaching, 22(3), pp. 183–190.
Hunston, S. (2002) Corpora in Applied Linguistics. Cambridge: Cambridge University Press.
Leech, G. (1999) ‘The distribution and function of vocatives in American and British English
conversation’. In: Hasselgård, H. and Oksefjel, S. (eds.) Out of Corpora: Studies in Honour of Stig
Johansson. Amsterdam: Rodopi, pp. 107–118.
Mallicoat, S. (2015) Women and Crime, 2nd edn. Los Angeles, CA: Sage.
McEnery, T. and Hardie, A. (2012) Corpus Linguistics: Method, Theory and Practice. Cambridge:
Cambridge University Press.
Meyers, M. (1997) News Coverage of Violence against Women: Engendering Blame. London: SAGE.
Monckton-Smith, J. (2012) Murder, Gender and the Media: Narratives of Dangerous Love. New
York, NY: Springer.
Moore, D. (2008) ‘Feminist criminology: Gain, loss and backlash’. Sociology Compass, 2(1), pp.
48–61.
Office for National Statistics (2016) Compendium Homicide. Available at: https://www.ons.gov.uk/
peoplepopulationandcommunity/crimeandjustice/compendium/focusonviolentcrimeandsexualoff

617
Amanda Potts & Federica Formato

ences/yearendingmarch2015/chapter2homicide#relationship-between-victim-and-principal-suspe
ct (Accessed: 25 July 2017).
Rayson, P., Berridge, D., and Francis, B. (2004) ‘Extending the Cochran rule for the comparison of
word frequencies between corpora’. In: Purnelle, G., Fairon, C., and Dister, A. (eds.) Le poids des
mots: Proceedings of the 7th International Conference on Statistical analysis of Textual Data.
Louvain-la-Neuve: Presses Universitaires de Louvain, pp. 926–936.
Rosulek, L. (2008) ‘Manipulative silence and social representation in the closing arguments of a child
sexual abuse case’. Text & Talk, 28, pp. 529–550.
Rühlemann, C., Bagoutdinov, A., and O’Donnell, M. B. (2015) ‘Modest XPath and XQuery for
corpora: Exploiting deep XML annotation’. ICAME Journal, 39, pp. 47–84.
Silvestri, M. and Crowther-Dowey, C. (2016) Gender and Crime: A Human Rights Approach. London:
SAGE.
Stubbs, M. and Gerbig, A. (1993) ‘Human and inhuman geography: On the computer-assisted analysis
of long texts’. In: Hoey, M. (ed.) Data, Description, Discourse: Papers on the English Language in
Honour of John McH Sinclair on his Sixtieth Birthday. London: Harper Collins, pp. 64–85.
Tabbert, U. (2016) Language and Crime: Constructing Offenders and Victims in Newspaper Reports.
Basingstoke: Palgrave.
van Leeuwen, T. (2008) Discourse and Practice: New Tools for Critical Discourse Analysis. Oxford:
Oxford University Press.

618
 Index

affect 32–33, 58–59, 450–452, 461, 520 discourse: critical see Critical Discourse Analysis;
agency 2–3, 5–6, 79, 815, 212, 219, 266–267, digital 482, 486–487; gendered 117, 215, 224,
437–439, 446 408, 411, 423, 465, 468, 559, 565; hegemonic
Agnes 244–245, 248, 274, 298, 369 169, 222, 229, 423, 461, 567; institutional 15,
anthropology 164, 184, 213, 254; cultural 97, 354, 365, 382, 525; macro- 332, 345; political
100, 136, 140; feminist 137; linguistic 93, 41, 329; public 486, 509, 583; workplace
103, 138, 140, 275, 324–325, 354; 212–214
sociocultural 55 discourse analysis: anthropological 93, 103, 136;
critical see Critical Discourse Analysis;
Baxter, J. xxvi, 6, 198, 408–409, 411, 422, feminist poststructuralist (FPDA) xxvi–xxvii,
425, 432 405, 422, 424–426, 433; multimodal 9, 485
binary: gendering 290, 299, 301; non-binary 25, discrimination 7, 9, 77, 184–185, 328–329, 388,
37, 56, 69–70, 72–73, 75–78, 290; opposition 402, 472–474, 511
4, 25, 231, 440, 444, 447 dispute 74, 275, 304, 306–307, 316
Bourdieu, P. 8, 10, 94, 102, 109, 400, 588 dominance approach 2, 325
burkini 15, 404, 439–447 double bind 189–190, 423, 467–469
Butler, J. 4, 28, 38, 71, 93, 273–274, 340, 451, drag king 272, 276–277, 279, 283–284
473, 546–547
ethnicity 25, 42, 55, 69, 113, 305–307, 326–328,
camgirl 167–168, 170–172 438–439, 545–546, 553
class 25–26, 28, 74, 101, 123, 153, 231, 306, ethnography 8–9, 75, 93–94, 100–102, 108–109,
539, 545 123–124, 138, 150–151, 164–165, 544; digital
Communities of Practice (CofP) 5, 26, 95–96, 164–165, 174; elite 102, 108; linguistic 102,
150–152, 160, 215 108–110, 164, 306; of communication 93,
contextualisation cues 185–186, 188–202, 139–140, 182
204–206, 227–228, 230–231, 234–235 ethnomethodology 8–9, 101, 139, 182–183, 243,
Conversation Analysis (CA) 8, 110, 139, 254, 274, 293
182–183, 243, 262, 290; feminist 258–260, embodiment 8–9, 33, 44, 71, 276
268, 305, 329
Corpus Linguistics (CL) methods: annotation 560, femininity 30, 63–64, 200, 207, 214, 217–219,
568; collocation 363, 562–563, 572–574, 583, 423, 468–470, 495–496, 535; hegemonic 5,
592–593, 602, 613; concordance 374–375, 222, 235, 423; normative 14, 72
563, 565, 591, 593, 614; frequency analysis feminism: first wave 137, 324; neoliberal 486,
583, 602; lemmatisation 592, 598; word list 505, 529; second wave 53, 137, 324–325, 327,
562; XML mark-up 602–603, 608–609, 615 409, 496, 500, 504; third wave 71, 229, 324,
Critical Discourse Analysis (CDA): 101, 329, 326, 499–500, 505
341–342, 355, 368, 485, 494, 587–588, 593; fieldwork 93–94, 102–103, 122, 154, 164–166,
feminist 329, 486, 494, 497, 505; 168, 307; ethnographic 72, 75, 96, 138, 153,
multimodal 485 164, 307, 544
Foucault, M. 137, 402, 408–410, 422, 442
deficit approach 2, 325, 391, 537–538
desire 3–4, 123, 155, 276, 450–453, 457, 460–461 gay 29–30, 37–39, 53–55, 121, 275, 339,
difference approach 2, 100, 325, 567 368–369, 374, 567, 595

619
Index

gender: as a social construct 222, 244, 261; cis- 5, 7, 327, 548, 598; non-normative 327;
52, 69, 74–75, 84, 101, 289–301; diversity normative 42, 63
69–72, 84–85, 330; dominance 219, 325; media: mass 360–361, 386–387, 389, 445, 494,
identity 30, 72, 217–218, 228, 261, 305–306, 512, 588; news 15, 168, 410, 466, 587–589;
340; ideology 103, 136–137, 142, 360; social 166, 171–172, 377–378, 390–391, 472,
inequality 220, 325, 401, 474, 524, 537; 487, 543–545
non-binary 293, 529; order 9–10, 16, 79, 219; metaphor 7, 274, 324, 344, 400, 490–491
-queer 69, 166, 328, 410; stereotypes 197–200, methods: ethnographic 121, 125, 132, 147–148,
208, 388, 468, 490; trans- 69–70, 84–85, 328, 152–153, 285; linguistic 99, 341, 343, 614;
593; see also gendering, norm mixed 12, 84; qualitative 14, 109, 114, 117;
gendering: andro- 290; binary 290, 299, 301; quantitative 84, 94, 108, 589
categorical 290, 301; cis- 290, 299, 301; #metoo 17, 148, 325, 365, 474–475
hetero- 290, 295, 297–298; non-binary 290; Middle East 159, 326, 399, 404, 423–425, 433
repro- 290 misogyny 364, 467, 472–474
Goffman, E. 4, 182, 186, 227, 231, 274, 482–484, motherhood 221, 370, 405, 409, 413–419
528, 530–531, 547 Mumsnet 409–413, 418–419
Gumperz, J. 14, 181–183, 187, 200–201, multimodality 9, 272, 301, 481481, 486–487,
227–228 491, 497, 544–545

Halliday, M. A. K. 125, 127, 183, 372, 485, 511 nationality 113, 246, 487, 553
homosociality 450–451, 453, 457, 460 nomination 411, 602–603, 610–611, 614
humour 197–198, 200–201, 207, 218, 369–370 norm: cultural 109, 112, 116, 152, 296; gender 41,
Hymes, D. 9, 165, 182–183, 275 79, 221, 282, 313, 474; gendered 214, 216–217,
221, 304–305, 316; linguistic 109, 117, 566;
identity: construction 191–192, 212–213, 219, sexual 347, 491; social 3, 227, 277, 520, 523, 577
544; gender 30, 72, 217–218, 228, 261, normativity: cis- 289–290, 301; gender 75, 255,
305–306, 340; in the workplace 212, 230; see 294, 340, 351, 474; hetero- 121–122, 125,
also performance 133, 248, 289, 328–329, 340–341, 343–344;
ideology: gender 103, 136–137, 142, 360; sexuality 289–290, 293–294, 296, 298, 301
language 97, 102, 144; of difference 489
indexicality 25, 37–38, 93, 98, 110, 214, 544, 546 observation: ethnographic 80, 189, 200, 213, 454,
inequality 16–17, 173, 219–220, 329–330, 332, 551; participant 70, 75, 94–100, 111, 138–140,
384–385, 546–547; gender 220, 325, 387, 401, 153–154, 182–183, 285, 454
468, 474, 537; social 185, 325, 332, 342, 355 online dating 405, 450, 453–455, 554
Interactional Sociolinguistics (IS) 14, 181–182,
184–185, 197, 200, 212, 226, 228–229, 231, patriarchy 2, 40, 153, 405, 438, 505
233–234 patronizing 116, 496, 564–565, 572, 577–578,
intersectionality 5–6, 40–41, 43, 56, 173, 234, 580–582
326–327, 496, 545–546 performance 4, 55, 63, 207, 213, 273–274, 285,
intersex 71–72, 245–246, 290, 328, 589, 592, 483, 530, 546–547; see also performativity
593, 595–596 performativity 4, 28–29, 37–38, 71, 273–274,
intertextuality 188, 234–235 324, 546
interview 45, 81–82, 97–98, 102, 111–112, persona 26–29, 34, 38–39, 43, 46–47, 57,
153–155, 182–183, 220, 344–345, 426 172–173, 364–365
phobia: homo- 121, 340, 344, 347, 350, 368, 374;
Lakoff, R. 2, 187, 383–384, 490 trans- 86, 121
leadership 197–198, 200–201, 387–388, 405, pitch 30, 38, 44–47, 59, 73–74, 309–310, 388
422–423, 465–467 platform literacy 167, 173
lesbian 37–38, 41, 53–54, 131, 228, 235, 341, politeness 198, 227, 582; im- 577, 582; mock
567, 590, 597 568, 572, 577–578, 580–582
LGBT 16–17, 101–102, 234–235, 243–244, 328, politics: feminist 324–325, 384; gender 473;
344, 346, 487 identity 402, 438, 544, 546; of difference 402;
sexual 42, 45, 500
marginalisation 64, 99, 133, 154, 219, positioning theory 190, 308, 412, 418
404–405, 456 poststructuralism 2–3, 229, 326, 399, 404–405,
masculinity 55–56, 63, 200, 203–205, 214–215, 437, 450, 461, 465, 496; feminist 15, 401, 409,
217–219, 326–327, 547–548, 551; hegemonic 418, 422, 424, 437–438, 447–448

620
 Index

power: gender and 108, 259–260, 598; imbalance semiotic landscape 461, 482, 488–489; linguistic
7, 43, 121, 212; relation 5–6, 96, 103, 112, 482, 488–489
402, 409–410, 438, 447–448; structure 56, 292 sexism 56, 290, 383, 472–474, 524, 529–530
practice: cultural 114, 317, 405, 543–544, 547; sexual: a- 291, 456, 487, 550, 589; bi– 53–54,
discursive 5, 93, 215, 343, 423, 438–439, 471; 73, 341, 369, 409, 574, 589, 593, 597; hetero-
524, 543; embodied 33, 44, 325; gendered 37–38, 102, 121–123, 126, 249, 289, 340, 348,
6, 79, 279, 364; institutional 125, 437, 447; 450–451, 587–588; homo- 260, 345–346, 373,
interactional 8, 301, 362–364; language 98, 567–568, 589–590, 594–595; pan- 86; trans- 5,
100, 304–305, 317, 342–343, 426; linguistic 244, 274, 589
34, 38–39, 328, 342–344, 422–423; local social order 2–4, 7–8, 26, 138–140, 145, 244,
150–151, 161; peer language 305, 307, 255, 306, 329, 488; see also gender order
316–317; semiotic 97, 481, 547; sexual 63, social semiotics 481–482, 485–486, 494, 505
150, 167, 243, 343; social 40, 71, 151–152, speech: event 115, 275, 285, 361, 364, 422, 425;
245, 274, 437–438; social media 166, 487, gay 53–54; lesbian 45; style 44, 204, 275
543, 545, 547, 550–551, 553; speech 47, 98, straight 53–54, 56–57, 76, 79
275; stylistic 29; see also communities of subjectivity 96, 153–155, 401, 410, 437–438,
practice 442–443
prosody 310, 313, 373, 458, 574
public engagement 333, 383–386, 392 text trajectory 354–355, 357–359, 365
textual analysis 143, 425, 447, 453, 550
queer: linguistics 101–102, 125, 229, 284, 328, transgression 115, 171, 173, 515, 520–521
340–342; theory 229, 328, 339–340, 343, 489 triangulation 109, 113, 117–118, 583, 602, 615

race 56–57, 64, 101, 275, 326, 402, 438, 484, variation: language 63, 150–151, 153,
503, 553 161; linguistic 34, 48, 150; pitch 37, 44;
racism 7, 64, 163, 387, 405, 447–448, 497, 524 sociolinguistic 52, 328, 583
recording: audio 33, 138, 142, 185, 244, 285, 291, violence: against women 258, 262, 266, 358, 475,
356; video 110, 114, 138, 244, 262, 277, 282– 606, 614; domestic 16, 261, 263–264, 266,
283, 291, 307, 356 332; sexual 324, 331, 356, 358, 360, 383
reflexivity 12, 94, 96, 100–102, 112–113, 168, voice: creaky 30–33; double 6, 471, 551–552
213, 219, 231–232, 284–285; self– 292, 426
women’s movement 136–137, 324
Sacks, H. 113, 182–183, 243–244, 246–247, workplace: discourse 212–214, 218; interaction 6,
264–265, 304, 309, 311 200, 208, 422; research 212, 218

621