What is Discourse Analysis?

‘What is?’ Research Methods series

Edited by Graham Crow, University of Southampton
ISSN: 2048–6812
The ‘What is?’ series provides authoritative introductions to a range of research
methods which are at the forefront of developments in the social sciences.
Each volume sets out the key elements of the particular method and features
examples of its application, drawing on a consistent structure across the whole
series. Written in an accessible style by leading experts in the field, this series is an
innovative pedagogical and research resource.

What is Online Research? Forthcoming books:

Tristram Hooley, Jane Wellens and John
What is Qualitative Interviewing?
Marriott
Rosalind Edwards and Janet Holland
What is Social Network Analysis?
What is Narrative Research?
John Scott
Molly Andrews, Mark Davis, Cigdem Esin,
What is Qualitative Research? Lar-Christer Hyden, Margareta Hyden,
Martyn Hammersley Corinne Squire and Barbara Harrison
What are Qualitative Research Ethics? What is Inclusive Research?
Rose Wiles Melanie Nind
What are Community Studies?
Graham Crow
What is
discourse
analysis?
Stephanie Taylor

L ON DON • N E W DE L H I • N E W Y OR K • SY DN EY
 Bloomsbury Academic

An imprint of Bloomsbury Publishing Plc

50 Bedford Square 175 Fifth Avenue

London New York
WC1B 3DP NY 10010
UK USA

www.bloomsbury.com

First published 2013

© Stephanie Taylor, 2013

This work is published subject to a Creative Commons

Attribution Non-Commercial Licence.

You may share this work for non-commercial purposes only, provided
you give attribution to the copyright holder and the publisher. For permission
to publish commercial versions please contact Bloomsbury Academic.

Stephanie Taylor has asserted her right under the Copyright, Designs and Patents Act,
1988, to be identified as Author of this work.

No responsibility for loss caused to any individual or organization acting on or refraining

from action as a result of the material in this publication can be accepted by Bloomsbury
Academic or the author.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.

EISBN: 978-1-8496-6905-4

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.

Typeset by Newgen Imaging Systems Pvt, Ltd, Chennai, India

Contents

Series foreword vii

Acknowledgements ix

1 Introduction 1

2 Theories and common concerns 7

3 Four examples of discourse analysis 29

4 The usefulness of discourse analysis for

social science researchers 53

5 The challenges of discourse analysis 67

6 Criticisms of discourse analysis 77

7 Summary 87

Further reading 91

Glossary 95

Bibliography 103

Index 111
Series foreword

The idea behind this series is a simple one: to provide concise and
accessible overviews of a range of frequently used research methods
and of current issues in research methodology. Books in the series
have been written by experts in their fields, following a brief, to write
about their subject for a broad audience who are assumed to be inter-
ested but not necessarily to have any prior knowledge. The series is
a natural development of presentations made in the ‘What is?’ strand
at Economic and Social Research Council (ESRC) Research Methods
Festivals which have proved popular both at the Festivals themselves
and subsequently as a resource on the website of the ESRC National
Centre for Research Methods. Methodological innovation is the order
of the day, and the ‘What is?’ format allows researchers who are new
to a field to gain an insight into its key features, while also providing a
useful update on recent developments for people who have had some
prior acquaintance with it. All readers should find it helpful to be taken
through the discussion of key terms, the history of how the method
or methodological issue has developed and the assessment of the
strengths and possible weaknesses of the approach through analysis of
illustrative examples.
Discourse analysis was an obvious choice for inclusion in the series
because it is a vital method for understanding a key aspect of social life,
namely what people say and how they say it. Making sense of patterns of
communication is by no means as straightforward as might be imagined,
for even apparently mundane everyday exchanges have hidden dimen-
sions to them that discourse analysis has the capacity to reveal. It can be
controversial to explore the ways in which things that are routinely treated
as ‘just talk’ are much more important and interesting than that, because
discourse analysis has the potential to highlight issues of power and the
creation and contestation of meaning, but the association of controversy
with a method may be taken as an indication of its capacity to go beyond

vii
viii Series foreword

comfortable but superficial understandings. Discourse analysis certainly

encourages us to think differently, and in this book Stephanie Taylor’s care-
ful exposition of the method reveals the insights that can be gained by
doing so.
The books cannot provide information about their subject matter down
to a fine level of detail, but they will equip readers with a powerful sense
of reasons why it deserves to be taken seriously and, it is hoped, with the
enthusiasm to put that knowledge into practice.

Graham Crow
Series editor
Acknowledgements

I would like to thank the many colleagues and students who have contrib-
uted to my own understanding and use of qualitative research methods,
including discourse analysis, in discussions at the Open University and
elsewhere, especially Margaret Wetherell, Martyn Hammersley, Jean
McAvoy, Karen Littleton and the members of the Discourse Group and
Qualitative Work in Progress Group. I would also like to thank Meg Barker,
Martyn Hammersley (again) and Jovan Byford for their invaluable feedback
on the manuscript in progress. This book is dedicated to the memory of Jill
Reynolds who was a friend, colleague and co-author and is sadly missed by
the Qualitative Work in Progress Group here at the Open University.

ix
1  Introduction

This book has been written as an introduction, to explain and demon-

strate some of the possibilities which discourse analysis offers for contem-
porary social research. It is for novices as well as experienced researchers
in the social sciences and related disciplines who want to read about
other people’s discourse analytic research and perhaps conduct their
own similar projects. The two major themes of the book are as follows:
discourse analysis does not refer to a single approach or method, and its
varied forms cannot be distinguished without an understanding of their
premises. The aim is therefore to help readers understand the connections
between the theoretical assumptions which a discourse researcher makes
about the social world, the research topic or problem being investigated,
the data chosen for analysis and the aspects of the data which are studied.
Any research project represents a unique combination of these features,
and indeed part of the attraction of discourse analysis is the wide range of
possibilities which it encompasses.
Chapter 1 develops an initial definition of discourse analysis and then
outlines the content of the subsequent chapters, with some suggestions for
using the book effectively as a practical guide, and building on its content.

What is discourse analysis? 

This question is difficult to answer succinctly because the term ‘discourse
analysis’ refers to a range of approaches in several disciplines and theoreti-
cal traditions. Discourse analysts in sociolinguistics, sociology and social
psychology, to mention just a few possibilities, are likely to differ in the
sources they refer to, and also, to some extent, in the problems and
research questions which they set out to investigate. This variety is poten-
tially confusing, but also positive in that it expands the possibilities for
new researchers and projects. This book provides an overview of discourse
analytic research as a rich and interdisciplinary field which continues to

1
2 What is discourse analysis?

change and develop in new directions. Later chapters will discuss some of
the theoretical background and examples of research studies. This section
begins a discussion on the meaning of the term discourse analysis.
One starting point is that discourse analysis usually refers to a research
approach in which language material, such as talk or written texts, and
sometimes other material altogether, is examined as evidence of phenom-
ena beyond the individual person. To understand this, imagine looking at
some old letters, written several decades or even centuries ago. Each letter
will of course be interesting for what it conveys about the writer’s situation,
opinions and feelings. However, it can also provide more general evidence
of society at that earlier time. For instance, passing references may suggest
what is taken for granted, including the priorities and values shared by
members of society. Some of the words used may even be offensive to a
contemporary reader, for instance, because they are linked to assumptions
about class or gender or race which have since been questioned. In addi-
tion, some of the writing may seem ‘old-fashioned’ in its style and level of
formality because there have been changes in the directness with which
people express opinions or state disagreements. If letters between both
parties to the correspondence have survived, they will provide further
insights into how people communicated at that time, including the con-
ventions which operate in particular relationships, business or personal.
In short, each letter, however private its original purpose, is potentially of
interest as evidence of social phenomena, in a way that the writer could
not have anticipated. This is the level of interpretation employed by a
discourse analyst.
To explain the variety of discourse analytic research, it is useful to distin-
guish two lines of academic work. They do not encompass all the variations
in discourse analysis and in practice, they tend to converge particularly
around the study of difference and inequality and the workings of power
in society, as some of the studies discussed later in the book will indicate.
Nonetheless, the distinction is helpful as a starting point.
The focus of the first line of academic work is the nature of language.
Traditionally, linguistics, or ‘linguistics proper’, can be defined narrowly as
‘the study of “grammar” in a broad sense: the sound systems of language
(“phonology”), the grammatical structure of words (“morphology”) and of
sentences (“syntax”) and more formal aspects of meaning (“semantics”)’
(Fairclough, 2001a, p. 5). However, sociolinguists and many other academ-
ics consider language as inseparable from its social contexts. Among many
 Introduction 3

other aspects of language use, they study differences in how people speak
(and write) which are linked to class and other social categories, or to a
particular activity, situation, role and purpose. Summarized somewhat
crudely, this line of writing and research can therefore be said to have
extended from the concerns of ‘linguistics proper’ to explore features of
language linked, again, to social phenomena. Discourse analytic research
in this line includes investigations of the details of how language varies
across contexts and can mark social difference, and how children acquire
competence in language use and, again, how that competence is linked to
identity and social difference.
The second academic line can be said to have originated in the study
of society and people as social beings, especially in sociology and social
psychology, and then developed to incorporate a focus on language. In
other words, the move has been in the opposite direction, from social
phenomena to language. For example, discourse researchers have analysed
public and private language use as a way of accessing the collective, though
not necessarily coherent, ‘world view’ of a society. Some aspects would be
the ways in which people and their activities are categorized, valued and
located in relationships of dominance and subservience. Language use
may also be analysed as one activity or practice (some would say, the most
important practice) which people engage in as part of their ongoing social
lives and relationships. Through the analysis of language and language
use, the researcher therefore builds up a picture of society and how it
functions.
The variety in discourse analytic research is also partly given by the kinds
of data used by researchers. Discourse analysts can study other forms
of representation from language, such as pictures and film, or consider
language use alongside other practices, but most work with some kind of
language data (the focus of this book). Some researchers investigate histori-
cal material, like the old letters already mentioned, but most take as their
data contemporary material related to language and communication. As
already noted, this is analysed following the principle that language pro-
vides evidence of social phenomena. For example talk, perhaps from focus
groups or people participating in interviews, will be of interest not (just)
as straightforward reporting, like witness statements. (Many analysts will
also avoid interpreting it as a direct report of the speaker’s inner thought
processes and feelings.) Instead, the analysis may focus on how the interac-
tion of speakers is shaped and constrained by its social contexts, from the
4 What is discourse analysis?

immediate conversational interaction to the larger situation of the place,

time, state of the nation and so on. The analysis may explore the styles
of speaking, and the ‘to and fro’ of the talk. It may consider the details
of how the talk is constructed, including grammatically, and the implica-
tions of the choice of one possible word or structure over an alternative.
It may consider the functions of the talk, or examine the assumptions
which apparently underlie what is said. Words, expressions and arguments
may be analysed as social and cultural resources associated with certain
social groups and contexts, recycled and modified by individual language
users in particular instances of communication. The analysis may explore
more dramatic aspects of the talk, for instance, when someone speaks ‘as’
a certain character or role (a concerned parent, an authoritative expert)
or appears to repeat a well-rehearsed anecdote, or address an audience
beyond the immediate situation. Some of these foci are associated with
particular named approaches to discourse analysis, but in many or even
the majority of studies, the features to be studied will be decided according
to the priorities of the research, drawing on the wide range of possibilities
given by previous published work.
The remainder of this book looks more closely at some of these pos-
sibilities. To assist the exploration, it will be useful to give an initial answer
to the question ‘What is discourse analysis?’ as follows: Discourse analysis
is the close study of language and language use as evidence of aspects of
society and social life. Later chapters of the book return to this definition
and extend it.

The chapters 
Chapter 2, ‘Theories and common concerns’, presents an overview of the
main theoretical traditions relevant to discourse analysis in the social
sciences. Although the focus is on key ideas, not individual theorists, the
chapter does introduce some of the ‘names’ readers may encounter in
their reading of other sources. It is intended as a starting point for further
reading which researchers can build on as they follow their own interests
and develop new projects.
Chapter 3, `Four examples of discourse analysis’, discusses four recent
articles published in academic journals, each of which presents find-
ings from a discourse analytic study. The examples have been chosen
to demonstrate the variety of discourse analytic approaches and also of
 Introduction 5

the kinds of research problems they have been used to address. Together,
the four articles refer to a range of subject fields, project designs and types
of data. The first article, by Kirsten Bell, is based on a study conducted with
people attending cancer support groups; the second, by Jovan Byford, on
an investigation of political rhetoric; the third, by Elizabeth Stokoe, on a
close analysis of talk from neighbours engaged in disputes; and the fourth,
by Ruth Wodak, Winston Kwon and Ian Clarke, on a study of meetings
in a business organization. The four articles variously discuss racism and
prejudice, health, risk and personal responsibility, morality and gender,
and leadership and consensus. The chapter provides an overview of each
research project, its theoretical grounding, the empirical work and the data
which were analysed, and the discourse analytic or discursive approach
which the researcher has adopted.
Chapter 4, ‘The usefulness of discourse analysis for social science
researchers’, begins with a general discussion of the reasons for using this
form of research and for analysing language data. The chapter also dis-
cusses two ‘half-reasons’ which are based on a partial misunderstanding of
the premises of a discourse analytic approach. The chapter then describes
some of the different kinds of data which can be analysed discursively
and considers practical aspects of obtaining discourse data. The chapter
includes sections on the collection of new data, the selection of already
existing or ‘found’ material as data and the production of transcripts.
Chapter 5 considers some of the problems or challenges faced by the
discourse analyst. These include deciding on the right discourse analytic
approach, and beginning to analyse data and develop an interpretation
and argument. The discussions of interpretation and analysis continue the
practical guidance provided in Chapter 4.
Every research approach has its critics and Chapter 6 addresses some
common criticisms of discourse analysis. For example, the chapter answers
the challenges sometimes made that discourse analysis is out of date as an
approach or is ‘just’ about words, or that it does not take enough account
of people or has limited practical applications. The chapter also discusses
some more specific criticisms, such as that discourse analysts should not
collect data through interviews.
Chapter 7 summarizes the key points covered in previous chapters and
presents a list of suggestions and references for readers who would like to
build on the practical introduction to discourse analysis which is presented
in this book. The book also includes a glossary of key terms.
6 What is discourse analysis?

Using this book 

Like other text books, ‘What is discourse analysis?’ has been written as a
guide and reference tool. Readers are likely to turn to chapters and sec-
tions which seem pertinent to their immediate concerns and practical
problems. However, the book has also been written to be read from start
to finish. The concepts and terms which are introduced in Chapter 2 are
referred to in the very detailed discussions of empirical studies presented
in Chapter 3. The discussions in Chapters 4 and 5 refer back to those stud-
ies, and to previous chapters, and the arguments presented in Chapter 6
presume some awareness of theories and concepts introduced earlier in
the book. Readers are therefore recommended to work through the chap-
ters in order, at least initially. It will also be useful to refer to the glossary of
key terms at the end of the book.

Summary 
Chapter 1 has introduced the topic of the book, offered an initial answer to
the question ‘What is discourse analysis?’ and outlined the contents of the
remaining chapters.
2  Theories and common
concerns

Introduction 
Chapter 1 proposed a definition of discourse analysis as ‘the close study of
language and language use as evidence of aspects of society and social life’.
This chapter explores the theoretical underpinnings of discourse analysis
and some of the main concerns of researchers, including the different
terms they use. The first three sections will discuss theories which estab-
lish a connection between language and social phenomena and therefore
inform discourse researchers’ arguments for the status of language data as
evidence. The fourth section discusses some relevant terminology, includ-
ing different definitions of the key term ‘discourse’. The fifth and sixth
sections introduce some of the most important concerns which underlie
discourse analytic research in different traditions, and the seventh section
outlines the links between discourse analysis and social psychology.
To introduce the first two sections, it may be useful to return to the
example of letters, introduced in Chapter 1. The discussion there of course
assumed that readers understand what is meant by ‘a letter’, even though
the use of letters is increasingly superseded by electronic communications.
A formal dictionary definition, from the Shorter Oxford Dictionary (1965),
is that a letter is ‘a missive in writing, an epistle’, a definition which might
prompt further consultation of dictionaries. A simpler definition is that a
letter is a communication on paper which is conventionally sent through
a postal system. This probably explains the ‘meaning’ of a letter sufficiently
for someone to understand what the word refers to, for instance, in
order to translate it from one language to another. However, it does not
encompass the kinds of meanings which might be of interest to a social
researcher. Consider, for example, the difference between using a letter or
an email. A letter would probably seem more appropriate to an invitation
for a special occasion. Similarly, in a personal relationship, especially one in
its early stages, a love letter would probably carry greater significance than

7
8 What is discourse analysis?

a declaration of love in an email. More negatively, in a correspondence

about a dispute, perhaps in a workplace or between neighbours, sending
a letter might raise the issue to a new level of seriousness. In short, a letter
can carry meanings of formality and occasion (the invitation), commitment
and lastingness (the love letter), seriousness and legality (the dispute).
These meanings are cultural, made familiar through the many experiences
of being part of a society, including seeing the post delivered, receiving
letters and sending them, and reading or viewing stories in which letters
mark key dramatic points. Additional meanings will attach to details like
the quality of the paper, whether the letter is printed or handwritten and
whether it is addressed to a named person or merely ‘The Occupier’. And
there are many other associations, of emotion and of images, such as a
bundle of letters tied with red ribbon, a card decorated with balloons and
so on.
These meanings and associations are not fixed. They have accumulated
over time and may now be fading because of the increasing use of electronic
communication. In addition, the meanings of particular letters will depend
on their context, including the context of a time (today, 20 years ago or a
few years into the future), the purpose of the letter, and the relationship
between the sender and recipient(s). The relevant point for this discussion
is that a letter is not just a letter but potential data for a social researcher.
It may provide evidence relevant to the exploration of society and social
phenomena such as celebrations, or changes in intimate relationships, or
the settlement of disputes, or institutionalized communication or the legal
profession: the exact focus will be given by the topic which the researcher
has chosen to investigate.
This example need not be elaborated any further but it introduces the
three approaches to meaning which will be discussed in the following sec-
tions, in relation to language. The first is the exploration of social meanings
as systems or aggregates, accrued over time. The second is the study of
meaningful practices, such as inviting, declaring, threatening and also writ-
ing, signing, witnessing, consulting a lawyer, entering a contract and many
more. The third concerns the use of language for the communication of
meanings. All three sections adopt a broad approach and at some points
juxtapose theorists who are usually associated with separate traditions of
academic work. As indicated in Chapter 1, the aim is to introduce the field
and enable readers to begin pursuing lines of interest to be followed up in
further reading.
 Theories and common concerns 9

Theorizing language as systems or

aggregates of meanings
The example of the letters introduced the notion that there are aggregates
or groups of meanings which ‘hang together’ by association rather than
logical links. The theorization of such systems of meaning is strongly
informed by the work of the French philosopher and social theorist Michel
Foucault (1926–84). His work is perhaps less academically fashionable than
it was two or three decades ago, but its influence persists, including in
work which may not acknowledge it directly. Foucault’s writings are too
wide-ranging and complex to be summarized easily. His own research was
primarily historical in the sense that he attempted to unravel the foun-
dations of contemporary knowledge by tracking connections between
ideas back through a variety of documentary sources. For example, in one
famous study, he suggested that madness acquired some of the stigma of
leprosy partly because of the chance circumstance that the same buildings
which had accommodated leprosy suffers were later used to incarcerate
people who were mentally ill (Foucault, 2001). He argued that although
the connection was subsequently forgotten, the negative association was
transferred from one set of sufferers (lepers) to the other (people with
mental illness). It influenced the treatment of the mentally ill and persisted
into contemporary thinking, given fortuitously by history yet having
important social consequences for social attitudes and practices around
mental illness.
This example also illustrates a second premise: such an aggregate of
ideas and meanings or, in Foucault’s term, ‘discursive formation’ is a form
of knowledge which is inextricably linked to the workings of society.
To understand this point in relation to the example mentioned above,
consider, first, the difference between the terms ‘mad’ and ‘mentally ill’,
and second, how the categorization of people as mentally ill inevitably
involves a chain of connections. The authority of the experts who can
ascribe such categories is linked to a justification for treating supposedly
mentally ill people in certain ways, the traditions and institutions which
provide the ‘knowledge’ underpinning the justification and the power
structures and institutions through which the treatment is administered.
The words and terms used in various contexts, officially and colloqui-
ally, are part of the chain; the language is inseparable from the social
phenomenon.
10 What is discourse analysis?

The notion of the social world as shaped by historically accrued, partially

contingent knowledge and language has further implications. It inverts
the commonsense assumption that knowledge, recorded or unrecorded,
follows faithfully from the details of the world which pre-exists it, like a
description of what ‘is’. A classification like ‘mad’ may even be seen to pro-
duce what it purportedly describes (a person different to most others?).
This is a challenge to the possibility of objective knowledge and it points
to a further premise: people do not necessarily learn by observing. Rather,
what they see is largely determined by already existing, socially circulated
knowledge. This point undermines the concept of ‘truth’, changing its
status from what ‘is’ (or was) to what is generally believed and accepted.
There can be more than one truth! It also implicates knowledge with
power since the workings of the social world will be closely connected to
the accepted truths of that world, such as which people ‘are’ right, or best,
or most important, or deserving of reward, and which have the power to
dictate or influence the accepted truths. This is discussed further in the
section ‘Knowledge, Truth and Power’.
This section has indicated one aspect of discourse analysis, that is, the
study of language as evidence of a system or formation of meanings and the
connections of those meanings to society, including the power relations
within society. However, such an analysis can be criticized as static because
it maps meanings onto language as if they do not change. This potentially
conflicts with two other premises which are particularly relevant to the
analysis of talk. The first is that meaning is ‘indexical’, that is, it depends on
and changes with context. A simple example is provided by the reference
of a pronoun such as ‘her’, ‘him’ or ‘it’. The meaning of the pronoun will be
given by the surrounding physical environment of objects and people, or
by the preceding talk. For example, ‘it’ may be something which a speaker
points to or touches, or something which is not present but has already
been made a topic of conversation.
The second premise is that meanings are constituted through what is
‘done’, that is, through practices and processes. One example would be
how information can be made ‘official’ through processes such as publica-
tion or the release of statements. Another would be in talk, the way that
a speaker’s identity may be indicated not by what is said but by how it is
said, for example, with authority or obedience. Each of these points opens
up further directions for investigation, some of which are illustrated in
the studies discussed in Chapter 3. The current chapter continues with a
 Theories and common concerns 11

discussion of theories which have contributed to the study of social activi-

ties or practices, and in particular, language use and talk as an everyday
practice.

Theorizing language as social practice 

A key writer in this tradition is Harold Garfinkel (1917–2011) who developed
the sociological field of ethnomethodology. The following short quotation
sets out its main points:
Ethnomethodologists seek to understand how we ‘do’ social life
and how ‘the properties of social life which seem objective, factual
and transsituational, are actually managed accomplishments or
achievements of local processes’. (Zimmerman, 1978, p. 11, quoted in
Stokoe, 2003, p. 320)
Ethnomethodology is therefore concerned with the study of practices
and processes. In addition, it begins with the study of what is ‘local’. It chal-
lenges macro-level explanations of social order by explaining the function-
ing of society as based in micro-level, rule-governed interactions between
people. Garfinkel famously encouraged his students to disrupt the order
of local situations, for example, by behaving at home as if they were guests
rather than members of the family. The purpose was to demonstrate the
tacitly understood rules and forms of behaviour which regulate ordinary
life practices and become obvious only when they are ‘breached’. In this
view, as people competently function in each small situation of daily life,
drawing on the knowledge and skills they have acquired as members of
society, their myriad activities combine to constitute an orderly whole.
Society is therefore seen to operate from the ‘bottom up’, although eth-
nomethodologists would reject any neat macro–micro hierarchy.
Following from ethnomethodology, talk and spoken interactions can be
analysed as orderly social activities. This is the approach of conversation
analysis (CA), based on the work of Harvey Sacks (Jefferson, 1992). It is
sometimes referred to as a form of discourse analysis and sometimes as a
separate field, also known as the study of ‘talk in interaction’. Conversation
analysts are interested in talk as a social practice and in the meanings
which are established within the immediate conversational context. The
data they analyse are usually ordinary conversational interactions (often
referred to as ‘mundane’), such as phone conversations or talk at the
12 What is discourse analysis?

dinner table. These are electronically recorded and then transcribed in

close detail, for example, to include emphases, pauses, overlapping talk
and sometimes other details like intonation. There is a preference for ana-
lysing ‘naturally occurring talk’ which would have taken place whether or
not the researcher (or recording equipment) was present, rather than an
interview set up for the purposes of the research. This kind of investigation
of talk as a social practice potentially (some would say definitely!) conflicts
with the analysis of meaning systems which is associated with the tradition
discussed in the previous section. Meanings are assumed to be established
within the context of the talk, rather than being given by society and then
imported into the situation being studied.
An analysis of talk as a social practice also challenges many established
ideas about communication and about people themselves. The analyst
studies how talk is shaped to the interactional context, to perform func-
tions within that context. To understand this, consider the kind of talk
which might be labelled ‘apologizing’. This would often be assumed, by
people other than ethnomethodologists, to follow on from some prior
mental activity: a person feels apologetic or thinks something like ‘I was
wrong’, and then expresses that feeling or thought in talk. As another
example, telling a story about something that happened in the past might
be assumed to follow from the mental activity of ‘remembering’. However,
in the practice-focussed logic of ethnomethodology and conversation
analysis, apologizing or storytelling talk occurs in response to other talk,
as part of a sequence of steps or moves like a social dance. The telling of
the story or the apology occurs at the point where it is made appropriate
by the ongoing social interaction. It is therefore understandable as part of
that interaction. In summary, what a person says can be accounted for only
with reference to interactional processes; it is not described or explained in
terms of some kind of individual and internal mental or emotional prompt.
This view has implications not only for interpretations of language but also
for models of people as language users. It has contributed to important
debates in psychology, discussed in more detail in the section ‘Discourse
analysis in psychology’.
A rather different emphasis on talk as an activity, usually associated with
the concept of performance, derives from the work of the philosopher J. L.
Austin (1911–60). He emphasized the distinction between the label given to an
activity and the words used to carry out that activity. Interrupting, for exam-
ple, may be accomplished using a variety of expressions (‘Excuse me’, ‘Just a
 Theories and common concerns 13

minute’, ‘But’), not to mention non-verbal sounds and actions (clearing the
throat, coughing, raising a hand and so on) but it seldom involves the words ‘I
interrupt’. This may seem a rather obvious point but Austin’s work is important
for discourse analysts because it draws attention to the functions of language,
especially but not exclusively of talk. These functions can be understood only
by considering its use in context.
This section and the preceding one have discussed two different
theorizations of language and meaning, in terms of systems and practices.
Together and separately they challenge many everyday assumptions about
language, including how it works for communication. The next section
discusses theories of language as communication.

Theorizing language as communication 

Communication between two people can be represented very simply
using a three-part model of sender, communication channel and receiver
(Wertsch, 1990). A piece of information which originates in the person who
is the sender, perhaps as a thought or observation, is conveyed through the
channel to the receiver. If the model is elaborated to include language as
the channel of communication, the information is first encoded or trans-
lated into language by the sender, then decoded or translated again by the
receiver. The model implies that the information is communicated without
distortion: neither the encoding and decoding processes nor the transmis-
sion in language alters its meaning in any way. The communication process
functions so efficiently that it can be ignored. This is the assumption which
operates when what someone says is assumed to be entirely intelligible as
the utterance and intended meaning of that individual speaker.
However, this model can be challenged. The problems with it become
obvious if it is applied to the example discussed earlier, that is, if the com-
munication channel consists of letters. The model would imply that the
meaning of a letter originates entirely in the writer and has no additional
meanings which are given by its social context. It would therefore discount
any meanings which arise as part of the social practices around language,
such as the style of the letter, and it would discount too any indexical
meanings belonging to the correspondence or communication itself.
These might include the kinds of emotion- or affect-laden meanings which
develop over an extended communication (such as pet names or special
occasions referred to by lovers, or a particular grievance which becomes
14 What is discourse analysis?

the charged focus of a dispute between neighbours.) The model would

imply, too, that the language of the letter has only one meaning which will
not change in different contexts, including different times.
Most, if not all, academic discourse analysts would reject the simple
three-part model, as James Wertsch (1990) discusses. The point of present-
ing it here is to draw attention to this rejection and, in doing so, to indicate
some of the interests of academics who study language as communication,
particularly sociolinguists. Their ‘social view of language’ (Kress, 2001, pp.
29–38) can be presented as a rejection of the study of languages as stable,
coherent, intact systems of words and grammar which can be understood
without reference to the contexts in which they are used, or to the lan-
guage users themselves. A social study of language would assume that a
standard named language, such as English or French, has many different
versions and variations. There are the ‘dialects’ associated with communi-
ties defined by geographical location, class, age, multilingualism, among
other possibilities. There are the differences in how people use languages
in specific contexts and relationships, and for different purposes: these
various ‘registers’ incorporate distinctive vocabulary and grammar (think
of the different ways in which children speak to their friends and to their
parents or teachers). Most of these variations carry social meanings and
values (think of the preferred speaking style for a TV news presenter, or
a government spokesperson). Some of the variations are more within
the language user’s control than others; in a specific situation (the pupil
in trouble who is talking to the head teacher) the user will make choices
about the language to be used but some variations, such as class and
regional dialects, may be relatively inflexible.
The approach known as Critical Discourse Analysis is specifically
concerned with the connections between language use and ‘dominance’
(van Dijk, 1998, pp. 367–93) and the ways in which elite groups use lan-
guage to exercise power in society. A Critical Discourse Analyst will be
interested in how a group influences the kinds of accepted ideas which
were referred to in the section ‘Theorizing language as systems or aggre-
gates of meanings’. As just one example, those who have access to the
media will be in a better position to influence which political arguments
are promoted and to establish consensus, even at the level of setting the
terms and categories which become taken for granted (as in the example
of ‘mad’ or ‘mentally ill’) with all the consequences attached to that
language.
 Theories and common concerns 15

This short overview suggests further possibilities for the analysis of

language use. Looking at a text (a piece of talk or writing or other language
use), the analyst might consider the words used and their associations,
as discussed in the section‘Theorizing language as systems or aggregates
of meanings’, in connection to the larger social environment, the power
structures and relationships within it and also the position and purpose
of the particular language user. Features of pronunciation and grammar
might mark dialect, such as a regional accent, or indicate the formality
related to a particular situation. The analyst might look at the social values
attached to variations. In popular entertainment and drama, these have
often been invoked as a basis for rather cruel comedy. For example, dialect
can be used to indicate the foolishness of a character such as an upper
class twit or country bumpkin, or an immigrant who speaks a heavily
accented, grammatically non-standard form of English. Characters can be
mocked for using register inappropriately, perhaps by being overly formal
in an attempt to look important, or overly colloquial in a situation which
requires formality. The humour derives both from how the characters nor-
mally speak, based in prejudice against certain groups of language users,
and from the social ignorance exhibited when characters are unable to
adapt their talk appropriately to a situation and purpose. The examples are
crude but they indicate some of the power attached to the variations in
a language and to language competence. This point is summarized by the
sociolinguist Gunther Kress as follows:
In critical linguistics the social is prior; it is a field of power; and
power (and power differences) is the generative principle producing
linguistic form and difference. Individuals are located in these fields
of power, but the powerful carry the day, and the forms which they
produce are the forms which shape the system. (Kress, 2001, p. 36)

Terminology 
The discussion of theories in the previous three sections now makes pos-
sible a fuller explanation of the terminology around discourse analysis. Many
writers use the term ‘a discourse’ to refer to the language associated with the
kind of system or aggregate of meanings which Foucault calls a ‘discursive
formation’. For example, a ‘discourse of education’ might refer to all the ter-
minology, theory and argument associated with education or, more usefully,
16 What is discourse analysis?

with one approach to education, such as a child-centred discourse. In some

discussions, the term ‘discourse’ may not be confined to the language but be
extended to encompass the relevant language users, the institutions and so
on. Stuart Hall summarizes Foucault’s notion of a ‘discourse’ as
a group of statements which provide a language for talking about –
a way of representing the knowledge about  – a particular topic a
particular historical moment … Discourse is about the production
of knowledge through language. But … since all social practices
entail meaning, and meanings shape and influence what we do – our
conduct – all practices have a discursive aspect. (Hall, 1992, p. 291)
Understood in this sense, a discourse pre-exists any particular occasion of
language use and is available to be drawn on or used selectively, according to
the situation and purpose. It is a ‘resource’ for talk. Analysts may refer to dis-
cursive resources or, more generally, cultural resources. Some analysts discuss
particular kinds of resources, such as a ‘narrative resource’ or an ‘interpreta-
tive repertoire’. Critical discursive psychologists (e.g., Edley and Wetherell,
1995; Wetherell, 1998) have preferred the term ‘repertoire’ to escape the
totality of Foucault’s notion of a discourse: for example, in research about
masculinity, they would be interested in the multiple and often conflicting
repertoires in play, rather than in an overarching discourse of masculinity.
Even within a specific tradition like this, the definitions of resources are likely
to vary and be reviewed for the purposes of a new study.
Somewhat differently, ‘discourse’ can refer to the use of language as a
practice, and especially to talk. For example, Norman Fairclough, who con-
tributed to the founding of Critical Discourse Analysis, defines ‘discourse’
as ‘language as a form of social practice’ (2001b, pp. 229–66). Analysts may
refer to the discourse of a user or group of users (‘politicians’ discourse’),
and also to the discourse data which are being analysed.
Following from both these uses, discourse analysis refers to the study
and interpretation of multiple aspects of talk or language use, as discussed
in the sections ‘Theorizing language as social practice’ and ‘Theorizing lan-
guage as communication’. However, the issue is complicated because some
influential writers have developed their own definitions. For example, the
sociolinguist James Gee calls discourses ‘social practices’ (Gee, 1992, p. 107)
and then defines them at greater length as follows:
Each Discourse in a society is ‘owned’ and ‘operated’ by a socio­
culturally defined group of people. These people are accepted as
 Theories and common concerns 17

‘members’ of the Discourse and play various ‘roles’, give various

‘performances’, within it. Each Discourse involves ways of talking,
acting, interacting, valuing, and believing, as well as the spaces and
materials ‘props’ the group uses to carry out its social practices.
Discourses integrate words, acts, values, beliefs, attitudes, social
identities, as well as gestures, glances, body positions, and clothes.
Discourses are ways of displaying (through words, actions, values,
and beliefs) membership in a particular social group or social
network (people who associate with each other around a common
set of interests, goals, and activities). Being trained (apprenticed) as
a linguist meant that I learned to speak, think and act like a linguist
and to recognize others when they do so (not just that I learned lots
of facts about language and linguistics). So ‘being a linguist’ is one of
the Discourses I have mastered. (p. 107)
This definition does not limit a ‘discourse’ to language but includes
non-linguistic elements (‘ways of acting’, ‘spaces and materials’, ‘values and
beliefs’, ‘gestures’, ‘body positions’). It emphasizes practice but also contains
the notion of a discourse as a resource which makes available ‘ways of talk-
ing, acting’ and so on, linked to the performance of an identity associated
with a group. That membership will raise issues of power and inequality,
for instance, around who is or not included.
Discourse analysis can therefore involve the investigation of language,
meanings, resources and practices. It may be associated with critical
language study, critical linguistics, sociolinguistics, communication stud-
ies, pragmatics, semiotic analysis, discursive psychology and, following
Foucault’s work, studies of governmentality. These various terms and fields
will not be explained in detail (an almost limitless task) but the studies
discussed in Chapter 3 refer to some of them. The next two sections of
this chapter will summarize some concerns shared by discourse analysts
working in the different traditions summarized so far.

Knowledge, truth and power 

Discourse analysis is strongly associated with critical research and many of
the issues which researchers investigate relate to the status of knowledge
or truth claims. The previous sections have presented arguments that
18 What is discourse analysis?

language is not a neutral vehicle through which knowledge is conveyed

from one person to another, because meanings will be created and
changed in the process of communication. Additional meanings accrue
historically. Because of this multiplicity, communication always involves
selection between alternative words, grammatical forms, registers and so
on, with each alternative carrying different meanings. A famous example
is that the terms ‘terrorist’ and ‘freedom fighter’ might be used to refer to
the same person. Clearly those alternatives are linked to different political
positions, allegiances and actions, for example, around how the person in
question ‘should’ be treated.
However, many choices are less clear-cut. Michael Billig and others (Billig
et al., 1988) have discussed some of the ‘ideological dilemmas’ around the
multiple meanings and contradictory consequences attached to a single
notion or situation. These can be illustrated with the example of the nation
as, supposedly, a group or community with a common territory and inter-
ests. Susan Condor (2011) shows how speakers confront a dilemma around
change and continuity in the nation, specifically, whether it is possible to
reconcile a positive view of the contemporary, increasingly diverse national
community with the narrow shared history which supposedly defines the
national character and limits the membership of the nation. Condor’s
research shows how speakers attempt to resolve the dilemma in order to
support diversity and the nation.
Obviously multiple accounts of the ‘same’ phenomenon are possible.
‘The nation’ is changing and stable, diverse and the same, new and old.
This kind of inconsistency may appear relatively easy to live with. However,
attempts to avoid it in relation to the nation are linked to political ide-
ologies of national purity and historical continuity which continue to be
used as a justification for horrendous violence. One concern for discourse
analysts is the multiple versions of what is ‘known’, the contests around
them and the power relationships implicated in these contests.
Theorizing multiplicity also brings into question the ‘objective’ knowl-
edge which is associated with science. In a famous discourse analytic
study, two sociologists, Nigel Gilbert and Michael Mulkay (1984), examined
scientists’ accounts of their experiments. In a conventional view of science,
scientists conduct experiments in order to make original discoveries which
add to what is already known. The findings can be checked or validated
by repeating the experiments and producing the same results. However,
Gilbert and Mulkay showed that the scientists’ accounts of their work,
 Theories and common concerns 19

especially in informal talk, were not consistent with this conventional view.
For example, the scientists’ talk indicated a practical conflict between using
their time to do original work or to replicate experiments. Originality was
more highly valued. The scientists would therefore describe themselves as
repeating an experiment in a slightly different way, or developing a different
explanation for the effect which has been observed; they emphasized their
own originality while suggesting that other scientists do the less interesting
work of exact replication to check previous findings. The point of Gilbert
and Mulkay’s research is not to discredit the scientists but to show that
even in this supposedly ‘objective’ field, language has contextualized social
functions. ‘Scientists employ forms of talk which enable them to accom-
plish both self-validation and the attribution of originality’ (Mulkay, 1985,
p. 145).
The psychologist Kenneth Gergen criticized the notion of objective
knowledge operating within his own discipline (1985) and suggested that
social psychologists should not ally themselves with natural scientists or
experimental psychologists. He emphasized the importance of language
for shaping meaning, arguing that people understand themselves and
the world in terms which are ‘social artifacts, products of historically
situated interchanges among people’ (p. 267). He suggested, therefore,
that psychologists should study language as part of ‘human meaning
systems’ (p. 270), rather than the world or the mental events which the
language might be purported to represent. This is now a principle of
discursive psychology, discussed in the section ‘Discourse analysis in
psychology’.
Many discourse analysts, including those associated with Critical
Discourse Analysis, investigate the versions of knowledge which have
become accepted as truth (i.e., as if there is no other version) and which
advance the interest of particular groups in society. Their interest is partly
in the processes through which ‘truth’ is established. The analysts may look
at how language is used to present and perpetuate a version, sometimes
referred to as an ideology, so that it comes to be taken for granted, or how
persuasive arguments may be advanced. ‘If the minds of the dominated
can be influenced in such a way that they accept dominance, and act in
the interest of the powerful out of their own free will, we use the term
hegemony’ (van Dijk, 1998, p. 372) One issue here will be, of course, who
is doing the arguing and how they can make their arguments heard, for
instance, because they have access to news media.
20 What is discourse analysis?

Following Foucault, some researchers have explored the origins and

implications of taken-for-granted ideas and categorizations. Such research
potentially undermines established authority and institutions. It will be
useful here to look at the example of the work of Nikolas Rose (1989; 1996)
which employs Foucauldian theory as a basis for research on the history
of official interventions in family life in the United Kingdom, including
those that have taken place through health and educational policies. For
example, he investigated how the discourses of education associated with
compulsory schooling gave rise to new identities, such as the identity
of a bad learner or child with learning difficulties (Rose, 1989). His work
indicates that for a bad learner to exist, there needs to be a framework of
norms and expectations about good or normal learning. This in turn rests
on a set of institutions, an education system, which defines and monitors
such learning, and also on a legal system which makes participation in the
education system compulsory. It has implications which extend into other
areas of life. For instance, the notion of ‘good’ parenting expands to involve
giving attention to a child’s educational success, encouraging children to
learn to read and so on.
Rose’s work is an example of the study of governmentality, of how
power works indirectly in contemporary social worlds, often referred to as
neo-liberal democratic societies. The argument is that people have come
to govern themselves, regulating their own behaviours in order to achieve
outcomes (such as successful school careers for their children) which they
have learned to believe are desirable. Rose traces the complex processes
through which people come to monitor and modify their own behaviours
in order to comply with social norms, and the role of knowledge within
these processes. He coined a new term, the ‘psy disciplines’, to refer to
psychology and related fields which formulate the knowledge which drives
many of these processes.

Culture and identity 

Another important group of concerns for discourse analysts relate to cul-
ture and identity. Culture might be defined as the social knowledge which
is distinctive to a particular society as a whole, or to groups within it, and
is therefore a source of identity. This overlaps with James Gee’s definition
of a discourse, quoted earlier, which relates a discourse to a group almost
as a form of insider knowledge. Extending this idea, the group might be
 Theories and common concerns 21

assumed to share a culture or body of knowledge which confers a specific

worldview (e.g., think of the different discourses of settlers and indigenous
people in a colonized country: see Dixon et al., 1997, for an analysis which
considers these). The group might also share discourses as language prac-
tices, such as ways of speaking, which mark them as members, to each
other and outsiders. These assumptions indicate several possible direc-
tions for research, into the discourses which distinguish particular groups
and create identities, the ways in which discourse practices are acquired
as part of the process of learning and becoming part of the group and the
conflicts which different discourse practices may give rise to, for example,
when ways of speaking are misunderstood, or prompt discrimination.
However, these assumptions also raise a number of issues which new dis-
course researchers need to consider.
One concerns the boundaries of cultures. It is easy to exaggerate the
distinctiveness of the discourse of, for example, women and men, or young
people and old. People do not live in neatly bounded, exclusive communi-
ties with their own entirely separate languages and social knowledge. They
understand each other and, even more importantly, they are members of
multiple groups simultaneously, or, to put it another way, they have mul-
tiple identities. It is therefore illogical to interpret language use as direct
evidence of group membership. For example, when a woman speaks, she
may not be speaking as a woman, or as a citizen of her country, or as a
daughter, or mother, or member of her profession, even though she has
a claim to all of those identities and more. In addition, even if she intends
to speak as a woman, she is not necessarily representing the view of all
women, or national citizens, or whatever. This is a caution against inter-
preting a participant as the ‘voice’ of an identity. Discourse analytic studies
of identity have employed several alternative approaches, separately or in
combination.
One involves the analysis of the shifting identity positions which
are taken up in talk. Within the different situations, interactions and
relationships which make up people’s social lives, they occupy different
‘positions’, each with an associated point of view and interests (Davies and
Harré, 1990). Some positions are conferred by others, as when a person
in authority speaks ‘as’ the authority, thereby positioning another person
as a subordinate. Other positions are more actively taken up or claimed
by people themselves. Discourse analysts study ‘positioning’, defined
by Bethan Benwell and Elizabeth Stokoe (2006) as ‘the process through
22 What is discourse analysis?

which speakers adopt, resist and offer “subject positions” that are made
available in “master narratives” or “discourses”’(p. 139). As indicated by the
references to multiple actions (adopting, resisting, offering), this kind of
analysis focuses on practices. It might extend to consider other actions
including claiming and contesting. The term ‘identity work’ is often used
to encompass these active negotiations around available and aspired to
identities.
Identity may also be discussed and analysed in terms of performance.
This concept has been mentioned already with reference to the philosopher
J. L. Austin. Austin’s work is cited by the feminist philosopher Judith Butler
who introduced the notion of gender identities as performance. In other
words, she argued for a shift from a conceptualization of being a gender to
doing or performing it, ‘though not a doing by a subject who might be said
to pre-exist the deed’ (Butler, 1990, p. 33). In other words, the person does
not pre-exist the performance, like an actor playing a part, but is made
or constituted through the practice or process of performance. Butler
suggests that performance is not a once-and-for-all accomplishment but
involves ‘a stylized repetition of acts’ (p. 179) which constructs someone’s
gender identity. Butler’s primary purpose is political; she is constructing
a basis for feminist politics which does not assume that ‘women’ share a
single essential or ‘foundationalist’ identity (p. 189). Her work is theoretical
rather than empirical but the concept of the performance of identity is
central to many discourse analyses, including studies of gender. As with
ethnomethodology, the focus is therefore on practice, on ‘doing’ rather
than ‘being’ (a particular kind of person).
A somewhat different discourse analytic focus involves the investigation
of the discursive or cultural resources which are associated with a particular
identity. These can be considered as setting a range of possibilities for an
identity as, for example, a man (Edley and Wetherell, 1995, 1997) or a single
woman (Reynolds, 2008). The analysis is usually linked to positioning and
the ways in which possible identities are taken up, resisted or otherwise
negotiated.
Implicit in all of these accounts is the notion that identities, and social
worlds, are emergent. People and their lives do not follow machine-like
cycles or repetitions (despite Butler’s reference to a stylized repetition).
Instead, they are understood to be part of an ever-ongoing flow in which
actions and interactions produce novel circumstances and situations.
It may be possible to look back and trace the pathways which led up to
 Theories and common concerns 23

the present, but the future is always uncertain, unfolding or hatching out
of the present in unforeseeable ways. People and their lives are always in
the making, never finalized or wholly predictable.

Discourse analysis in psychology 

This section is included because although discourse analysis has been
widely used across the social sciences, and in related disciplines such as
sociolinguistics, education and gender studies, it has a special status in
psychology. Many of the key texts on discourse analysis as an empirical
approach have been written by social psychologists. In addition, discourse
theories have led to the development of a new sub-discipline, discursive
psychology. This final section introduces some of the main debates associ-
ated with discourse analysis and discursive psychology.
An early and highly influential text, Discourse and Social Psychology by
Jonathan Potter and Margaret Wetherell, first published in 1987, challenged
established quantitative and qualitative methods for researching attitudes
and opinions (Potter and Wetherell, 1987). Potter and Wetherell criticized
the standard questionnaire design in which the respondent chooses an
answer from a scale (‘Agree strongly’, ‘Agree’, ‘Disagree’, etc). They also criti-
cized less structured interviews in which participants are invited to state
their opinions on an issue. Their arguments followed from the two broad
premises which have already been introduced in this chapter. The first is
the assumption that meaning is social, derived from larger discourses or
formations rather than reflecting the mental activities of individual speak-
ers. The second is the conceptualization of talk and other communications
as orderly social practices, shaped by the communicative situation (e.g., an
interview) rather than, again, by the particular individuals who are party to
the communication. Following these ideas, Potter and Wetherell rejected
the notion of an attitude as somehow fixed and internal to a person, ena-
bling translation into a choice on scale or into a neat statement of opinion.
A further premise here is that talk or language (or other representations)
cannot be a simple vehicle for an opinion which already exists: the medium
inevitably shapes the message, or even creates it.
These arguments, and in many cases Potter and Wetherell’s own work,
continue to be important for discourse analytic research. They inform
investigations of communication and the situated meanings which are
constructed, and co-constructed, within the immediate context of an
24 What is discourse analysis?

interaction (such as a conversation) and the action of communicating.

Following these arguments, many discourse analysts avoid treating talk as
evidence of the fixed opinion or thought processes of individuals. This dis-
tinguishes their work from most other social research. Instead, they analyse
either groups of meanings as discursive resources (as already noted, these
can be labelled in different ways, for example, as ‘discourses’ or ‘interpre-
tative repertoires’) or else the social practices through which meaning is
constituted, especially the social practices of talk. Many analyses combine
these foci.
Potter and Wetherell’s work has also made an important contribution
to ongoing discussion, within psychology and in the social sciences more
generally, about the nature of the person and, relatedly, about identity
(‘who I am’). This discussion generally rejects what might be called a com-
monsense theory of the speaker or language user, that is, as an individual
container for an internal mental and emotional machinery which emits
or expresses outputs into the social world. One project for discursive
psychologists has been to re-interpret ‘mental’ processes as social activi-
ties. By considering, say, remembering as a kind of talk rather than as the
expression of a cognitive process, discursive psychologists have challenged
established theories and understandings of the person.
Discursive approaches also tend to challenge our sense of ourselves
as free agents. One contrasting theory is of the person, or subject, as
subordinate (or ‘subjected’) to the workings of power in society. This
suggests that the position which a person occupies within the power and
knowledge systems discussed by Foucauldian scholars (a ‘subject posi-
tion’) determines who he or she is. Identity is therefore given by external
circumstances and situations rather than internal aspects of an individual.
Nikolas Rose’s work, mentioned earlier in this chapter, explores this mak-
ing or ‘subjectification’ as a complex and extended process (1989; 1996). In
his words, subjectification involves ‘apparatuses, practices, machinations,
and assemblages’ including aspects of organizational and institutional life.
He suggests that the process produces an illusion of freedom and agency.
People learn to discipline themselves to act in certain ways yet believe that
they are freely choosing to do so. As noted previously, Rose has explored
the subjectification associated with education together with the resulting
forms of behaviour and monitoring through which, as he describes it,
people ‘govern’ themselves. In a more recent example of research which
 Theories and common concerns 25

employs similar concepts, Rosalind Gill (2008), analysing advertisements,

has discussed the sexual subjectification by which an identity for a young
woman as agentic or empowered is tied to certain demanding require-
ments, to be ‘beautiful  … sexy, sexually knowledgeable/practised and
always “up for it”’ (p. 35).
A different theory of identity and the subject is associated with social
constructionism and (some areas of) discursive psychology. This is the
concept of a distributed self, originally proposed by Jerome Bruner
(1990). The underlying argument is that people are inseparable from
the contexts in which they function, including the contexts of their
relationships with others. Visually, this might be represented as a shift
from a neat diagram of a person as contained within the outline of the
body, to multiple images of the same person in different situations, all
superimposed on each other, with the outline of the body blurring into
the different backgrounds. The implications for psychology are that the
cognitive processes associated with the container model of the person
are re-interpreted as occurring across the whole range of social practices
which make up a person’s life. An example of how this might occur was
indicated in the earlier discussion of ‘remembering’ and ‘telling a story’.
As one writer summarizes it, in this view the mind is ‘immanent in dis-
course practices’ (Herman, 2007, p. 308), ‘spread out as a distributional
flow in what participants say and do’ (p. 312) and located in all of their
‘socio-communicative activities unfolding within richly material settings’
(p. 308). This striking idea has been hailed as part of a ‘second cognitive
revolution’ (Harré and Gillett, 1994) because it assumes that ‘discursive
phenomena, for example, acts of remembering, are not manifestations
of hidden subjective, psychological phenomena’ (p. 27). It can be linked
to classic observational empirical studies of work which investigate the
combined functioning and capacities of a technological system and
the worker within it, as a single unit (e.g., Goodwin, 1994). There are
similarities, and some direction connections, to research in the areas of
distributed cognition and Actor Network Theory.
Discursive psychological theories of talk as a social practice generally
assume that the person is active and to some extent operating independ-
ently or agentically, rather than following the dictates of the larger situa-
tion or society. As already outlined, a speaker is assumed to be positioned
by others and by society as having a certain identity, but also actively to
26 What is discourse analysis?

position her or himself, for example, by choosing between available subject

positions (in talk); in other words, identities are assumed to some extent
conferred and to some extent actively claimed and contested. This is a key
assumption in critical discursive psychology (e.g., Wetherell, 1998; Edley,
2001; Seymour-Smith, 2008).
Working from a different theoretical position, the social psychologist
Wendy Hollway has criticized the concept of the subject position as
inadequate (Hollway, 1998). She notes that people occupy many different
subject positions, yet the identities which these make available do not
carry equal importance. Some additional explanation is needed for the
emotions attached to some subject positions. Bringing together discourse
and psychoanalytic theories, she explains this in terms of ‘investment’.
Some discourse analysts discuss this issue in different terms. For example,
in a narrative analysis it can be interpreted as a question of continuity and
why some identities persist beyond an immediate, transient situation (see
Taylor 2010). An additional point to note here is that in psychoanalytic
theory the term ‘subject’ is associated with a sense of self and a subjective
(i.e., not objective) view of the world, in contrast to the notion of people
as ‘subjects’ because people are shaped by or ‘subjected to’ the workings of
power in society. The work of Hollway, with Tony Jefferson, has contributed
to a psychosocial research approach (Hollway and Jefferson, 2000).
There is no agreement on a resolution to these various problems around
a theorization of identity or the subject in discourse theory. However,
discursive approaches may be brought together with other theories, as
in Hollway and Jefferson’s work. As another example, the self-actualizing
identity project proposed by theorists of reflexive modernization, such as
Anthony Giddens, can be understood and explored in discursive terms (see
Reynolds et al., 2007). The original theory proposes that in contemporary
or late modern Western societies, a process of individualization oper-
ates through which each person shapes ‘who I am’ as their own ongoing
and reflexive identity project. A discursive psychologist approach draws
out the multiple possibilities and also the conflicts and constraints and
negotiations involved in the process of identification. A further example is
the work of Scharff (2008; 2011) which variously brings together discursive
psychology with ethnomethodology, and with performativity and affect
theory.
 Theories and common concerns 27

Conclusion 
This chapter has discussed a wide range of ideas in order to provide an
overview of the main theories and issues associated with the field of dis-
course analysis. It is therefore not possible to produce a unified summary,
and nor would it be especially appropriate since discourse analysts do not
agree on every point or follow a single composite approach. Nonetheless,
for the purposes of the book, it will be useful to expand the definition given
at the end of Chapter 1, as follows: Discourse research involves the analysis
of language data as evidence of social phenomena, theorizing language as
communication, practice or selective constructions derived from accrued
social meanings.

Summary 
Chapter 2 has introduced some of the theories and issues which shape
different approaches to discourse analysis.
3  Four examples of discourse
analysis

The previous chapters have discussed discourse analysis in general terms.

This chapter becomes more specific, looking at four published examples
of social research to show how discourse analysis is used by researchers.
The examples are journal articles which have been chosen to encompass
a variety of projects, problems, forms of data, analytic approaches, theo-
retical traditions and disciplines. All four fall under the broad umbrella
of discourse analysis as it is defined in this book. Some of the researchers
use that exact term to describe their work, while others refer to discur-
sive practices and constructions, and the analysis of rhetoric, ideological
dilemmas and categorization, following some of the theoretical traditions
discussed in Chapter 2.
The account of each article begins with an outline of the research project.
It sets out the substantive area and problem being addressed and discusses
the major claims or findings. It also describes the data collection and
analysis and discusses the analytic concepts which are employed and some
important premises of the research. Of course any single study will never
be a sufficient model for future research but each of the studies discussed
in this section does provide a useful illustration of how a researcher used
discourse analysis in empirical work. Readers who are interested in finding
out more about the research studies may want to refer back to the original
article and to other work by the same academics, although the summaries
below do provide enough detail for the introductory purpose of this book.

Discourses of health and illness 

Example 1
Kirsten Bell (2010), ‘Cancer survivorship, mor(t)ality and lifestyle
discourses on cancer prevention’, Sociology of Health and Illness 32(3):
349–64.

29
30 What is discourse analysis?

The first article has been chosen as an example of discourse research on

health and illness. It is also a discourse analytic project which combines
different methods of data collection, interviewing and observation and
different forms of data, including transcribed talk and the researchers’
fieldnotes. Another point of interest is that the data were originally col-
lected in two different languages.
Discourse research on health and illness presents an interesting chal-
lenge to ‘commonsense’ in that these are being considered as social rather
than straightforward biological issues. The author of the article, Kirsten
Bell, conducted an ethnographic study with people who were attend-
ing support groups after being diagnosed with cancer. The research has
obvious applications, for example, in a medical support context, but it is
primarily a sociological study.
Bell’s interest was in how cancer patients and cancer survivors engage
with lifestyle discourses about ‘the importance of lifestyle factors (par-
ticularly excess body weight, a poor diet and a sedentary lifestyle) in
contributing to cancer and their role in managing disease recurrence or
progression’ (p. 351). These lifestyle discourses can be found in scientific,
medical and health sources (‘scientific literature’, p. 349; ‘public health and
health promotion discourses’, p. 350). In another version, they are part of
‘complementary and alternative medical therapies’ (CAM) (p. 350). The
discourses will be recognizable to most people: they can be regarded as
part of commonsense or shared cultural knowledge, not only in Canada
where the research was conducted but also in similar affluent Western
societies. In addition, Bell notes that people who attend support groups
have been found to develop ‘collective understandings’ (p. 351) so that
these discourses may be regarded as part of the more local knowledge or
culture of the groups themselves.
To conduct the research, Bell attended three cancer support groups
for English-speaking people with different cancers and also interviewed
some of the group members individually. A second researcher on the
project attended a group conducted in another language, Cantonese, and
conducted interviews with some of its members. Because the researchers
were participant-observers in the groups, this is described as ‘ethno-
graphic’ fieldwork. The inclusion of the Cantonese speakers was intended
to capture evidence of cultural variation, with issues this might raise. One
example which Bell found was that advice on nutrition and diet in some
respects conflicted with ‘Chinese dietary practices’ (p. 356).
 Four examples of discourse analysis 31

The research project combined three kinds of formal data. The first was
the researchers’ field notes, which were collected over quite an extended
period (eight months) of their attending the support groups. The second
was the transcripts from the recorded interviews, and the third the writ-
ten notes from the interviews. The extracts presented in the article are
obviously a very small part of the data. For the project as a whole, Bell
analyses field notes from eight months of observations by two researchers,
and also the recordings, transcripts and additional notes from the indi-
vidual interviews. Some of this material would have existed in duplicate,
in the original Chinese and an English translation. In addition, Bell refers to
medical literature and other background material which provided relevant
background to her analysis, including many other studies from both medi-
cal and social research traditions. Bell used a software package to manage
her data. She notes that this was used only in the initial stages of coding,
presumably to organize the data into codes or broad categories.
The article includes extracts from transcripts and also from field notes,
as in the following example:
Sandra volunteers that in her own case she thinks that it [the cause
of her cancer] was genetics but admits that it may also be due to
the fact that she was overweight. Daphne expresses surprise at this
[Sandra is on palliative chemotherapy and has lost a considerable
amount of weight since her diagnosis with metastatic cancer] and
Sandra responds, ‘Oh, you should have seen me, I was a real porker!’
She then looks down at her stomach and sighs that she is still
overweight – pinching her gut with her fingers to demonstrate her
point. (p. 353)
This account is written by Bell and includes background information,
about Sandra’s therapy and weight loss, which Bell has presumably col-
lected in an interview or through attending previous meetings. The field
notes describe an incident with details which indicate that they are the
researcher’s interpretation and summary. For example, the expression ‘a
considerable amount of weight’ (emphasis added) indicates a value judge-
ment and the descriptions that Sandra ‘volunteers’ and Daphne ‘expresses
surprise’ indicate not only what these participants did (they spoke) but
also the researcher’s interpretation of their feelings. The field notes are
fairly full and were almost certainly based on a combination of rough notes
taken during the meeting and the researcher’s memory of what happened;
32 What is discourse analysis?

ethnographic researchers typically write up their field notes as soon as

possible after observing a situation. As the extract indicates, Bell is using
as evidence not only the words but also the feelings and actions of her
participants. This contrasts with some other discourse approaches which
retain a strict focus on the words used (see Example 3, by Elizabeth Stokoe,
in this chapter).
The discourses which Bell discusses are ones that she has found in
her larger data set and labelled, sometimes following on from previous
researchers who had identified similar patterns or features (the analysis
presented in Example 2 in this chapter, by Jovan Byford, similarly builds on
the findings of previous research). However, other analysts might focus on
different divisions and use different labels. For example, it might seem obvi-
ous that discourses of CAM are different to scientific medical discourses,
but for Bell’s research the more important point is that both reinforce
lifestyle discourses. The discourses also reflect the analyst’s rather than the
participants’ interpretation; Bell’s participants would not necessarily divide
or label the discourses in the same way or even be aware of them.
The lifestyle discourses discussed by Bell link disease management, sur-
vival and (continuing) remission to certain behaviours and ways of living,
such as eating healthily and doing sufficient exercise. The discourses there-
fore attribute cancer, at least partly, to a patient’s failure to take sufficient
care of her or himself. By implication, they potentially position patients
negatively, as to blame for their own diseases. Bell herself does not make a
judgement about the importance of lifestyle factors in preventing cancer
or cancer recurrence. Her interest was in how the people who attended
the support groups responded to lifestyle discourses.
The analysis explores the participants’ ‘complex and ambivalent engage-
ment’ (p. 349) with the lifestyle discourses. Bell notes that the discourses
are not straightforwardly accepted. Instead, the participants sometimes
resist and sometimes selectively take up the advice implicated with the
discourses. For example, Bell observed that the people who attended the
support groups sometimes made jokes about the advice they received
from visiting speakers who advised them on nutrition. However, they
would also talk in terms of lifestyle factors when they described steps that
they themselves were taking to prevent recurrence of their cancer, such as
eating carefully or going to the gym. Bell also notes some gender variation
in how these discourses were taken up, with women more likely to refer
to advice on diet and the importance of nutrition, and men more likely
 Four examples of discourse analysis 33

to discuss exercise. She also notes that there was some resistance through
alternative discourses. For example, she cites a participant who suggests
that eating what you like is part of a decision to enjoy living in the moment;
she is countering a lifestyle discourse with another discourse which might
be called ‘living life to the full’ (although Bell does not use that label).
Lifestyle discourses are individualistic in that they present health as the
responsibility of the individual person, in contrast, say, to environmental
discourses which might link cancer to problems on a larger social or geo-
graphical scale, such as pollution. The lifestyle discourses also link health
advice to morality, to notions of being ‘good’ or ‘bad’, and to prescriptions
about how people should discipline themselves to behave: it is noticeable
that the advice tends to present healthy options as ones which people are
likely to find less attractive than the unhealthy ones. Bell connects the
lifestyle discourses to processes of subjectification, discussing how the dis-
courses position people as certain kinds of subject, that is, as citizens who
carry responsibility for their own health and illness. Following the work of
Nikolas Rose, discussed in Chapter 2, this research is in the Foucauldian
tradition and develops the concept of the self-governing, self-policing
‘neoliberal subject’. Bell associates lifestyle discourses with ‘Neo-liberal
rationality (which) emphasizes the entrepreneurial individual who is called
upon to enter into his or her own self-governance’ (p. 350). Following from
this ‘rationality’, there is a potential implication that people who become
ill are bad citizens because they have brought the illness on themselves by
not managing their lives properly. Bell suggests that the patients who are
her participants are to some extent aware of being positioned as respon-
sible for becoming ill. She describes this implication and discusses how
they variously accept and resist this positioning. She describes how they
talk about ‘the sense of guilt, blame and judgement that such discourses
produced’ (p. 360), and also how they resist these discourses by drawing
on alternative explanations which would not make the illness their own
responsibility, such as genetic or environmental causes of cancer.
Social researchers who are not discourse analysts often approach partici-
pants as informants on their own experiences and feelings and opinions,
using talk collected in interviews and focus groups. A discourse analytic
study like Bell’s similarly treats the talk and the researchers’ observations as
straightforward reporting. Its distinctive feature is the additional analytic
concepts it employs, such as ‘discourses’ and ‘governance’. Unlike many
discourse analysts, Bell is not particularly interested in the exact words
34 What is discourse analysis?

used by participants or the details of their interactions. She employs the

concepts of ‘a discourse’ and ‘governance’ as operating in meanings and
behaviours above or beyond the level of talk.

Learning from this example 

The article by Kirsten Bell presents an example of the following:
• a discourse analytic study of health and illness;
• an analysis of discourses which refers to the work of Foucault, linked
to the work of another theorist, Nikolas Rose, and the discussion of
governance or governmentality;
• a project which combines discourse analysis and ethnography;
• a relatively large-scale research study involving more than one
researcher;
• discourse analysis of several forms of data in combination;
• a research problem or focus which was not the starting point for
the project but emerged in the fieldwork, partly as a result of the
researchers’ observations and partly because the participants them-
selves referred to it (this kind of emergent focus is typical of both
ethnographic and discourse analytic research);
• an analysis which treats talk as a report of participants’ feelings and
experiences;
• an analysis which approaches participants as active, in sometimes
resisting dominant or hegemonic discourses, and also sometimes
inconsistent. This is a complex picture.

The construction of prejudice 

Example 2
Jovan Byford (2006), ‘Distinguishing “anti-Judaism” from “anti­
semitism”: Recent championing of Serbian Bishop Nikolaj Velimirović’,
Religion, State & Society 34(1): 7–31.
This second article has been chosen as an example of interdisciplinary
research which brings together history, political science and social psy-
chology. It is also an example of discourse analysis conducted across two
languages. The project presented in the article is part of a long tradition
of discourse analytic research into prejudice and racism. There is now a
 Four examples of discourse analysis 35

significant body of findings on forms of talk and argument through which

social inequalities are rationalized and perpetuated. Researchers have
investigated the contemporary language and language practices which
attach a negative value to certain groups or categories of people. For
example, in a study of ‘white’ New Zealanders talking about indigenous
(Maori) New Zealanders, Wetherell and Potter (1992) noted how talk about
‘culture’ could be used instead of direct references to ‘race’. The central
argument of such discourse analytic research is that these new formula-
tions exclude people without violating contemporary social taboos against
the open expression of prejudice. In other words, these kinds of talk and
writing function as negatively as the more explicit ‘biased, prejudiced and
racist’ talk which is now seen as ‘inconsistent with the general values of
tolerance’ (van Dijk, 1992, p. 115).
Discourse analysts have also explored the kind of political rhetoric
which attempts to redefine the nation in order to exclude a negatively
valued group. This may be done by constructing a version of history
which appears to justify the exclusion, for example, through reference to
past conflicts, or by associating an authentic national identity with only
a limited part of the current national population. Such rhetoric is part of
the ongoing exercise of power in society, inextricably linked to domina-
tion, exclusion and sometimes violent oppression. The article analyses an
example, looking at details such as how people are categorized and valued,
the justifications for the valuing and the construction of arguments and
national histories. A central premise of research of this kind is that what is
real or ‘true’ cannot be separated out from the many ways it can be, and
is, described. This is not a denial that there is an objective world ‘out there’
but an assumption that such a world is always viewed and interpreted
from a particular perspective and set of interests. There are multiple ver-
sions of reality. Observers, including researchers, are insiders with a partial
view. Discourse analysts do not try to check the truth of any version or
claim but try to understand how it is constructed. By showing that it is
a construction and interpretation, they therefore open an apparently
authoritative and ‘truthful’ claim to contestation.
The focus of Byford’s research is contemporary accounts of a historical
figure which potentially function as exclusionary rhetoric against a racially
defined group (Jewish people). The article introduces Bishop Nikolaj
Velimirović (1880–1956), who is venerated as a spiritual leader, religious
philosopher and theologian. He is respected both within the Serbian
36 What is discourse analysis?

Orthodox Church and in Serbian nationalist political circles. He was can-

onized by the church (i.e., officially named as a saint) in 2003. What makes
this an interesting case, from the perspective of research on prejudice, is
that the veneration of the bishop persists in contemporary Serbia in spite
of a number of controversies surrounding his personal biography, the
most important of which is that he was virulently anti-Semitic. The project
looked in detail at the various discursive and rhetorical strategies that
Velimirović’s supporters (both from within the church and outside it) use
to avoid the controversy, deny that the bishop was anti-Semitic or in some
instances even excuse his stance towards Jews.
Byford’s project is a case study, focusing on a specific situation and con-
text but making claims which have a wider importance and application.
First, he discusses the national context, Serbia, as an example of an Eastern
European nation-state which is in transition from communism and a
possible site of the rise of the politics of the far right, in a revival of the
fascist movements of the period before World War II. The case is therefore
located within the larger scale of historical and political change. Second,
he discusses the rhetoric around this case as examples of patterns of argu-
ment, justification and denial which have been employed more widely,
both in racist and prejudiced talk in more local contexts (the concern of
many social psychologists who analyse discourse, as noted above) and in
academic and political arguments around anti-Semitism, the particular
form of racism which is invoked in the case.
The data for the study were all collected in a relatively short period of
four months, centred on a point of celebration of the key figure, Bishop
Velimirović. The main data are audio recordings of interviews or ‘conversa-
tions’ between the researcher and 12 Serbian public figures, conducted in
July and August 2003, shortly after the bishop was canonized. An unusual
feature of the research is that these participants are not anonymized; they
are identified ‘by name, rank and occupation’ on the grounds that they are
all public figures. However, they are not discussed as individuals but for their
status as influential ‘memory-makers’ and contributors to political debate.
Another form of data is published and broadcast material in the public
domain. Byford analyses a range of texts (‘newspaper interviews, com-
mentaries, editorials, serialized texts, debates and readers’ letters as well
as a small amount of radio material’) which had been produced in the
preceding months (May and June 2003). The project therefore combines
collected and ‘found’ data consisting of both spoken and written language.
 Four examples of discourse analysis 37

The analysis draws on the researcher’s knowledge of the context, including

its recent history. The original data are in Serbian and most of the analysis
was conducted in that language (the researcher is bilingual). The article is
in English and data extracts are presented in translation but sometimes
also with the original Serbian, for example, when details of wording or
grammar in that language are relevant.
Part of Byford’s analytic approach is to examine his data for features of
talk and argument (‘rhetoric’) which have been noted by previous analysts.
For example, he shows how some speakers set up a contrast between an
extreme position and a supposedly more moderate and, by implication,
excusable claim. One common strategy involves creating the distinction
between ‘real’ anti-Semitism (of the kind propagated by the Nazis), and
‘biblical’ anti-Semitism (a ‘critique’ of Jews found in parts of the Christian
Bible, including the allegation that Jews killed Christ) with the latter being
presented as acceptable. By scrutinizing the rhetoric of Velimirović’s
supporters, and situating it in the appropriate historical and ideological
context, the research suggests that the distinction between the two sup-
posedly different forms of anti-Semitism is in fact false.
Why is this important? Because, Byford argues, ‘this exercise in the
redrawing of the boundaries of antisemitism can play an important role
in the denial of prejudice’ (p. 17). The insistence that there is an ‘accept-
able’ form of anti-Semitism helps to perpetuate anti-Jewish prejudice by
presenting it as an acceptable and inevitable aspect of the Christian faith.
Like the references to ‘culture’ studied by Wetherell and Potter (1992), the
arguments employed by Velimirović’s supporters are a contemporary form
of prejudiced talk. Such talk can make negative categorizations of a group
appear excusable, unproblematic and, potentially, normal, acceptable and
justified, with all that can follow from that. Byford notes that there are
‘increasing antisemitic incidents in Serbia’ and suggests that the contem-
porary celebration of Velimirović is ‘the most powerful ideological source
of anti-Jewish prejudice in Serbian culture from which contemporary
antisemitism derives much of its legitimacy and authority’ (p. 27).
This is therefore a wide-ranging study which refers to social, political
and intellectual activities taking place over an extended historical period.
It traces connections between political and religious ideas, political pur-
poses and activities and the talk and writing of specific individuals, such
as politicians and journalists. It illustrates the possibilities of a complex
‘multi-layered’ analysis of broadcast talk and political publications within
38 What is discourse analysis?

national and institutional contexts, including the context of the Orthodox

Christian church. It also moves between different levels of data, linking
‘micro’ details of talk and written language (the use of particular words like
‘just’); larger discourses of religion and nation; contemporary and historic
identities, including both group and individual identities, and ‘macro’
accounts of history and politics. The analysis builds on a theorization of the
working of power in society by which competing claims to authority, enti-
tlement and even social presence are assumed to be played out in talk and
argument, sometimes as the justification for other actions. Byford inves-
tigates anti-Semitism as a ‘lived ideology’ (a concept he takes from Billig
et al., 1988), that is, as the power relations which are in an ongoing state
of renegotiation, contest and reinforcement within the everyday activities
and communications of the people who make up society. Byford’s interest
is in the anti-Semitism which is ‘regularly disseminated in speeches, books,
articles, sermons and everyday talk’ about this figure, Bishop Velimirović.

Learning from this example 

The article by Jovan Byford presents an example of the following:
• a discourse analytic study of prejudice and racism;
• a cross-disciplinary study (politics, history, social psychology);
• an analysis of the discourse of politicians and public figures;
• a case study centred on an event celebrating a particular figure;
• a study which combines analyses on different levels, investigating
the intersections of argument and rhetoric with political ideologies,
nation-making and the rewriting of history;
• an analysis of racist political rhetoric which links the celebration of a
single controversial politicized figure with larger political movements
and historical trends;
• a research project which combines collected and ‘found’ discourse data.

Gender in talk 
Example 3
Elizabeth H. Stokoe (2003), ‘Mothers, single women and sluts: Gender,
morality and membership categorization in neighbour disputes’,
Feminism & Psychology 13(3): 317–44.
 Four examples of discourse analysis 39

The third article presents analyses of talk about disputes between neigh-
bours. It is an example of discourse research which looks at a social problem
and also contributes to a major area of social science research, on gender.
It is a piece of feminist discursive research in which gender identities are
considered as ‘performances, constructions and enactments, rather than
rigid and unchanging essences’ (p. 318). This is consistent with the theories
of social practice and identity discussed in Chapter 2, including the work
of Judith Butler. In addition, Stokoe’s article is an example of research from
discursive psychology which presents fine-grained discourse analysis in the
conversation analytic tradition, examining the details of interaction and, in
particular, categorization.
The starting point for the project which Stokoe discusses is the relation-
ships between neighbours. This topic links the research to the particular
social and cultural context of the United Kingdom, since one element is the
conventions around how neighbours behave. However (as with Bell’s study),
similarities can be assumed in many other countries, giving the research
a wider relevance and potential application. In addition, as already noted,
the research concerns gender. In the opening sections of the article, Stokoe
makes connections between these various foci. She argues that ‘as neigh-
bours describe, report and account for their own and others’ activities, they
display the social order, which, in turn, regulates everyday neighbouring
practices’ (p. 319). In their talk, speakers refer to ‘people’s actions as either
appropriate to or as breaches of the moral or social order’ (p. 319), and in
doing this, Stokoe notes, they refer to values, to ‘‘good’ or ‘bad’ neighbours,
women and men’ (p. 320). This value-laden talk can partly be taken as evi-
dence of the existing social order, a snapshot of society and its beliefs.
These premises follow from the ethnomethodological tradition. As
discussed in Chapter 2, the major premises of ethnomethodology are
that society is never static but always in the process of being made and
re-made, and sometimes subtly altered, through the ongoing practices
of everyday life. The ‘macro-level’ of society is inseparable from these
‘micro-level’ practices. Stokoe is considering talk as one such micro-level
practice, looking at its effects and consequences. She explores connections
between ‘neighbour relationships, gender and morality’ (p. 317) through an
analysis of how women are categorized in the talk, the positive and nega-
tive values which are invoked in these categorizations (e.g., around being
a good or bad mother) and how the talk functions as part of the ongoing
‘constitutions of the gendered social and moral order’ (p. 340).
40 What is discourse analysis?

More specifically, the analysis draws on the work of Harvey Sacks, the
originator of the approach known as Conversation Analysis (CA). Stokoe
uses a variant of CA called Membership Categorization Analysis (MCA).
This investigates how talk functions partly through the kinds of grouped
or associated meanings which were discussed in Chapter 2. MCA does
not refer to discourses but to ‘categories [which] may be hearably linked
together by native speakers of a culture’ (Stokoe, 2003, p. 321). Stokoe
cites a famous example from Sacks’ work: when we hear ‘‘The baby cried.
The mommy picked it up’, we will of course assume that the ‘mommy’
is the ‘mommy’ of the ‘baby’. This is because our cultural knowledge of
the larger category of ‘family’ (a Membership Categorization Device) links
the categories ‘mommy’ and ‘baby’ with certain actions (‘category-bound
activities’) and characteristics (‘natural predicates’). MCA explores how a
brief reference to one part of the MCD may invoke other parts, ideas of
what is normal and so on.
Like CA, this can be seen as a ‘technical’ approach (Schegloff, 1997) in
that it begins with a close examination of what the participants say and
builds up an interpretation and argument from the details of their talk,
in this case, about good and bad neighbour behaviour by women and
men. However, it also draws on the analyst’s own insider knowledge as
a member of society since she explores the ‘inferences’ derived from the
social context in which the language is being used, that is, the wider social
meanings of categories, as well as the meanings which come out of the
immediate conversation.
It is important not to assume that Stokoe is interpreting her speakers as
planning their talk in advance, like master strategists. In the approach she
uses, talk is understood as part of living, that is, as a form of action which,
like body movements (walking, gesturing, changing facial expressions) car-
ries meanings for (most) people around us, is more or less ongoing, and is
done mostly without thinking or preparation. As someone interested in
the working of the physical body might slow a film to study the details
of the movements of limbs and muscles, so analysts like Stokoe use audio
recordings of talk, transcribed in great detail, in order to examine how con-
versation is ‘done’. This is therefore very different from an unravelling of the
intended implications of a pre-planned statement, like a political speech
or a public relations release. A study like Byford’s which uses the latter
forms of data might assume that speakers are presenting well-rehearsed
arguments which perhaps were planned and carefully worded in advance;
 Four examples of discourse analysis 41

however, as discourse analysis this will still assume that the talk is an action
which occurs in context and is shaped by that context. It will be a unique
situated version, even if to some extent it repeats arguments and wording
which have been prepared and used before.
Stokoe’s study uses two forms of data. One consists of televised record-
ings in which people talk to chat show hosts about problem neighbours.
The other is recordings from centres in which mediators attempt to resolve
disputes between neighbours by talking to the people involved. As in most
conversation analytic studies, Stokoe’s recordings are transcribed in close
detail to show the irregularities of ordinary talk: repetition and re-starts;
sounds which are not words (‘mmm’, ‘hhh’); pauses; overlaps between the
talk of different speakers; and some features of how the words are spoken
(e.g., more or less quickly and more or less loudly).
This is one of the data extracts presented in the article. It is part of the
talk from a mediation session in which people are talking about a neigh-
bour with whom they have a dispute:
1 E: [I mean I came home the other day and she was (.) a:rguing with
2 somebody at the top of her voice in the street (.) [and it was eff’ing
3 ? [(?)
4 E: this and [eff’ing that
5 L: [yeah (.) she’s eff’ing all the time=
6 E: and I can’t beli↑e::ve that somebody would have such a showdown
in the
7 street=
8 G: =have you heard her shouting at the kids why don’t you piss off? to
9 these tiny little kids (.) at the gate? (.) I’ve heard her saying it
10 (.) you know (.) this I don’t know [. . .]
11 C: is she has anybody actually approached her
12 L: she [( )
13 E: [how can she [( )
14 G: [how can you ↑talk to the woman?
15 L: you just get a mouthful
Following the principles of MCA, Stokoe’s analysis of this talk considers
the categories of people involved and the activities which are linked to
these categories. In this extract, she notes that ‘the category ‘woman’ is
linked to activities including ‘arguing’, ‘eff’ing’, ‘shouting at the kids’ and
saying ‘why don’t you piss off’ (p. 340). To understand the significance of
42 What is discourse analysis?

this link between category and activity, think about the different mean-
ings which would be invoked if the same activities were linked to another,
male category of person, such as ‘man’ or ‘youth’. As Stokoe notes, there
are ‘moral’ issues being invoked in this account: ‘Here, “bad” women swear
in the street and in front of children, they are foul-mouthed, they argue in
public places and are bullies.’ The account of their neighbour which these
speakers build up is not only ‘about’ a particular incident but also about
wider social values and expectations regarding women, their roles and
behaviours. Stokoe comments: ‘In constructing such categorizations, the
participants display their category knowledge (e.g., “women” should not
“swear”) and position themselves as powerful knowers of the “right” way
for women to act (Nilan, 1995).’ In short: ‘Neighbour disputes about noise,
vandalism and communal spaces are enmeshed with moral assessments
about appropriate behaviours for women’ (p. 340).
Stokoe is interested in the gendered nature of the neighbours’ com-
plaints for several reasons. First, as she comments, they probably reflect
the continuing cultural association of women with home, local neighbour-
hoods and ‘community and domestic space’. Second, they show how
‘women’ and their inappropriate activities become the focus of neighbour
disputes’ (p. 332). She suggests that this is a form of ‘cultural regulation’
because the disputes function to reinforce established meanings around
gender and gender identities: ‘in order to maintain “viable” femininity,
women must engage only in those activities conventionally associated with
their category.’ The disputes about the proper behaviour of neighbours
are therefore also about the proper behaviour of women more generally,
and the policing of one, in the broad sense of watching and attempting
to control, is part of the social policing of the other. In the terms used in
this book, gendered discourses are reinforced. A third claim which Stokoe
is making is that this analysis of how gender is invoked in one kind of dis-
pute, between neighbours, is likely to be relevant to other situations since
gender is almost always present as a potential point for comment: it is ‘a
pervasive resource’.
In her more recent work, Stokoe (2010) has extended the concept of
MCDs in an analysis of the talk of men who are being interviewed by the
police. There is a strong pattern of the suspects categorizing themselves
in a certain way, as the kind of men who do not hit women. This is not
just a matter of self-description; the categorization functions as a denial
that they have committed an offence. Stokoe is arguing against the kind of
 Four examples of discourse analysis 43

interview-based research which treats a discourse as a generalized resource

drawn on to talk about something. By showing a much more specifically
situated use of categorization to do something, and do something ‘conse-
quential’ for the speaker, Stokoe emphasizes the functional nature of talk
in interactional contexts.

Learning from this example 

This article by Elizabeth Stokoe presents an example of the following:
• a discourse analytic study of gender;
• a discourse analytic study of a social problem with findings which
have practical applications (Stokoe’s website provides further details
of her involvement in related work);
• Membership Categorization Analysis, a form of conversation analysis;
• an analysis which emphasizes the functional nature of talk;
• a project which uses naturally occurring data;
• a project which combines found data and newly collected data;
• an analysis with a focus which emerged from the data (an important
principle in conversation analytic research); the research was not set
up as a project about gender;
• an analysis of closely transcribed talk which avoids any interpretation
in terms of the speakers’ intentions, motives, emotions or other ‘inner
states’.

Leadership in an organization 
Example 4
Ruth Wodak, Winston Kwon and Ian Clarke (2011), ‘“Getting people on
board”: Discursive leadership for consensus building in team meetings’,
Discourse & Society 22(5): 592–644.
The fourth article discusses discourse analytic research in the area
of workplace and organization studies. In the words of the authors and
researchers, the research adopts ‘an interdisciplinary discourse-oriented
approach to leadership in meetings and teams, studying discourse in use’
(p. 593). It investigates how leadership is ‘accomplished’ or ‘performed’ in
the chairing of workplace meetings, particularly the skills and ‘strategies’
employed (p. 594). The article suggests that this approach contrasts with
44 What is discourse analysis?

previous writing and theorizing in which leadership is discussed either as

a style of setting goals and giving rewards (transactional leadership), or in
terms of the leader providing a role model (transformational leadership).
The research project is therefore presented as a novel investigation of an
established topic. The research is described as ‘critical’, first because of this
challenge to established understandings of leadership, and second because
of its concern with the links between leadership and power.
The article sets out its rationale for adopting a discursive approach to
the study of leadership. The premises are that the effectiveness of a team
depends on shared agreement, or consensus, among its members, and
the formation of such consensus depends on good leadership. Within an
organization, the people officially designated as leaders are potentially
powerful, but not all of them will fulfil this potential. The successful
performance or enactment of leadership requires ‘discursive skills’ and
strategies and these can be investigated through an analysis of the talk
which occurs in meetings, although the full context of this talk needs to
be understood.
The research was conducted in an Australian company, referred to pseu-
donymously as Defence Systems International (DSI). Two field researchers
observed meetings and other group interactions (workshops, an away
day) over a six-month period, making field notes. They also conducted
individual interviews with some of the staff who attended the meetings.
The data set consists of the transcripts from 150 hours of recordings of the
meetings and interviews, although at some stages of the analysis a further
selection is made. The interpretation and discussion of the transcripts is
informed by the background knowledge which the researchers obtained
from the fieldwork. This is referred to in their accounts of the participants,
the issues under discussion, the influence of people not present and
subsequent events. The ethnographic element of the research is therefore
used to locate the meeting transcripts within the larger context of the
company.
The project combines this ethnographic element with Critical Discourse
Analysis (CDA). CDA refers to a family of approaches rather than a single
method (Wodak, 2007). It is interdisciplinary, drawing on sociology and
sociolinguistics, and explores language as an element of social processes by
considering a range of features or aspects. Critical Discourse Analysts study
interactions and sequences of talk, like conversation analysis researchers
such as Elizabeth Stokoe, and they are also interested in the argumentative
 Four examples of discourse analysis 45

and rhetorical aspects of talk and texts, like those which are the focus of
Jovan Byford’s study. Critical Discourse Analysts also investigate the kinds
of patterns which Kirsten Bell and others call discourses (see Fairclough,
2001b, pp. 229–66). More distinctively, CDA involves a micro-analysis of
details of grammatical and communicative (‘pragmatic’) aspects of talk
including, in this study, pronouns and tenses, and speech acts, such as
questioning and telling.
The discourse analytic approach adopted in the study is described in
detail. The data set of the transcripts from the interviews and meetings
was analysed (in part or in full) in four stages, referring to different levels.
The first stage was a computer-assisted corpus linguistic analysis, using
software which detects recurrent uses of ‘keywords’ (see Yates, 2001,
pp. 93–146, for a detailed example). This is close to the ideal of an objective
or technical analysis, in that it is statistical; however, there will inevitably
also be judgement and interpretation involved, in this case, in the research-
ers’ initial selection of a subset of meetings which they thought were most
relevant to the company’s strategy. They then conducted the keyword
analysis on the transcripts from these selected meetings.
In contrast, in the second stage of the analysis, the researchers
conducted an interpretive analysis of agenda items and related topics
of discussion, looking for patterns of the discussion (including ‘topic
elaboration’ and ‘argumentation patterns’) in order to identify the
‘macro-topics’, ‘primary’ and ‘secondary’ topics. The third stage of
the analysis was to classify the various strategies of the leaders of the
meetings, and the fourth to consider how these strategies were used to
‘achieve consensus’. The multi-level analysis can therefore be described
as moving, broadly, from what was discussed (topics) to how it was
talked about, then to how the leaders intervened in the discussions and
produced agreement.
The researchers identified five ‘discursive leadership strategies’ which
were linked back to the two styles of leadership identified by previous
writers (transactional and transformational). Following the multiple foci of
CDA, these strategies are discussed in terms of various features, including
the grammar and also the communicative functions (linguistic-pragmatic)
of the talk. Although some of these features might seem sufficiently
specific to be identified unquestionably, the analysis is not as objective
or technical as, say, counting key words. The analysis is interpretive. The
researchers conduct it as informed observers, able to understand the
46 What is discourse analysis?

talk in context because of the knowledge they have built up through

the ethnographic study, including their reading of ‘confidential company
documents’.
The five strategies identified by the researchers are as follows:
(1)  Bonding. This is defined as ‘the discursive construction of group
identity that supports motivation to reach consensus and a
decision’ (p. 603). The authors suggest it is evidenced in a leader’s
use of the personal pronouns. Part of the analysis was to count the
relative uses of individual (‘I’) and group (‘we’) references by the
talk of someone chairing a meeting.
(2)  Encouraging. This is a strategy to open up general discussion and
encourage other speakers to participate in order to promote a
sense of the decisions being made by the group as a whole. The
features of the talk which are taken as evidence of this include
linguistic-pragmatic features such as questions to invite others to
express their opinions, and ‘the questioning/supporting of existing
propositions, via repetition, positive back-channeling and explicit
praise’ (p. 604).
(3)  Directing. In contrast to the previous strategy, this involves closing
down or resolving the discussion and is evidenced, unsurprisingly,
by opposite features: making firm statements rather than asking
open questions; summarising (‘So . . .’), and privileging of the chair’s
views over those of others, including by ‘blocking’ what they say.
Although it might seem possible to detect features like this by
looking for specified details (statements, not questions; particular
words), identifying the strategy also requires interpretation. For
instance, the researchers cite examples of someone pointing to
an architectural drawing and saying, ‘Look at this building’, which
closes down the previous discussion, or making a joke, which has
the same effect.
(4)  Modulating. This strategy involves contextualizing an issue in
terms of urgency by linking it to larger imperatives or threats, for
example, by suggesting (or denying) that there is only a limited time
in which to achieve something. Again, it is difficult to reduce this
strategy to features or categories which can be identified without
interpretation. The analyst will have to look at the functions of a
particular utterance in context.
 Four examples of discourse analysis 47

(5)  Re/Committing. This is the strategy of moving from a shared

understanding to an agreement to act. It often involves a reference
to the future, so grammatically, a change of tense. It may also
involve reference to people’s duties or responsibilities, or an appeal
to their values.
The researchers then discuss extracts from transcripts of meetings to
show how the chair of the meetings uses the different strategies and how
their use constitutes a leadership style. On the basis of the analysis, the
researchers argue that ‘an egalitarian leadership style positively influences
the formation of consensus within a team and, importantly, increases the
likelihood of a durable consensus being achieved’ (p. 612).
It is important to note that the same leader can adopt different styles
in different contexts, and the same strategies can to some extent be
employed as part of different styles. For example, directing can close down
discussion, as part of a more authoritarian style (corresponding to trans-
actional leadership) but it can also be part of summarizing and therefore a
step towards reaching consensus in a more egalitarian (transformational)
style. As with other forms of discourse analysis, a constant or predictable
connection is not assumed between a ‘type’ of person and a form of dis-
course or discursive practice.
More unusually, at some points in the analysis, the researchers speculate
about the mental states of the speakers, suggesting, for example, that a
chair has previously made a decision which he does not disclose during
the meeting. This contrasts with approaches, such as conversation analy-
sis, which avoid attributing intention to speakers (see the discussion on
discourse analysis and psychology in Chapter 2). In this study, there is some
implication that leadership strategies and styles are employed intention-
ally, at least to some degree. This would suggest that the research findings
have practical implications: people can be taught to be more effective
leaders. However, this is not stated directly.

Learning from this example 

The article by Ruth Wodak, Winston Kwon and Ian Clarke presents an
example of the following:
• a discourse analytic study of leadership in organizations;
• a project which combines Critical Discourse Analysis and ethnography;
48 What is discourse analysis?

• an analysis of meeting talk;

• a study which combines analyses on different levels, linking a macro
issue, the exercise of power through leadership, to language practices,
including the use of argument and rhetoric, and the details of the
language itself, including linguistic features;
• a relatively large-scale study involving more than one researcher;
• an analysis which is not confined to speakers’ talk but also speculates
about their intentions and other mental states.

Concluding comments 
The four articles which have been discussed in this chapter present some
of the possibilities and varieties of discourse analytic research. All four arti-
cles are interdisciplinary, investigating issues and situations which are of
interest to academics in several areas of the social sciences. However, the
analytic approaches employed by Jovan Byford and Elizabeth Stokoe are
more closely associated with discursive psychology (Stokoe’s article was
published in a psychology journal), including in the sources they cite.
As already noted, the researchers and authors describe their studies
in different ways and employ different concepts. Kirsten Bell uses the
term ‘discourse’ to refer to a family of words and images associated with
a particular topic. Her discussion of ‘lifestyle discourses’ and ‘health pro-
motion discourses’ includes reference to a ‘nutritional discourse’ which
distinguishes between ‘good’ and ‘bad’ food. The discourse therefore links
specific words (such as ‘diet’ or ‘healthy eating’) to values (here, ‘good’
and ‘bad’ categorizations for particular foods and eating habits) which in
turn are connected to larger ideas (what you eat affects your health; you
have a responsibility to monitor your eating; eating well can prevent the
onset or recurrence of cancer) and consequently to activities and aspects
of society which are not so easily contained in language (cancer support
groups; nutrition as a field of professional training and qualification; the
distinction between medical and other forms of treatment; the pressure
which cancer patients may feel to monitor their behaviour and even take
responsibility for having brought the disease on themselves).
Jovan Byford mostly uses the term ‘discourse’ to refer to the talk or
language of a particular speaker or source. He refers to ‘public discourse’,
for talk and publications which are directed to a wide public audience, and
 Four examples of discourse analysis 49

‘ecclesiastical discourse’, for talk which is by and about religious matters, in

contrast, say, to official political communications. He also refers to ‘the dis-
course of the Right’, meaning both the language of certain speakers, from
a particular political position, and the words, images and meanings which
make up or constitute their politics. Elizabeth Stokoe is interested in the
discourses that ‘shape our gendered world’ and in how ‘taken-for-granted
“facts” about women’s appropriate behaviour and characters are perpetu-
ated in discourse’, that is, in people’s talk. For Wodak et al., the analysis of
discourse involves multiple levels, from ‘strategy’ to the details of wording.
Published accounts of research findings from particular projects, like
these four articles, do not usually include detailed descriptions of how the
analyst worked with her or his data. Nonetheless, there are some points
which can be noted.
The first is that all four projects were conducted by the academics
who wrote up the research, sometimes with the involvement of others.
Discourse analysis is often the work of a single academic who works
through a data set, becoming familiar with it, exploring and making con-
nections and building an interpretation and argument which is relevant to
the starting concerns of the project. This may place a practical limit on the
scale of a discourse analytic project, especially the quantity of data which
can be worked with. However, joint analysis is possible, although usually
only in a small team. (An example of how this might be done is presented
in Chapter 4.) Bell’s project involved a second researcher, although there
is no indication that the analysis was a joint process. Kwon and Clarke
conducted the data collection for their relatively large project, and then
worked together with Wodak on the analysis.
A second point to note is that all the analyses were conducted across
the data sets. In other words, they were not confined to discrete items
like a single speech or document or even a single interviewee. Discourse
analysis is concerned with language use as a social phenomenon and
therefore necessarily goes beyond one speaker or one newspaper article to
find features which have a more generalized relevance. This is a potentially
confusing point because the publication of research findings is generally
presented through examples and the analyst may choose a single example
or case to exemplify the features to be discussed, but those features are
only of interest as a social, not individual, phenomenon. Occasionally
the particular case may be chosen precisely because it has a wider social
50 What is discourse analysis?

importance, as with Byford’s discussion of Bishop Velimirović as a focal

figure in contemporary Serbian politics.
In all four studies, the researchers also worked with far more data than
can ever be written up or published. Wodak et al. did include extended
transcripts in appendices to their article, but these were still only a small
part of their full data set. This links to a further point to note: as part of
the familiarization process, the researchers in all four studies had collected
background material. Such material is sometimes separate to the data
which are analysed and sometimes has a different function in the study.
For example, Bell’s attendance at the support group sessions, recorded in
her field notes, presumably provided some of details which assisted her
analysis (background on her participants and their experiences; practical
information about different treatments and their side effects, etc.). Her
description of her research as ‘ethnographic’ marks this backgrounding
as part of the research process. Some of her observations are included in
her field notes. Byford had clearly researched Serbian history and collected
other information as part of his extensive data set, all of which was neces-
sary background to his analysis (e.g., the biography of Bishop Velimirović
and the history of the treatment of Jewish people in Serbia and former
Yugoslavia). Stokoe’s analysis of Membership Categorization Devices draws
on her own cultural knowledge, for example, about the meanings and
associations attached to motherhood. She had also clearly learned about
neighbourhood mediation services and the kinds of problems they com-
monly deal with, and had probably watched far more broadcast material
than she eventually included in her data set. The analysis by Wodak et al.
refers to details of the company’s situation and the relationships between
staff which the researchers had learned during their fieldwork. As these
examples indicate, a discourse analytic project often involves the kind of
background or support familiarization which will enable the researcher to
conduct an analysis as an insider (or partial insider), and this may some-
times blur with the data collection.
The fourth and final point to note here is that the four analyses refer
back to, and build on, the findings of previous research, and the concepts
used. Like other research, discourse analysis is always located within a
tradition of ongoing research, substantively, that is, in terms of the topic or
problem being investigated, and methodologically, in terms of the particu-
lar approach and concepts which are employed. Two of the studies refer
to named approaches, Membership Categorization Analysis and Critical
 Four examples of discourse analysis 51

Discourse Analysis, but both developed these approaches to accommo-

date the requirements of the specific studies. The process of conducting a
discourse analytic study is discussed in general terms in the next chapter.

Summary 
Chapter 3 has discussed four published studies as examples of discourse
analytic research.
4  The usefulness of discourse
analysis for social science
researchers
The previous chapters have introduced theories and issues associated with
discourse analysis, and discussed examples of published discourse analytic
research. This chapter is the first of two which offer more practical guid-
ance for researchers planning their own projects. It begins with a discus-
sion of some of the reasons why researchers use discourse analysis. It then
describes some of the different kinds of discourse data. The final section
discusses briefly how to obtain data although, given the focus of the book
on analysis, this cannot be a full guide to conducting empirical research.
The section includes advice on collecting new data, selecting already exist-
ing or ‘found’ material as data and producing transcripts.

Reasons to use discourse analysis 

This section discusses four possible reasons to adopt a discourse analytic
approach in a research project, and then two more which have been
labelled ‘half-reasons’ because there is often confusion about their rel-
evance for discourse analysis.

Reason 1: To work with interesting data

Many aspects of research are enjoyable. Researchers become closely
involved in the problems which they investigate. They work with theories
which present them with new insights and questions. They have the satis-
faction of producing findings which contribute to the further development
of those theories, add to general understandings of the social world and
also have direct practical applications. Discourse analytic research is also
notable for the intrinsic interest of the material which researchers work
with. As previous chapters have indicated, this can include conversations
and interviews, observations of what people do and texts which have been
produced for a wide range of purposes.

53
54 What is discourse analysis?

Reason 2: To investigate the practices and interactions of

ordinary social life
The research projects outlined in Chapter 3 show how language data can be
analysed in investigations of ordinary talk: between disputing neighbours
or people attending a support group or colleagues in a work meeting, and
by politicians using language in ways which are familiar from newspapers,
media interviews and party political broadcasts. Discourse analysis enables
researchers to explore these everyday situations and practices as part of
larger processes and social phenomena, such as a contemporary political
project to construct a new nation; the racism which potentially excludes
certain people from the nation; the leadership practices which facilitate
the running of a large business organization; the social norms linked to
gender identities which function to exclude and stigmatize certain people
in more local contexts and interactions; and a society’s beliefs about health
and illness which may function to encourage lifestyles, for example, in rela-
tion to exercise and diet, but also to blame people for becoming ill. The
main argument for approaching talk and language data as evidence of the
practices and interactions of ordinary life comes from ethnomethodology,
discussed in Chapter 2.

Reason 3: To investigate a particular culture

Investigations of culture through texts of different kinds (images as well as
writing) often draw on discourse analytic approaches, although they may
not always be described as such. As Chapter 2 discussed, any study of culture
raises issues of definition. Because this book is primarily concerned with
social research, it is less focussed on art, music and literature, sometimes
described as ‘high culture’, than culture in the more anthropological sense
of lifestyle and shared customs and beliefs. The problem then becomes the
limits of the culture which is being referred to. Are the research findings
applicable to a national population as if that were bounded and uniform?
A further problem here can be that different languages can be assumed
to relate to distinct cultures, which logically is not the case (Bell’s study,
discussed in Chapter 3, collected data from two language groups within
the same nation-state, Canada). In short, the boundaries of any so-called
cultural group are inevitably porous. The researcher must therefore decide
whether any claims from the research have a broad reference, for example,
Western culture, or a narrower one such as one age group in one country,
or one kind of organization.
 Discourse analysis for social science researchers 55

Most discursive researchers would probably make claims about a more

local and specific cultural environment but they would also understand
culture as complex and divided. An example of this position comes from
Nigel Edley and Margaret Wetherell (1995) in their research on masculinity.
They said: ‘The cultural environment in which people live is not coherent
and integrated … but embodies a whole range of differences and contra-
dictions.’ Following Antonio Gramsci, the Italian Marxist, they suggest that
‘society’s common sense’ is ‘a fragmented body of ideas and practices,
littered with the “sediments” or “deposits” of age-old ideas and theories’
and ‘an established stock of taken-for-granted understandings (which) can
always provide the raw materials for the construction of markedly differ-
ent world views’ (p. 165). In the terms presented in this book, this common
sense can be analysed in terms of discourses or discursive resources. The
Membership Categorization Analysis which Elizabeth Stokoe employs (see
Chapter 3) similarly refers to categories and connections which are taken
for granted and can be understood to be part of common sense.
Common sense can therefore be understood as the shared possession
of a whole society but it does not provide a single coherent cultural view-
point for everyone in that society; its fragmented and mixed up character
(with sediments from the past) can provide a basis for different and even
conflicting arguments and perspectives. Discourse analysis informs the
investigation of that kind of complexity, including its contradictions and
conflicts, rather than any generalized claim about a society or cultural
group. Analysts may investigate the conflicts and associated contests
around power through an analysis in terms of ‘ideology’, as in the study by
Jovan Byford. A key source for this approach is the work of Michael Billig
(1996; 1999).

Reason 4: To research identity

Researchers may employ discourse analysis to investigate ‘big’ categories
of identity, such as national or gender identities, or they may begin with
a narrower focus, such as illness or neighbour disputes or creative work,
and then find in the course of the analytic process that identity becomes
relevant. Some theoretical issues around identity were discussed in
Chapter 2. Discourse analytic research on identities may focus on identities
in context, as they are variously claimed, conferred, rejected and modified
in ongoing talk, for example, between school pupils (Edley and Wetherell,
1997) or between people in a counselling situation (Edwards, 1998).
56 What is discourse analysis?

Another approach is to consider the identities associated with a particular

life situation or problem, as evidenced in language data from many dif-
ferent people in that situation, such as cancer patients and their partners
(Seymour-Smith et al., 2002; Bell, 2010), or people who live in one country
but claim the national identity associated with another (Scully, 2009; 2010),
or people at a similar point in their lives (Reynolds, 2006; 2008; McAvoy,
2009). The second approach may be linked to a theorization of identity
or subjectivity which would not be described wholly or exclusively as dis-
cursive. Psychosocial research is broad-ranging but in some versions brings
together an analysis of discourses and an interpretation using concepts
associated with psychoanalysis (Hollway and Jefferson, 2000). Narrative-
discursive research considers the resources and contingent constructions
of sequence and consequence in people’s talk, including their construc-
tions of biographical narrative (Taylor and Littleton, 2006; Taylor, 2010).

Half-reason 1: To obtain practical information about people’s

lives and experiences
This is probably the commonest purpose for which researchers collect lan-
guage data, including talk, but it is labelled here as a ‘half-reason’ because
it is not a priority for most discourse analysts.
Research for information tends to assume that language is referential or
transparent; it is a channel or medium for carrying information from one
person to another. Journalists make this assumption when they interview
eye witnesses or experts, asking them to describe or explain something
that has happened. However, most discourse researchers start from a dif-
ferent premise; they are interested in talk or language use as functional
and constitutive.
The distinction can be confusing for two reasons. The first is that the
same data can be employed for studies with both purposes; the differ-
ence is in the analysis, not the data collection. (The same language data
can of course be repeatedly analysed in different ways: see Van Den Berg
et al., 2003) The second is that although it is usual to make this distinction,
between language as transparent and language as functional or constitu-
tive, in practice the approaches are difficult to separate entirely. Using an
interpretive or constructionist approach like discourse analysis does not
exclude the use of talk as information. Moreover, even the writers who
argue that language is constitutive also use language referentially in their
 Discourse analysis for social science researchers 57

own work, for example, when they describe what they did and present
arguments and findings. An analysis will normally treat participants’
accounts at least partly as information and in most research it is possi-
ble to accept these accounts as broadly truthful and reliable. As Martyn
Hammersley has argued: ‘No knowledge is certain, but knowledge claims
can be judged in terms of their likely truth’ (1998, p. 66).

Half-reason 2: To explore people’s opinions

Social psychologists, and especially those who employ discourse analytic
approaches, have criticized the kinds of research which employ language
data in order to investigate attitudes and opinions. They reject research
which treats people’s talk (or writing) as evidence of what they will say or
do in future. They point out, for example, that an expression of prejudice
is not necessarily proof that in some future situation the speaker will
act in a prejudiced way (see Potter and Wetherell, 1987). In particular,
discourse analysts criticize any assumption that a stated opinion is the
straightforward expression of an attitude, as if this were some fixed men-
tal state. They therefore avoid research which employs questionnaires
or interviews with the purpose of collecting opinions or measuring or
evaluating attitudes.
This rejection of conventional opinion polls and attitude research fol-
lows from two arguments associated with discourse analysis. The first is
the argument that language is not transparent, as discussed above. What
someone says will be shaped by the immediate context, including by what
has come before. It is not direct evidence of the workings of the person’s
thoughts or mind. This links to the second argument, concerning the
nature of the person. As discussed in Chapter 2, discourse analysts and
discursive psychologists would reject the whole model of the person as a
kind of container for inner states or processes.
What are the practical consequences for discourse analytic research?
Discourse analytic research may investigate the ideas and ways of talking
which function to exclude certain groups of people in society (Wetherell
and Potter, 1992). This is likely to involve an exploration of discursive
resources. They may consider the functions of (apparently) prejudiced talk
in a specific context (Condor, 2011). Jovan Byford’s research, discussed in
Chapter 3, illustrates both these points. In summary, discourse analysis
explores ideas which are commonly held across society or particular social
58 What is discourse analysis?

groups or contexts, but they generally do not investigate or make claims

about the opinions of individuals.

Different forms of data 

Discourse analysts work with many different kinds of language material.
The research projects discussed in Chapter 3 illustrated the most common
types. Kirsten Bell interviewed participants and also compiled field notes
in her role as a participant-observer in cancer support groups. Jovan Byford
conducted interviews, which he recorded, and also collected a range of
texts which were publicly available, for example, in newspapers. Elizabeth
Stokoe arranged for mediation sessions to be recorded and also collected
recordings of television interviews. Wodak et al. recorded meetings and
other events taking place within a business organization. Discourse stud-
ies can also analyse data which are not linguistic. For example, a study of
nations and national identity might consider the ways in which flags are
displayed (Billig, 1995), research on gender might discuss still or filmed
images in advertisements (e.g., Gill, 2008).
Discourse analysts work with written language, for example, in the form
of official documents, news articles and online material from web pages
and forums. They analyse audio and video recordings. Although some
researchers do work directly with the recordings, listening and watching
repeatedly, the recordings are usually converted to written form through
selective copying of what is said. This process of transcription is discussed
below.
In summary, we can distinguish two categories of data. The first is new
material which is collected specifically for a research project, such as
interview and focus group recordings, and the second is already existing
material (TV and radio programmes, newspapers, etc.). A researcher who
collects new material may need to recruit potential participants, with
all the steps this involves, including appropriate ethical procedures. The
discourse analyst will not always be the interviewer or focus group leader.
For example, Bell’s data included recordings from a focus group conducted
by someone else and Taylor and Littleton’s research on creative work and
creative careers (Taylor and Littleton, 2008) used interviews conducted
by someone else, employed because she had particular interviewing skills.
A possible intermediate category of material is the recording of talk and
 Discourse analysis for social science researchers 59

interactions which would have taken place anyway, whether or not the
researcher was involved. Examples of this were Stokoe’s mediation ses-
sions and Wodak et al.’s meetings. This kind of data is often referred to as
‘naturally occurring’. It is favoured by researchers working in the tradition
of conversation analysis but is also used by other discourse researchers, as
in Wodak et al.’s CDA study.
The material which discourse researchers work with does not automati-
cally acquire the status of data. It needs to be collected and selected and
often transformed in other ways. The next section discusses some of the
processes involved. It provides practical guidance on obtaining data for
discourse analysis. However, note that this is not a general guide to setting
up and conducting an empirical project so it does not include advice on
project design, obtaining access and contacting participants, implement-
ing appropriate ethical procedures and other necessary aspects of any
research project.

Obtaining data 
This section will discuss the practical side of discourse research, specifically
issues of data collection and data processing, including transcription and
selection.
Planning any research project, whatever the data to be used, involves
establishing a theoretical and empirical starting point. This is given partly
by key theoretical sources and previous empirical research. Both of these
will shape the particular problem or research questions which a new
project addresses. The researcher will need to decide what a data analysis
might contribute, what specific analytic approach to use and what data
are appropriate to it. (As just one example, a conversation analyst would
probably not use interview data.) These decisions will help the researcher
either to devise an appropriate method of data collection, such as setting
up interviews or focus groups, arranging to record naturally occurring talk,
observing and recording in a situation in which the researcher is also a
participant, or to select the already existing data which are relevant to the
project.
The studies discussed in Chapter 3 referred to data which had been
obtained in several different ways: ‘found’ data (the different kinds of news
material analysed by Byford, and the television interviews analysed by
60 What is discourse analysis?

Stokoe), new data obtained through interviews (by Byford, Bell and Wodak
et al.), ‘naturally occurring’ data (the recorded mediation sessions analysed
by Stokoe and the meetings analysed by Wodak et al.), and observation
data obtained through participation (by Bell and her co-researcher, and
by Kwon and Clarke, recorded in field notes). The last mentioned form of
data is associated with ethnographic research. This is a research approach
in its own right so it will not be covered here. The remainder of this section
will address three questions: ‘How do I select “found” data?’ ‘How do I col-
lect data through interviewing?’ and ‘How do I obtain naturally occurring
data?’ The section leads into the discussion of the process of analysis in the
next chapter.

How do I select ‘found’ data?

A researcher who decides to analyse already existing material might
appear to have an easier task than one who collects new data. However,
there will still be a number of steps to follow before available material
can be approached as research data. A major problem with found data is
usually to narrow down the available material to a manageable quantity.
Newspapers (now mostly available online), web forums, advertisements
and official documents are all potentially excellent research material, as are
existing sound and visual recordings from the web (again), television and
radio but the researcher will almost certainly obtain more material than
can be analysed. The problem will be to make a manageable and appropri-
ate selection.
A further issue is to decide what material potentially constitutes data
and what provides background information: as the studies in Chapter  3
illustrated, the distinction may not be clear. In general, the data for a
discourse analytic study will be explored in terms of one of the analytic
concepts associated with the approach, such as strategies, discourses,
subject positions, dilemmas and Membership Categorizations.

How do I obtain naturally occurring data?

As already noted, ‘naturally occurring’ usually refers to talk which would
have happened anyway, whether or not the researcher had been there.
One way to obtain this is to analyse talk which has already been recorded
for another purpose, which is a form of found data. An example would be
when the researcher chooses to analyse television or radio programmes,
or online material. The practical advantages of using already recorded
 Discourse analysis for social science researchers 61

material are obvious; the disadvantages are that there may not be suf-
ficient information available about the context (e.g., when the recording
took place). The material may have been edited. There may also be some
ethical issues involved about who has the rights to the material and, in
some cases, whether it is appropriate to discuss speakers who cannot
remain anonymous, for instance, because they were recorded for television
or radio and named in the programme.
Alternatively, the researcher may arrange to obtain new data of this
kind; for example, Wiggins (2004) invited families to record their meal-time
conversation and many other researchers have placed recording equip-
ment in schools and universities (of course with permission). Some of the
most famous conversation analytic research has used recorded telephone
calls (e.g., in the work of Harvey Sacks; see Jefferson, 1992). A study which
seeks to obtain new data of this kind will of course involve the same steps
(planning, obtaining ethical approval, calling for volunteers, etc.) as other
research projects involving participants. These steps are not detailed in
this book since it focuses primarily on analysis.

How do I collect data through interviewing?

The most popular way to obtain data for discourse analysis is prob-
ably through interviewing. There are many published texts on research
interviewing and this section will not attempt to provide a general guide
but will focus on points relevant to interviewing for a discourse analytic
project.
The first point to note is that the data which will be used are the talk,
not the speakers. It is always tempting to attempt to ‘understand’ the per-
son behind the talk but this is not the aim of most discourse analyses. The
purpose of the interview is therefore an occasion for the collection of talk
data. Because an interview is a form of interaction, a further purpose will
be for the interviewer (who may not be the same person as the researcher)
to engage in an interaction with the interviewee. There are many differ-
ent types of interviews, from television interrogations of politicians to
job recruitment interviews to consultations between medical doctors or
therapists and their patients. A research interview is different from any
of these but can occasionally take on aspects of all of them, especially if
the researcher has not thought through the purpose and nature of the
interview as an occasion. Most discourse analysts will aim for an interview
which is something like a friendly conversation, that is, courteous, informal
62 What is discourse analysis?

and two-way, with the interviewer acknowledging what is said, contribut-

ing responses and perhaps some thoughts and experiences of her own.
The interaction between researcher and participant begins well before
the interview. Because discourse theories emphasize the situated nature
of talk, it will be relevant for the researcher to pay careful attention to the
way the project is presented to potential participants and the expectations
which may be established, for example, in how the interview is set up and
what they are invited to talk about. Most research participants volunteer
or agree to be interviewed because they have some interest in the general
topic of the research. However, they may also have misunderstood the
topic, however carefully they have been briefed, and even the purpose of
the interview, making assumptions, for example, about the kind of infor-
mation or details which are required.
The researcher has a responsibility towards the participants. Ideally
they will enjoy the interview: it is a situation in which they are receiving
attention and interest. Occasionally they may be involved because they
disagree with the researcher’s position, so that the interview is a form of
contest (this might have been the case with Byford’s interviews, referred to
in Chapter 3). In either case, the researcher is the beneficiary, receiving vital
contributions to her or his research project, and the interviewee deserves
to be treated with courtesy and, if at all possible, not caused distress or
offence.
Many discourse researchers prefer to conduct their own interviews but
others work from the recordings and transcripts of interviews conducted
by someone else. The research data will include both the answers provided
by the participant and the questions and other contributions from the
interviewer since these are all part of the interaction. The most common
type of interview is a semi-structured interview, that is, one in which the
interviewer has a prepared list of questions which is used as a general guide
but not a rigid template for the interview. For instance, the interviewer
may omit or rephrase questions or vary the order, depending on what has
already been said.
An important premise in discourse analytic research is that the talk
is prompted by the questions but not produced by them. The talk is
co-constructed in the interaction: this is part of its situated nature. It is
also shaped by the discursive resources which are available to speakers.
In addition, depending on the topic of the interview, much of the talk is
likely to be rehearsed, produced on this occasion as a situated version of
 Discourse analysis for social science researchers 63

what has been said before. These various assumptions suggest different
approaches to analysis (discussed in the next chapter) and also have impli-
cations for the researcher planning an interview. There is an argument that
an interview cannot ‘fail’ because any talk, or even an absence of response,
can constitute data and be analysed (see Jones, 2003, for an analysis of an
interview which went wrong). However, for novice researchers in particu-
lar, it will be preferable to have more talk rather than less, so ideally the
interview questions will be easy for the participant to understand; they will
be open-ended and interesting, inviting extended answers, and they will
encourage a flow from one topic or area to the next without any abrupt
change of focus to interrupt the interview as a conversation.
The researcher will of course need to decide how many interviews to
conduct. Since the aim is to analyse talk data, not participants, a discourse
analytic study seldom involves just one or two interviews. A fairly substan-
tial data set or body of interview material is appropriate. For a researcher
working alone, 15 to 20 interviews are probably the minimum. These might
be obtained from 20 participants, or 10 who are interviewed twice. Repeat
interviews, associated with a longitudinal study, can enable interesting
comparisons across different occasions of talk (see Taylor, 2012), as can
data from different sources, such as group discussions and one-to-one
interviews.
Regardless of the number of interviews conducted, they will almost
certainly be recorded for later transcription and analysis. Most researchers
in this field use audio recordings but video can provide useful additional
information, for example, about exactly who is speaking in a group inter-
view or focus group.

Producing transcripts
In its simplest form, transcription is the process of converting talk to written
language by writing down what was said. It is part of the analytic process
(Ochs 1979; Hammersley, 2010) but it is only the beginning. As the next
chapter discusses, the process of analysing interview transcripts requires
protracted and iterative re-reading and comparison, across different
interviews and different parts of the same interview. The initial purpose
is to find patterns in the data and this provides a useful, general guide for
transcription. The details to be transcribed will be those in which the analyst
seeks to find patterns. For example, if the interest is in discursive resources,
such as discourses, interpretative repertoires or narratives, then the most
64 What is discourse analysis?

important content of the transcript will be the words. If the analyst is

studying interaction, for example, in a conversation analytic study, then
the transcript will need to record details of the interaction, such as the
pauses or overlaps between successive turns of talk from different speakers.
Researchers have devised systems of ‘notation’ for recording various details
(hesitation, intonation, emphasis) and particular studies may devise new
symbols to indicate particular points of interest. Chapter 3 showed an exam-
ple of a detailed transcript from the article by Stokoe.
Transcription is not a purely technical process because it is always selec-
tive: it involves interpretation. There is almost no limit to the details which
can potentially be included, depending on the researcher’s interest, so it is
never possible to transcribe ‘everything’. Some analysts do all of their own
transcription, regarding this as a necessary part of the process of familiar-
izing themselves with their data. Others employ a professional transcriber
to produce a first version and in the near future, this may be done using
voice recognition software. However, the analyst generally needs to check
such transcripts against the original recordings and correct and add details.
Some researchers begin by working with recordings and part-transcripts,
then produce final transcripts of selected sections of the data rather than
attempting to transcribe the entire data set. Whatever choice is made,
transcription is a laborious and time-consuming task.
There are three further points to be made about transcripts. The first
is that a transcript is primarily a working document for the researcher.
Research publications such as reports, articles and monographs will
include transcript extracts but, as with other data, the discourse researcher
will always have analysed more material than can be presented to read-
ers. A transcript can seldom be published in full, partly for this reason of
length, partly because there will too much detail which is irrelevant to
the arguments of the researcher and partly because there will be details
which breach anonymity. (For example, a single reference to a detail like
a participant’s occupation or age or place of residence will probably not
enable others to identify her, but if at various points in her talk she dis-
closes several of those details, then taken together they may enable other
people to recognize her or find out who she is.) This therefore becomes an
ethical problem, since participants are usually promised anonymity.
The second point is that even the simplest transcription of talk will be
different from written language because the grammar of talk is different
 Discourse analysis for social science researchers 65

from the grammar of writing. This is a linguistic point which does not need
to be explored here, except to point out that people do not speak in ‘cor-
rect’ sentences. Newspapers sometimes present interviews as if these have
been directly transcribed, but the talk has generally been quite extensively
edited into the form of standard written language, in full sentences (indi-
cated with capital letters, full stops and commas). Unedited talk always
contains repetitions, incomplete utterances, false starts, hesitations and
self-corrections, so transcribed talk generally includes at least some of
these regularities. A ‘found’ transcript may provide suitable material for
analysis, but the analyst will need to be aware of how it has been edited
and what may have been omitted, or added.
A third point which sometimes concerns researchers is whether or
not they should ask participants to check the transcripts of interviews.
It is usual to offer people a transcript of their own interview but it is
important to be clear why this is done. First, it is a courtesy. Second, it
may also be done for ethical reasons, for example, to offer participants an
opportunity to ask for certain details to be removed, although that is not
usually necessary if participants have been promised anonymity. A third
reason, in a longitudinal study, is that a participant may be given a tran-
script as preparation for a subsequent interview, to remind them of what
was said. An argument against this is that remembering is selective (i.e., it
always involves forgetting) so that participants may be surprised and even
distressed to read in a transcript what they said on a previous occasion,
especially some time ago. A different reason for not returning a transcript
is that people are often disconcerted by the irregularity of transcribed talk
because they are not aware that this is different from written language, as
noted above. More prosaically, transcripts are very long so the participant
is being given quite a substantial reading task. If a reminder is really needed,
then it is probably more appropriate for the researcher to prepare a short
summary of key points covered in the previous interview.
Some researchers, such as oral historians, do give back transcripts as
part of a process of negotiating a factual record. The participant is encour-
aged to revise what was said. However, a discourse analyst is interested in
the situated nature of talk and would not expect people to give identical
accounts in different interviews. If a participant amends a transcript (e.g.,
by adding details) what results will be a composite of several versions
from different occasions. Such a composite is not a transcript in discourse
66 What is discourse analysis?

a­ nalytic terms so is not appropriate as data for the kind of analyses dis-
cussed in this book.

Summary 
Chapter 4 has discussed some reasons to use discourse analysis for a
research project, the kinds of data which can be analysed and the com-
monest ways of obtaining data.
5  The challenges of discourse
analysis

This chapter reviews the problems or challenges faced by the discourse

analyst. The content of this chapter also complements the section
‘Obtaining data’, in Chapter 4, as a practical guide, moving on from the
task of obtaining data to the process of analysis and interpretation. The
chapter is organized in a question and answer format.

How do I decide the right approach to

discourse analysis for my project?
As the previous chapters of this book have indicated, there is no one
‘right way’ to analyse discourse. Any discourse analytic project estab-
lishes its own connections between the researcher’s assumptions about
the world (theory), the topic being investigated, the interpretation of
‘discourse’ which is adopted, the materials to be analysed (data) and the
relevance of those materials to the topic (such as when talk is analysed
as ‘identity work’). A new project may build on previous research or it
may develop its own unique variant; the resulting project may or may not
be labelled discourse analysis. As a consequence, the researcher needs to
be aware of the assumptions underlying the methods she or he employs.
Most researchers will begin by reading published studies (like the ones in
Chapter 3) alongside the work of key theorists (such as those named in
Chapter 2). The reference lists for publications can indicate other relevant
sources. The broad approach given by previous work can then be revised
during the planning of the new project and new analytic concepts may
be incorporated when the researcher begins working with the data, but
at every stage it will be important to check decisions back against the
underlying theory.

67
68 What is discourse analysis?

How do I decide how much material to analyse? 

This is a difficult point. It is certainly possible to have too small a quantity
of data. The analytic process outlined in Chapter 5 involves searching for
patterns across a data set, and this would not be possible with, say, just one
interview or one document. Alternatively, with that amount, it would still
be possible to build on the findings of previous research, to note that the
new data illustrate patterns and features noted by other researchers. For
example, a conversation analyst might label details of a new extract of talk
with reference to the canon of conversation analytic studies. There is an
assumption operating in this approach that interactions can be described
in near-universal terms, but most other discourse researchers would not
generalize so widely and so they work with larger data sets.
On the other hand, there are practical limits to the quantities of data
which are manageable. For instance, recorded material probably needs
to be transcribed, which is a laborious process, as noted in the previous
chapter. More importantly, the researcher needs to familiarize herself with
all or most of the data, as discussed in the following section.
It might seem obvious that collecting more data will raise the standard
of the research and the reliability of the findings, but this is a misconcep-
tion. First, qualitative research does not and cannot use statistically
representative samples; the basis for generalization is different (see later).
Second, the research process is not equivalent to drawing a very detailed
picture, adding detail to achieve a closer representation of reality. Rather,
the researcher is building a chain of argument, linking (i) a theory of how
the social world works, and (ii) a second theory (usually called method-
ology) of how material collected in an empirical study can be treated as
evidence of the workings of that world, to (iii) the selection and presenta-
tion of particular material (data) as the evidence to support a new claim.
Adding more and more data does not inevitably strengthen the links; the
more important point is how they are theorized.

How do I manage and ‘process’ my data? 

Most researchers work with a combination of electronic files, for easy
storage, and hard copy printouts, for easy reading, re-reading and mark-
ing up. The software packages which are rather misleadingly referred to
as CAQDAS (computer assisted qualitative data analysis software) cannot
 Challenges of discourse analysis 69

analyse discourse data, or any other qualitative data, because the researcher
will always need to familiarize her or himself with all of the material and
interpret it, a process which inevitably involves decision-making. However,
software is extremely useful as another medium for marking up data like
transcripts, and for appending comments and sorting material into new
files, especially when there is more than one researcher working on the
project.

How do I begin the analysis? 

The process of analysing discourse data is not linear but exploratory and
iterative. In other words, the analyst’s task is not one of straightforwardly
‘translating’ or ‘decoding’ the data, one item at a time; rather, analysis
involves reading and re-reading an entire data set, comparing, notic-
ing and marking points of possible interest and returning to them later.
(The analysts whose work is discussed in Chapter 3 all worked ‘across’
their data sets, as noted in the section ‘Concluding comments’.) This is a
time-consuming process and most novices do not allow enough time for
it. The analysis phase of a discourse analysis project may overlap with both
the data collection and the writing up and will take longer than either. It
cannot be accomplished in one or two sittings.
Without computers (and many analysts do prefer to work with hard
copies of transcripts and other data), starting the analysis may involve
marking selected sections of the data, perhaps with different coloured
pens; assembling piles or files of ‘like’ material; scribbling notes and mak-
ing lists of possible points for follow up, often working from features and
concepts noted by previous researchers. Many analysts now use software
to facilitate the same operations. A standard word-processing program
can be used to copy and paste sections of data into new files and mark up
points of interest. So-called qualitative analysis software facilitates all of
these operations and is particularly useful for coordinating the work of a
team of researchers. Some software, such as the corpus analysis programme
used by Wodak et al., searches out particular word sets. However, as noted
above, most software is designed as a tool for data management rather
than analysis. The hard work of exploring, thinking and decision-making
still has to be done by the analyst.
As the analyst becomes familiar with all of the data, one activity will be
to mark connections between different pieces of material, even though in
70 What is discourse analysis?

the first instance the significance of the connection may not be apparent,
and then to assemble subsets or collections of material. This operation is
often referred to as ‘coding’. Initially, the codes may be very broad (e.g.,
‘work’, ‘home’, ‘finance’) and they will not be exclusive: the same piece of
material may be coded several ways. Coding cannot be done mechani-
cally, without active thinking, although it is often possible use a computer
search function in order to find potentially relevant material, for example,
by looking up key words relevant to a code. Coding can be a useful ‘way in’
to a large data set but it is not the inevitable starting point.
For interview data, another starting point can be to compare differ-
ent participants’ answers to the same or similar questions. When the
same participant has been interviewed more than once, there can be
comparisons across the interviews. Since the talk is situated, it is not
likely that someone will say exactly the same thing on different occa-
sions, but both the difference and the similarities can be surprising. In a
narrative-discursive analysis, one point of interest is the similarities which
follow from the rehearsed nature of many accounts, so that previous
versions, for example, of a speaker’s life history, become resources for
her own subsequent talk: Taylor and Littleton call these ‘local resources’
(2006, p. 33; 2012, p. 42).
More generally, similarities across interviews (whether in the actual
wording, the images or broad ideas, or the style or form of talking) can
usefully be grouped together for a later decision about how to relate the
material to analytic concepts such as discursive resources. A different
starting point will be to consider the talk as action: What is it doing in
this specific interaction and more generally? It may be useful here to think
about rhetorical work, of the kind discussed by Byford. This is talk or writ-
ing which makes an argument, sometimes by anticipating a challenge or
criticism (‘I’m not prejudiced but …’), indicating the argumentative and
dialogic nature of talk (Billig, 1987; 1999). The relevant analytic concepts
will be given, at least initially, by the approach which the researcher has
chosen (see question 1, ‘How do I decide the right approach to discourse
analysis for my project?’). As the analysis proceeds, new concepts may be
taken up. For example, an investigation of how participants resist dominant
discourses (as in Bell’s study, discussed in Chapter 3) might focus attention
on the uses of humour. The researcher would then return to the data look-
ing for jokes, laughter and so on. New questions arise: Are there different
kinds of laughter? What is the significance of speakers laughing together or
 Challenges of discourse analysis 71

one person laughing alone? This point would suggest new features to be
included in the transcription, and so on.
The researcher’s interpretation is based on evidence obtained through
a rigorous data analysis, but this analysis remains hidden to a great extent.
One reason is that the quantities of data involved cannot be presented, for
practical reasons of length and also, of course, to protect the anonymity
of participants. Discourse analysis is therefore an interpretative analytic
approach grounded in the evidence provided, for example, from tran-
scribed talk. The previous painstaking exploration of multiple interviews
has to be taken on trust. (This is not very different to the ‘off-stage’ work
involved in most kinds of qualitative research.)
A discourse analyst detects a pattern in the data and eventually labels
it for the purposes of the particular discussion. Clearly, this is a loose
process of labelling rather than a right/wrong identification. For example,
what one researcher calls a narrative, that is, a construction of sequence
and consequence, often with a link to temporality, could be a part of what
another analyst would call an interpretative repertoire.
The remainder of this section presents two examples of the beginning of
the analytic process, one of a researcher working alone and one of collabo-
rative analysis between two researchers. Although many discourse analytic
projects are conducted by a single researcher, there is a sense in which no
academic researcher is working entirely alone because she or he is neces-
sarily engaging with others by reading their work, listening to conference
presentations and presenting papers to journals and conference audi-
ences for feedback. Discourse analytic researchers are also likely to seek
opportunities to present data or work in progress in seminars and research
groups. Other people’s suggestions can provide an invaluable stimulus for
an analyst who is struggling with a large data set.

Example 1
A researcher is analysing interview and observation data which she has col-
lected in a project working with young people. Her first task is to re-read
her field notes and interview transcripts and notes, at least twice over for
each. She does this over several days in a number of sessions, some of them
quite short (even as little as half an hour). As she reads, she tries to keep an
open mind, putting aside, temporarily, her own views of her participants
and their lives and her knowledge of other research and of theories. This
is sometimes referred to as ‘bracketing’: the purpose is to approach the
72 What is discourse analysis?

data as data without thinking too much about the speakers themselves
or the situations or events which are referred to. A group discussion and
brainstorming of a sample of data can help the researcher to do this; the
views of other people draw her attention to her own assumptions.
She marks sections of the talk which seem interesting for any reason,
such as wording which is striking, or a memorable anecdote or interaction.
She also notes features which recurred within one speaker’s talk (wording,
references, accounts), possibly on different occasions, or across the talk
of different speakers, or across several kinds of data (e.g., links between
what one participant said and what the researcher had observed another
participant do).
The researcher then goes back to the marked sections of the data and
re-reads those. She re-examines sections of interview data on several levels,
asking herself questions.
• The first is the level of the interview interaction between the inter-
viewer and participant. She asks: What function does the talk have
as a response to questions and follow-on to previous discussion?
Her purpose here is to examine the talk as interaction between the
people who were present. A further step is to consider whether there
is interaction with other potential or imagined audiences. Are certain
features of the talk addressed to people who are not present, such
as potential critics? Her purpose here is to consider the rhetorical
aspects of talk. For example, negative prefacing comments like ‘I’m
not saying that …’ are generally seen to be directed, or oriented, in
this way (cf. Byford’s study in Chapter 3).
• The second level relates to the talk more generally. Possible ques-
tions to ask are: What features of the talk itself are noticeable? Are
there stories or anecdotes which have a rehearsed feel, or are even
repeated? What positive and negative valuing appears? What’s men-
tioned as ‘good’ or ‘bad’ for example? Are there instances of implied
cause and effect, that is, what’s spoken of as leading to or following
from something else? What categorization appears in the talk? The
purpose here is to examine the talk as a complex changing produc-
tion which is shaped both by previous talk and ideas (the speaker’s
own, and other people’s) and by its multiple, overlapping functions as
communication, with the interviewer and with other audiences asso-
ciated with particular subject positions (imagined listeners, society in
 Challenges of discourse analysis 73

general, etc.). This contributes to an analysis of the talk in terms of

discursive resources.
• The third level is the most straightforward and the one common
to other qualitative approaches, like ethnography, which treat the
speaker as an informant. She asks: What’s happening in the events
described in the talk?
She compiles files of some of these features and labels the patterns for
reference, but at this point she does not try to explain their importance.

Example 2
Two researchers are analysing 50 interviews which have been conducted
for a large project by a research fellow, then transcribed. The researchers
have listened to each recording and read each transcript as they received
them. They then begin the process of analysis.
• They choose an interview to study closely. Each listens again to the
recording and re-reads the transcript, making their own notes.
• They meet. Each reports what she’s noticed about the interview and
one makes expanded notes which she writes up as a record.
• They repeat this process over different meetings, discussing two or
three interviews at each meeting. They start to see similar points
recurring in different interviews, or at different points in the same
interview. The notes on each interview begin to reflect these patterns,
which they label (e.g., ‘big money talk’, ‘a story about childhood’,
‘competition’).
• The researchers continue the process for all the interviews, which
takes hours! By the time they have discussed about half their data,
they have labelled a number of patterns and are talking about how
these fit together, why they might be there (e.g., what functions
they might fulfil) and what implications they might have. A useful
point can be the examples which depart from a pattern (Taylor and
Littleton, 2006, discuss an example of an analysis which builds on
a pattern and exceptions). The researchers also consider how these
patterns fit with points which other analysts have noted, and with
theories. Some ideas and possible arguments begin to emerge in
their discussion. The researchers note these but are careful not to
close down the process of exploration. They also consider their own
preconceptions, discussing what they might be wanting to find or,
74 What is discourse analysis?

alternatively, avoiding seeing. What personal interests are they bring-

ing to the data? Which participants do they feel more or less interest
in, and why? Are there any interviews they should read again, putting
aside certain assumptions?
• They choose particular patterns which they have noted and return
to individual interviews where the pattern appeared, looking again at
the pattern within the context of the particular interview.

How do I develop an interpretation and argument? 

At this point, the researcher, whether working alone or with someone else,
will be very familiar with a large proportion of the data, not just one or
two interviews. She has picked out recurring features and compiled col-
lections of examples of these features. The next step is less systematic.
The researcher steps back from the data to the larger foci, which are the
following:
• the research questions which were the starting point for the whole
project,
• the issues given by the theoretical background to the project and by
previous related research.
By relating the patterns to analytic concepts (discourses, categories,
strategies), the researcher develops an account of what is happening in the
talk. This leads on to the questions: What could this be evidence of? What
do I want to argue on the basis of this evidence?
The point of a discourse analytic project using talk data is that the theory
underpinning the project indicates a particular connection between the
talk and the larger question or problem being addressed. The aim is not
to analyse the speaker as a person. For researchers working with talk data,
it will therefore be useful at this point to try to put aside what is known
of the person and focus only on the talk. This is more easily done some
time after the occasion of the interview, or with data from an interview
conducted by someone else.
The aim now is to build up arguments, from evidence to claims. It will
probably be useful to build on similarities and differences compared to
the evidence and claims of previous researchers in this area. This think-
ing process is the most difficult part of analysis and interpretation. It will
 Challenges of discourse analysis 75

almost inevitably involve many drafts of notes and trial arguments. On the
way, some arguments, even with evidence, will be put aside, while others
will be rehearsed and refined.
The eventual core argument which will be presented in a research text
(a paper or a chapter of a dissertation or thesis) will be the outcome of
this extended process of development; the process itself and most of the
data cannot be presented in full. Discourse researchers are always looking
for succinct data extracts which exemplify patterns that recurred across
an entire data set. Sometimes it is practical to discuss a single case in this
way (e.g., Taylor, 2011) but the discussion will make clear how far this case
is representative of the larger data set, including the ways in which it is
distinctive. Talk or field notes can provide illustrative examples of general
features but any example will inevitably be rich in specific details, some
of which are particular to one participant or speaker. Part of the task of
writing up the analysis will be to draw attention to the general argument
and claims being made on the basis of the entire data analysis.

What is the status of the findings? 

The findings of a discourse analytic study are an interpretation grounded
in a rigorous process of data analysis. Generalization is on the basis of the
patterns found in a large body of talk or other language data through
the analytic process described above. The broad assumption is that the
analysis has discovered social meanings and language-related practices
which are a feature of a wider context and population. However, there is
no assumption that either the meanings or practices are wholly determin-
ing or predictive of future practices, interactions or situations. Social life
is assumed to be emergent and highly complex. Consequently, although
any example of talk is from an individual speaker, potentially describeable
in general category terms (a woman or a man, of a certain age, occupation
and so on), the speakers are not discussed or analysed as ‘types’.

Summary 
Chapter 5 answered some common questions about the challenges of
discourse analysis, including the process of analysis.
6  Criticisms of discourse analysis

This chapter will be presented in the form of ‘critical statements’ followed

by discussion.

‘Discourse analysis is just about words’ 

This criticism encompasses several different issues. A statement like this
is usually part of a larger complaint that discourse analysis does not take
account of something else, such as non-verbal communication, body
language and emotions, visual representations, or objects, space and the
material world, including people’s bodies. A number of points can be made
in response. In addition, the criticism can refer to how discourse analysts
discuss their participants: this is explored as a separate statement in the
section ‘Discourse analysis doesn’t take enough account of people’.
The first point to make here is to agree that, yes, for some discourse
analysts the approach is about words and language. They investigate how
texts and talk in certain contexts contribute to the exercise of power.
Parliamentary speeches, news reports, official statements and other
forms of public communication, such as advertisements, can all be ana-
lysed with this focus. The interest is in how language is used to reinforce
norms, legitimate existing social structures, subtly obscure or rationalize
inequalities, play down problems and perpetuate an interpretation of
society which supports some interests and obscures others. Analysts aim
to expose these effects and in doing so, to challenge established orders
and elites. The analysis of the texts may extend beyond the words them-
selves to the visual elements of a text, such as the organization of a news
page with headlines and pictures. However, the main focus will almost
certainly be on the language. The approach known as Critical Discourse
Analysis, associated with Norman Fairclough, Ruth Wodak and Teun
A. van Dijk among many others, is particularly known for this kind of
research.

77
78 What is discourse analysis?

A second reason why some (but not all) analysts would agree with the
above statement is that words and language have a special status in society.
When people talk to each other their language use may incorporate ges-
tures (such as a nod of the head) and sounds which are not words (‘Mm’
or similar as an acknowledgement or agreement) but these are meaningful
within a communication system which is primarily verbal. There are special
cases in which an entire ‘conversation’ can be conducted without words,
but they are rare exceptions. Communication through language use is a
vital, rule-governed social practice in which people necessarily become
skilled in order to function as members of society. Talk and language are
therefore obvious forms of data for a social scientist who wants to study
society. In an analysis of such data, a researcher can draw on her or his
own expertise as a language user, as well as on previous findings about
the conventions of conversation. The importance of this argument can
be underlined by comparison with a research project in which the par-
ticipants are asked to draw a picture or take a photograph. They are being
asked to engage in an unfamiliar practice with few or no conventions. (Of
course this may not be quite the case if they are skilled artists or photogra-
phers, and the analyst shares those skills, but that would be an exceptional
situation and highly unusual in social research.) The consequence will be
that neither the participants nor the analyst will be able to ‘do’ things
with the media or interpret what is produced as either a communication
or a socially meaningful activity. Alternatively, the researcher may invite
them to talk about what they have produced, so that the analytic focus
becomes, again, language.
However, some discourse analysts would disagree strongly with the
statement that ‘Discourse is just about words’. Their argument, broadly, is
that there is no neat separation between the meanings in language and in
the social world more generally. To talk about people, things and the mate-
rial world is to bring them into language (and therefore ‘into culture’ in the
words of Edwards et al., 1995). Even if people and things are not specifically
named and described by a research participant, they inevitably function
within larger systems of meaning. Previous chapters have discussed exam-
ples of the connections between certain named or nameable identities
(a good learner, a healthy eater), the values and knowledge which makes
the identities available, the institutions through which the knowledge is
established and disseminated and so on. Some of this complex formation
may operate without being explicitly brought into language: indeed, one
 Criticisms of discourse analysis 79

purpose of the research may be to identify and name what has previously
been invisible and unacknowledged. Words and language are inevitably
linked with the social world and its ongoing activities. In this view, to say
‘discourse analysis is just about words’ is to ignore the social meanings
which are attached to language and words yet not confined to them.
These issues arise partly because of the range of data and analytic
approaches encompassed within the category ‘discourse analysis’.
Researchers who are interested in the difficult experiences of people
who have been socially marginalized may feel that a focus on words and
language amounts to a trivialization of their participants’ difficulties. To
study details of wording, grammar and punctuation may seem inappropri-
ate, even offensive, and a turning away from vital, apparently non-verbal
aspects of their participants’ lives such as feelings, their poor health or
physical injuries, bad living conditions or incarceration. A further motiva-
tion here may be a wish to look at the ‘whole’. It may seem disrespectful to
investigate only small details of an account.
One counter-argument is that the apparently non-verbal aspects of
communication are generally connected to language, so language data
can provide appropriate evidence. Another counter-argument is that
all research involves the selective reduction of a complex social world to
data. For example, a census or survey reduces people and their situations
to categories and statistics, yet it can prompt help and improvement of
their welfare. Discourse analytic research has always had a strong connec-
tion with social issues. As the previous chapters have indicated, discourse
researchers have challenged prejudice, racism and other injustices. The
analysis of words and language, often at the level of fine detail, is their
starting point for doing this.
A final point to note here is the issue of emotions, more recently linked
to the study of affect. Some psychosocial researchers have suggested
that discourse analysis ‘leaves out’ emotion. The assumption here is that
the taking up or contesting of a subject position is a verbal or intellectual
action which is separate from feelings. This argument is possibly a conse-
quence of an artificial separation of language data from the continuum
of meanings, lived experience and so on with which it is implicated, as
outlined earlier. A similar separation appears in the common contrasting
of the rational with the emotional. A counter-argument would be that
the types of discursive resources discussed by many analysts are laden
with values and other affective associations so that the kind of identity
80 What is discourse analysis?

work which may be investigated and discussed in terms of transcripts is,

again, not just about words (see Taylor and Littleton, 2012).

‘Discourse analysis doesn’t take enough account

of people’
Much of the discussion in this book has referred to the analysis of talk
data from interviews and other interactions. For novice researchers in
particular, there can be a tension between forms of data which appear
to push attention towards the individual speaker, understood in com-
monsense terms as agentic and expressive, and an analytic approach
which challenges this conception of the person (see Chapter 2) and
makes claims about social phenomena rather than particular people.
Researchers who collect their own data are likely to have a strong
impression of speakers as driven by emotions and intentions, yet dis-
course analysis is concerned with meanings and practices which are
primarily social, albeit impacting upon and constitutive of individuals
and often investigated through their particular practices, including talk
practices. (This is a reason to avoid giving participants pseudonyms or
prefacing their discussions of data extracts with mini biographies of the
speaker or writer: see Taylor, 2012.)
This tension is seldom directly addressed in published accounts of analy-
ses or in research methods texts. Harvey Sacks, the originator of conversa-
tion analysis, famously said about people and their talk: ‘Don’t worry about
how fast they’re thinking.… don’t worry about whether they’re “thinking”.
Just try to come to terms with how it is that the thing comes off. Because
you’ll find that they can do these things’ (Sacks in Jefferson, 1992, p. 11). This
is another way of saying that the focus is on talk as a practice, a form of
doing which (like body movements) is seldom pre-planned or consciously
performed. As a consequence, analysts tend to avoid using language which
implies a two-stage process of thinking and speaking. This is a subtle point,
not least because, first, speaking can shift from a practice which is per-
formed without intention to one which the speaker is very conscious of
planning (e.g., in unfamiliar interactions), and second, because of course
some analysts do investigate material which has been planned with great
care and attention, such as political speeches on sensitive topics (as in
Byford’s research, discussed in Chapter 3).
 Criticisms of discourse analysis 81

Nonetheless, discourse analysis remains an approach primarily used for

investigating collectively held meanings and collective practices. The addi-
tional issues of how these may be constitutive of the talk of an individual
and, further, constitutive of the person as an individual are ones addressed
by some social psychologists (e.g., Hollway and Jefferson, 2000; Wetherell,
2003; Taylor, 2011; 2010) but are not necessarily part of discourse analysis.
This territory is one which is increasingly discussed as the concern of psy-
chosocial research. For novice researchers, the important point to keep in
mind is that discourse analysis is an approach for social research. It analyses
language data but does not attempt to make claims about the characters
and personalities of the people who produced the language, written or
spoken. Discourse analysis is not about obtaining the kind of insights into
individuals which are commonly associated with confession or therapy.

‘Interviews are not an appropriate source of data’ 

Some discourse analysts oppose the use of interviews. They are mainly
discursive psychologists who work in the tradition of conversation
analysis (e.g., Potter and Hepburn, 2005). They argue that an interview is
not a natural form of talk because the interviewer is imposing the topic
on the participants and leading them to talk about things they wouldn’t
otherwise talk about. One counter-argument is that an interview is a
familiar situation in which people ‘naturally’ produce a particular kind of
talk, interview talk. A second, which rather contradicts the first, is that
the talk in interviews is not completely different to talk in other contexts
but will inevitably overlap in many respects, for example, in the ways in
which speakers follow conventions or present new versions of previously
rehearsed accounts (who I am, where I come from, where I work, a funny
thing that happened to me, etc.).
A somewhat different defence of interviews is that the term ‘naturally
occurring’ invokes a research paradigm which most discourse analysts
would reject, that is, one associated with an objective researcher who dis-
covers universal truths in a research situation which is ‘uncontaminated’
by the researcher’s intervention or the expectations of the participants.
This is generally referred to as positivism (see Hammersley and Atkinson,
2007). The contrasting position favoured by most social researchers is that
no form of data collection is ‘natural’ or ‘unnatural’ and any data analysis,
82 What is discourse analysis?

including discourse analysis, is inevitably interpretive, shaped to some

extent by the researcher, even though she or he will attempt to be rigor-
ous and systematic to limit personal bias. An analysis is always partial, and
incomplete.
This alternative position draws attention to data collection and analysis
as situated activities which occur within the context of the project, includ-
ing its socio-historic and cultural context. This suggests that the researcher
needs to take account of the expectations that participants may bring
to an interview or any other form of participation in research, including
to the ‘naturally occurring’ situations. It suggests too that both the data
collection and the data analysis will have meanings particular to the time,
place, participants, and so on. For example, the particular reference and
associations of a word or image may go unrecognized if the researcher
is not familiar with these larger contexts. A practical implication of this
interpretive position is that the discourse researcher needs to collect
additional material as part of the process of familiarizing her or himself
with the context. This material will inform the analysis even though it will
not be data. Chapter 3 noted the collection of such background material
within the four studies discussed.

‘Discourse analysis just produces interpretations,

not facts’
This criticism relates to some of the points made above. It is linked to an
assumption that facts and interpretations can be distinguished, and there-
fore to a notion of truth (see Chapter 2). Statistics, for example, seem to
be uncontroversial because the counting can be checked. They are not, it
seems, just one researcher’s interpretation but an objective measure of the
world as it is. However, a closer examination of the assumption will show
the distinction to be unfounded. As just one point, what is counted must
previously be defined, which involves interpretation and value judgements.
A simple example, discussed by Raghuram (2009), is that official statistics
on immigrants vary considerably depending on which people are deemed
to be immigrants. The same will apply to other categorizations of people.
A further issue of particular concern for discourse analysts derives from
the arguments of Schegloff (1997). His concern was that analysts would bring
their own interests to talk data and freely interpret the talk as being ‘about’
that. Against an analyst who would look at a conversation between a man
 Criticisms of discourse analysis 83

and a woman and decide that it is inevitably ‘about’ gender, Schegloff argued
that a close analysis on CA principles could show whether or not gender
was relevant, or salient, by revealing what the speakers themselves ‘orient
to’ or ‘make relevant’. This is also an argument against ‘reading off’ certain
behaviours as gendered. If a man speaks at the same time as a woman, is this
inevitably evidence of interruption and a performance of a gendered identity
of dominance? Might it not be an example of overlapping talk as a common
form of agreement in the talk of people who position themselves as equals?
Schegloff’s arguments are a useful challenge to the over-free interpreta-
tion of data. Given that any two speakers can be categorized in terms of
multiple differences (e.g., gender, age, sexuality, nationality, place of birth,
occupation, religion, body features such as height, education), how reason-
able is it to claim that their conversation is relevant evidence for a study
of any particular category? On the other hand, there seems no reason to
assume that people may not ‘orient to’ several different issues at once
simultaneously, so that talk may be ‘about’ gender as well as its ostensible
topic. Theorists of intersectionality (e.g., Phoenix and Pattynama, 2006)
argue that categorization is always complex; for example, the relevant iden-
tification will be as a raced, classed, gendered person of a certain age group.
Discourse research on prejudice and racism has shown how categories may
merge, so that ‘culture’ becomes synonymous with ‘race’. More subtly, the
conventional categories of research may be overtaken by new lived catego-
ries of identity and difference, such as the ‘postcodes’ currently invoked in
conflicts between some young people in London. A discourse analyst needs
to be alert to all of these possibilities, to be alive to the possible differences
between participants’ views and her own, and to pause repeatedly to inter-
rogate the assumptions and concerns she brings to the research.
These further issues can be followed up in the sources listed in the
recommended reading list at the end of the book. The more general point
to note is that yes, discourse analysis is interpretive as is other research,
qualitative or quantitative. Part of its theoretical foundation consists of
challenges to the notion of a simple, objective truth.

‘Discourse research doesn’t have

practical applications’
Again, this is a common criticism of qualitative research in general. As with
some of the other criticisms discussed above, it rests on two incorrect, or
84 What is discourse analysis?

at least debatable, assumptions about other kinds of research: that they

can produce factual, non-interpretive knowledge about the social world
and people’s behaviour, and that they can uncover ‘rules’ which enable
prediction and therefore interventions to produce desirable outcomes.
The first point concerns interpretation and was discussed earlier. The
second concerns models of the social world.
A useful starting point is that all research is partial, involving selection
and abstraction. Research findings which have practical applications are
generally based on a predictive model. This reduces complexity to estab-
lish fairly simple cause–effect relationships, often with significant success
(e.g., in many medical treatments). It is appropriate for some, but not all
issues. One way to understand this point is to think about different levels.
For example, a thousand people in a street could be studied as separate
biological organisms, or separate ‘feeling’ beings with their own personal
histories, or, going up a level, as a group functioning as a group (e.g., in
response to an event which creates panic), or, up another level, as located
within an urban or national environment, as part of a population and so
on. The difference here is partly a matter of scale, as if a camera were mov-
ing outwards and upwards, but also of the aspects of the person which
are being considered. How much heat and carbon dioxide do a thousand
bodies radiate? How many, going by population-level statistics, are likely
to be female, or under 25, or unemployed, or overweight? These kinds of
questions have answers which could, potentially, provide a basis for practi-
cal applications, for example, about environmental impact or the provision
of street furniture or policing or facilities like public toilets.
In contrast, discourse researchers embrace complexity and assume
that social phenomena and social worlds are emergent, as discussed in
Chapter 2. They are likely to ask different questions. What do these people
believe, for example, about staying healthy? How does that behaviour
discriminate against (some) others? What different expectations do they
place on women or men? They often explore what a predictive model
omits. Unlike researchers who set out to develop predictive models
which enable practical interventions, discourse researchers are interested
in exploring complex aspects of society and social life which in the first
instance are more likely to enhance understanding than interventions. The
answers to their questions may have applications, but ones which are less
straightforward. For example, Bell’s research, discussed in Chapter 3, might
usefully indicate why some people would resist lifestyle discourses, even
 Criticisms of discourse analysis 85

if they were informed relevant cause–effect relationships that have been

established, perhaps between improved nutrition and lower recurrence
of cancer. (Note: this cause–effect relationship is cited as a hypothetical
example only, not a reference to research findings.) In general, the practical
applications of discourse analytic research often relate to public education
and opinion-forming.

‘Discourse analysis is out of date’ 

The term ‘discourse analysis’ is probably no longer as widely used as it was
a decade or two ago. This is partly because in some areas of the social sci-
ences, the theoretical focus on Foucault’s work, strongly associated with
the term ‘discourse’, has been succeeded by attention to other theorists.
However, as the previous chapters of this book have shown, related meth-
odological approaches remain and continue so one reason why discourse
analysis may appear less prominent is that it is now often conducted under
different labels. Within social psychology, for example, research in the areas
of discursive psychology and rhetorical psychology is a development from
the sources which originally referred to discourse analysis. Some narrative
analysis similarly builds on discourse analysis (e.g., the narrative-discursive
approach used by Stephanie Taylor and Karen Littleton, 2006; 2008). Other
social researchers incorporate an analysis of discourse into a differently
named research approach, such as ethnography or psychosocial research.
Discourse analysis, as it has been presented in this book, is an approach
to talk and text data which derives from certain premises: language is
constitutive, meanings are socially derived and also situated, negotiated
and co-constructed and language use is a functional social practice. These
notions will continue to underpin the work of many researchers, even if
they do not label their work ‘discourse analysis’. This book has argued that
its value as an analytic approach lies in its breadth and flexibility.

Summary 
Chapter 6 has addressed some common criticisms of discourse analysis.
7  Summary

This book has introduced discourse analysis as a social research approach

which involves the close interpretation of language data and language
practices. It has shown that there is no one way of doing discourse analysis.
The common feature of the various forms discussed in the book is that
language data, like talk or documents, is not considered simply for the
information it conveys through description. Instead, an analysis considers
how language works as communication, its situated functions  – what is
‘done’ by talk or written language in a particular context – and the larger
systems or formations of meanings which pre-exist, enable and constrain
any instance of language use. These aspects can be explored in different
ways, deriving to some extent from different theories and academic dis-
ciplines. Chapter 1 introduced discourse analysis and Chapter 2 discussed
some of the theories and issues associated with it. It also introduced some
of the key concerns of discourse analysts and outlined work in social psy-
chology which has contributed to the popularity of discourse analysis as
an empirical research approach.
In Chapter 3, four examples of published discourse research were dis-
cussed in detail to show alternative possibilities for analysis. The functional
nature of language use can be investigated in terms of construction. The
researcher might ask: What is the effect of talking about a problem or issue
in terms of certain images or cause–effect relationships, such as when ill
health is talked about as the consequence of unhealthy living? (This ques-
tion is addressed in one of the studies discussed in the chapter.) Function
can also be explored in an analysis of talk as interaction. A question here
might be: How do people complain? How is complaining ‘done’ in order to
sound plausible and appropriate? And how is this accomplished in conver-
sation, between speakers? Or how is leadership enacted? What leadership
practices are effective in a common workplace situation like chairing a
meeting? Function can also be linked to a certain kind of argument and
rhetoric. Possible questions here might be: How does language function to

87
88 What is discourse analysis?

include or exclude certain categories of people? How does such language

have effects beyond words? How is racism ‘done’, for example, in political
talk?
Researchers who analyse larger systems of meaning may discuss these
in different terms, for example as discourses, Membership Categorization
Devices or discursive resources like interpretative repertoires. Some of
these terms are associated with distinct named approaches to discourse
analysis, such as Critical Discourse Analysis, Conversation Analysis or criti-
cal discursive psychology. This book has not attempted to separate these
completely. Instead, the aim has been to show the assumptions and foci
operating in different studies, regarding these as variants on discourse
analysis as it was defined at the start of the book, that is, as ‘the close
study of language and language use as evidence of aspects of society and
social life’. Researchers developing their own projects are likely to follow
the example of previous work but also shape their own approaches as
appropriate to their topics and investigations.
Why might they use discourse analysis? Chapter 4 set out a number of
possible reasons. For many researchers, the attraction and interest of dis-
course analysis derives from the data. The chapter discussed different types
of discourse data and the most common ways to obtain them. It noted
that some analysts work with already existing ‘found’ data while others
collect new material. The procedures for data collection are not necessar-
ily different to those entailed in other forms of social research which use
language material. Indeed, there is considerable overlap between discourse
analysis, ethnography, narrative analysis and other qualitative approaches.
(Some of this overlap was discussed in previous chapters.) Because so
many researchers are interested in working with interview material, the
chapter discussed some potentially distinctive features around interview-
ing for a discourse analytic project. It also presented an introductory guide
to transcription.
Any empirical research presents challenges. Chapter 5 discusses some
aspects of discourse analytic research which are likely to concern a novice
researcher. In particular, the chapter presented a detailed guide to the
process of analysis, from beginning to work with the data to developing an
interpretation and argument. This section included accounts of the steps
followed on two actual projects, by a researcher working on her doctoral
project and two researchers collaborating on a relatively large project
employing a research assistant.
 Summary 89

Discourse analysis has been a popular form of research for several decades.
During that time it has been subject to criticisms, for instance, that it is
‘just’ about words and that it fails to take account of some other aspects of
people and society. There has also been criticism of the most common form
of data collection associated with it, interviewing. Chapter 6 addressed these
points and others, arguing against the criticisms and showing the continuing
relevance of discourse analysis for social research.
Further reading

New discourse analytic research is likely to be published in the journals

Discourse & Society, Discourse Studies, Critical Discourse Studies, Text,
Narrative Inquiry, Qualitative Research, Qualitative Research in Psychology
and, with a more substantive focus, in Sociological Research Online, The
Sociological Review and many other sociology, social psychology and edu-
cation journals. The reading list below suggests some relevant books which
are likely to be available in a university library. Most are well-established,
even classic texts so not particularly new. It also includes some important
journal articles.
There are many available texts on discourse research and discourse
analysis, some providing the kind of general overview presented in this
book and some with a more specific focus.

Two large collections which cover similar ground to that presented (much
more briefly) in this book are as follows.
Wetherell, M., Taylor, S. and Yates, S. J. (eds) (2001), Discourse: Theory and
Practice, London: Sage.
Wetherell, M., Taylor, S. and Yates, S. J. (eds) (2001), Discourse as Data,
London: Sage.
As the titles indicate, the first collection provides theoretical background
with a combination of new writing and reprints from classic theory sources
and empirical studies. The second collection presents more general guid-
ance on empirical research and a series of ‘workshop’ chapters in which
researchers present examples of data analysis.

Two highly readable texts which provide a fuller introduction to the social
nature of language are as follows.
Fairclough, N. (2001b), Language and Ideology, Harlow: Pearson Education.
Hodge, R. and Kress, G. (1988), Social Semiotics, Cambridge: Polity.

91
92 Further reading

A series of Sociolinguistic Readers published in 1998 provide an introduc-

tion to many of the aspects of discourse analysis and Critical Discourse
Analysis which are of interest to sociolinguists and educationalists. They
include
Cheshire, J. (ed.) (1998), The Sociolinguistics Reader Volume 2: Gender and
Discourse, London: Hodder Arnold.

A general guide to Critical Discourse Analysis is provided by the following.

Wodak, R. and Meyer, M. (eds) (2009), Methods for Critical Discourse
Analysis, London: Sage.

Useful introductions to some of the theorists and theories discussed in

Chapter 2 are provided by the following.
Blackman, L., Cromby, J., Hook, D., Papadopoulos, D. and Walkerdine, V.
(2008), ‘Editorial: Creating subjectivities’, Subjectivity 22: 1–27.
Burr. V. (2003), Social Constructionism, London: Routledge.
Downing, L. (2008), The Cambridge Introduction to Foucault, Cambridge:
Cambridge University Press.
Lynch, M. (2011), ‘Harold Garfinkel obituary: Sociologist who delved into
the minutiae of daily life’, The Guardian 13 July, www.guardian.co.uk/
education/2011/jul/13/harold-garfinkel-obituary.

The following text provides a very clear introduction to ethnomethodology

and a guide to the uses and advantages of different research approaches.
Have, P. t. (2004), Understanding Qualitative Research and
Ethnomethodology, London: Sage.

A critical comparison of conversation analysis and (one form of) discourse

analysis is presented in the following.
Wooffitt, R. (2005), Conversation Analysis and Discourse Analysis: A
Comparative and Critical Introduction, London: Sage.

The practical applications of discourse analysis are discussed in the follow-

ing title.
Willig, C. (1999), Applied Discourse Analysis: Social and Psychological
Interventions, London: Open University Press.
And also in the concluding chapter of one of the collections mentioned
earlier.
 Further reading 93

Taylor, S. (2001), ‘Evaluating and applying discourse analytic research’, in

M. Wetherell, S. Taylor and S. J. Yates (eds), Discourse as Data: A Guide
for Analysis, London: Sage and The Open University.

A debate on the use of interviews is presented in the following articles.

Potter, J. and Hepburn, A. (2005), ‘Qualitative interviews in psychology:
Problems and possibilities’, Qualitative Research in Psychology 2:
281–307.
Smith, J. A., Hollway, W. and Mishler, E. G. (2005), ‘Commentaries on
Potter and Hepburn, “Qualitative interviews in psychology: Problems
and possibilities”’, Qualitative Research in Psychology 2: 309–18.

The most famous transcription notation system was developed by Gail

Jefferson but this is more detailed than many discourse analytic studies
require. Useful discussions on the principles of transcription are contained
in the following.
Hammersley, M. (2010), ‘Reproducing or constructing? Some questions
about transcription in social research’, Qualitative Research 10(5): 553–69.
Nikander, P. (2008), ‘Working with transcripts and translated data’,
Qualitative Research in Psychology 5(3): 225–31.
The second article by Pirjo Nikander particularly addresses issues around
discourse analyses of translated data.

The following sources specifically concern discourse research on gender.

Edley, N. and Wetherell, M. (1995), Men in Perspective: Practice, Power and
Identity, Hemel Hempstead: Prentice Hall/Harvester Wheatsheaf.
Speer, S. A. and Stokoe, E. (2011), Conversation and Gender, Cambridge:
Cambridge University Press.
Weatherall, A. (2002), Gender, Language and Discourse, Hove: Routledge.
Wodak, R. (ed.) (1997), Gender and Discourse, London: Sage.

As this book has indicated, many of the best-known texts and studies on
discourse analysis have been written by social psychologists. An early and
now classic book which introduced discourse analysis into psychology is
the following.
Potter, J. and Wetherell, M. (1987), Discourse and Social Psychology: Beyond
Attitudes and Behaviour, London: Sage.
94 Further reading

Other important sources from social and discursive psychology which

present slightly different positions are as follows.
Benwell, B. and Stokoe, E. (2006), Discourse and Identity, Edinburgh:
Edinburgh University Press.
Billig, M. (1996), Arguing and Thinking: A Rhetorical Approach to Social
Psychology (new ed.), New York: Cambridge.
—(1999b). ‘Whose terms? Whose ordinariness? Rhetoric and ideology in
Conversation Analysis’, Discourse & Society 10(4): 543–82.
Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D. and Radley, A.
(1988), Ideological Dilemmas, London: Sage.
Edwards, D. (1997), Discourse and Cognition, London: Sage.
Edwards, D. and Potter, J. (1992), Discursive Psychology, London: Sage.

The following sources develop new approaches for empirical research

which are broadly based on the forms of discourse analysis discussed in
this book.

(1) Visual research methods

Barker, M., Richards, C. and Bowes-Catton, H. (2012), ‘Visualizing
experience: Using creative research methods with members of sexual
communities’, in C. Phellas (ed.), Researching Non-Heterosexual
Sexualities, Farnham: Ashgate.
Henwood, K., Shirani,F. and Finn, M. (2011), ‘So you think we’ve moved,
changed, the representation got more what? Methodological and
analytical reflections on visual (photo-elicitation) methods used in the
Men as Fathers study’, in P. Reavey (ed.), Visual Methods in Psychology:
Using and Interpreting Images in Qualitative Research, London:
Routledge.

(2) Narrative analysis

Taylor, S. and Littleton, K. (2006), ‘Biographies in talk: A
narrative-discursive research approach’, Qualitative Sociology Review
II(1).

(3) Psychosocial research

Hollway, W. and Jefferson, T. (2000), Doing Qualitative Research Differently:
Free Association, Narrative and the Interview Method, London: Sage.
Glossary

Analysis
In this book, analysis refers to the close investigation of empirical material
(data) as potential evidence for the workings of society. It is therefore a
process of exploration with the aim of discovery. The researcher studies
the details of data and then considers their wider implications. There are
different ways of doing this. Any qualitative analysis, including discourse
analysis, tends to be time-consuming. It also generally requires the
researcher to develop her own explanation of the relationship between
the part (the details of the data) and the whole (the social phenomenon or
issue which is being investigated).

Construction
This term is linked to the assumption that language is not a neutral
vehicle of communication. To talk, or write, about something, or some-
one, involves choices about words and perspectives and associations.
Alternative versions are always possible. For example, a street could be
described in terms of its buildings, or trees and gardens, or the people
who live there, or of different activities associated with it. Each descrip-
tion constructs the street differently. Discourse analysts are interested in
construction as a process, and in the constructed, and co-constructed,
nature of people’s accounts, and the implications or effects of particular
discursive constructions.

Conversation analysis 
Conversation analysis (CA) is based on the work of the sociologist Harvey
Sacks, and its subsequent development by Emanuel Schegloff and Gail
Jefferson. It involves the close analysis of talk as interaction, following the
principles of ethnomethodology (see Chapter 2). The focus is on details
of speakers’ utterances and responses within the context of spoken

95
96 Glossary

interaction. Ordinary talk is investigated as a skilled, rule-governed, joint

activity in which meanings are established and changed through the
minutiae of what is said or not said, and how it is said (e.g., the mean-
ings conferred by pauses or by the talk of two speakers overlapping).
Conversation analysts generally work with very detailed transcripts.
They focus on the talk itself, excluding background information about
the speakers and context unless this is directly invoked in what is said.
They regard their analyses as ‘technical’, that is, as following closely from
the evidence of the talk itself, which they contrast with more ‘interpre-
tive’ analyses by other discourse analysts (see Schegloff, 1997; Wetherell,
2001).

Critical Discourse Analysis

This term refers to an interdisciplinary approach to the analysis of talk
and texts which considers the workings of power in society and the role
of language in establishing and perpetuating dominance. Some of the
best-known writers associated with Critical Discourse Analysis (CDA) are
Teun van Dijk, Norman Fairclough and Ruth Wodak. It generally involves
analysis on a number of levels, from linguistic features of a text to its loca-
tion within a larger social context.

Critical discursive psychology 

This term was formulated by Margaret Wetherell (1998) to describe the
approach she developed following her initial work on discourse analysis
in psychology (with Jonathan Potter, 1987). Critical discursive psychol-
ogy (CDP) focuses particularly on ways in which discursive work in
talk is implicated with social inequalities and the exercise and contest
of the workings of power within larger social contexts. A key aspect of
the approach is the analysis of discursive resources, such as interpreta-
tive repertoires (see definition). Wetherell’s own research has included
studies of masculinity and, more recently, affect. Other researchers
whose work is associated with this approach are Nigel Edley and Sarah
Seymour-Smith.

Discourse
This is the key term of the book yet also one of the most difficult to
define. It can refer to communication, specific language practices and a
 Glossary 97

complex set or system of meanings which is a resource for language use. As

Chapter 2 discusses, many writers have developed definitions of discourse
which encompass the foci of different analytic approaches.

Discourse analysis 
Discourse analysis is the close study of language and language use as
evidence of aspects of society and social life. It can involve the analysis
of language data as communication, practice or selective constructions
derived from accrued social meanings.

Discursive psychology 
Many of the most influential texts on discourse analysis and related forms
of empirical research, such as psychosocial research, have been writ-
ten by social psychology academics (e.g., Hollway and Jefferson, 2000).
Discourse analysis has led to the development of a new area of psychol-
ogy, discursive psychology, sometimes referred to as a sub-discipline.
Academics working in this area have drawn on theories of discourse and
discourse practice in order to challenge many traditional psychological
theories of the person and also the kind of ‘mental’ activities (such as
remembering) which are studied by some other psychologists, such as
cognitive psychologists.
Discursive psychology encompasses different assumptions and methods,
and many discursive psychologists would no longer refer to themselves
as discourse analysts. There are areas of overlap but also some points of
dispute, for example, about the ‘right’ kinds of data or the relationship
between ‘technical’ and ‘interpretive’ analyses.
Critical discursive psychologists such as Margaret Wetherell and Nigel
Edley are particularly interested in the issues around power which are
associated with the work of Michel Foucault. Many other discursive psy-
chologists, including Jonathan Potter, Derek Edwards and Elizabeth Stokoe
work more closely in the tradition of ethnomethodology and conversation
analysis.

Discursive resource
One point of investigation for discourse analysts can be the implications
and consequences of discourses, that is, the ways in which ideas, repre-
sentations (in various media), words and ways of speaking have effects
98 Glossary

in the social world. For example, to call someone a ‘good mother’ is to

invoke a category which is linked to certain words and activities (includ-
ing what it excludes, that is, all that is associated with bad mothering).
It is not a neutral categorization but affective, carrying strong emotional
associations and potential implications. These meanings are part of the
shared knowledge of members of society, largely carried in language, but
not exclusively confined to it. By using a term or speaking in a certain
way, a particular speaker can draw on all of this shared knowledge; it
is a discursive resource for a speaker. Part of the purpose of a discourse
analytic study can be to investigate discursive resources. Some analysts
discuss discourses as resources. Others may focus on a narrower set or
clump of words/ideas/ways of speaking, such as a narrative resource or
interpretative repertoire.

Identity
Discourse analysis is often employed in research into identities. At its
simplest, identity refers to who someone ‘is’, with all the range of possibili-
ties that can encompass, from the person’s name to a detailed description
including a biography, to the multiple roles that person takes in different
situations and relationships. Social researchers may study major identity
categories such as gender, race, age, nationality and also more specific iden-
tities associated with particular contexts or issues (see ‘subject position’).
Discourse analytic research on identities, especially by social psychologists,
has addressed some of the major problems of the social sciences, such as
how every person is both an individual actor with a unique identity and
sense of who she or he is, and at the same time a component of a larger
society, sharing identities with many others and apparently shaped by the
social environment to be typical of her or his class, generation, nationality
and so on.

Identity work
Discourse researchers may study identities in talk, such as the subject
positions available to a speaker, or how identities are constructed in talk,
for example, when a speaker describes herself or others in a certain way.
This kind of talk about identity is often referred to as identity work. The
concept is linked to the assumption that identities are partly given by
 Glossary 99

contexts (or ‘situated’) but also actively contested, constructed, claimed

and taken up in ongoing everyday talk and other life practices and
interactions.

Ideological dilemma
The notion of an ideological dilemma (or, more simply, a dilemma) in
discursive work comes from the work of Michael Billig and others (Billig
et al., 1988). They suggested that commonsense logic is often inherently
contradictory, so that speakers may find themselves making claims or
arguments which they recognize as inconsistent; they then try to repair
the inconsistency or otherwise resolve the problem.

Interpretative repertoire
Another widely discussed discursive resource is an interpretative rep-
ertoire. This has been defined as a ‘relatively coherent way . . . of talking
about objects and events in the world’ (Edley, 2001, p. 198) or, in a some-
what broader definition, ‘a culturally familiar and habitual link of argument
comprised of recognizable themes, commonplaces and tropes’ (Wetherell,
1998, p. 400). It is therefore similar to ‘a discourse’ but is perhaps more spe-
cific and particularly associated with the discussion of data. Researchers
who discuss interpretative repertoires are often interested in multiple and
conflicting resources and their different implications.

Narrative resource
One example of a discursive resource could be an established narrative
sequence such as the sequence of stages which supposedly constitute a
‘normal’ relationship and life course: courtship, coupling, parenting, etc.
This is not descriptive, since many or even most people’s lives do not
unfold like this. However, it is normative because it shapes expectations,
biographical accounts and identities (for a fuller discussion, see Reynolds
and Taylor, 2005), influencing how people talk about their own lives and
the lives of others and reinforcing certain social values, including the
positive valuing of heterosexuality (i.e., it is heteronormative). One form of
discourse analysis is a narrative analysis (or narrative-discursive analysis; for
example, Taylor and Littleton, 2006) of narrative resources.
100 Glossary

Rhetorical work
This concept comes from the work of Michael Billig (1987; 1999) who sug-
gests that talk is shaped, sometimes quite subtly, by a wider context of
media debates, political discussion and social issues. A speaker’s rhetori-
cal work is a reference to this context, perhaps in the form of a response
to a challenge or criticism which has not been made in the immediate
interaction. For example, a comment which begins ‘I’m not a racist but . . .’
is acknowledging the possibility that what will be said could be heard
as racist or prejudiced and that these positions are widely criticized. A
comment like this indicates how the speaker is simultaneously situated in
multiple overlapping contexts, as a party to the ongoing conversation, a
member of the larger society, an audience to current media debates and
so on.

Social constructionism
Within psychology, social constructionism refers to the tradition of theory
and empirical research which emphasizes how people’s understandings of
themselves and their worlds are not based on how things ‘are’ but on ideas
and meanings current within the society. This is consistent with some
of the premises of discourse analysis, such as the constitutive nature of
language.

Subject position
The concept of ‘a discourse’ carries implications for the nature of the
(social) world and also the people within it. For example, a discourse of
healthy eating creates certain identities, such as a healthy eater, a bad eater
and a good parent who provides a healthy diet for the family. Conversely,
the identity of ‘bad eater’ does not exist without a discourse of healthy
eating. The identity associated with a discourse like this is called a subject
position. It is a ‘position’ because it is created by the external framework
of the discourse. One possible focus of discourse analysis is the shifting
subject positions or positionings which are taken up in talk, often quite
subtly, for example, in the way someone speaks (e.g., as an authority) or is
spoken to, how a speaker describes herself and so on.

Subjectivity
This term can combine several different ideas. The contrast between the
objective and subjective is, roughly, between the outsider and insider
 Glossary 101

point of view, so that, conventionally, objective knowledge is based on

neutral, unemotional observation and subjective knowledge as based on
experience, enriched by emotion and personal involvement but perhaps
liable to bias. (Most contemporary social scientists would challenge this
neat division between the subjective and the objective.) A different notion
is of the subject as a person controlled or constrained by larger forces, as in
references to a ‘political subject’ who is in the power of a ruler or govern-
ment. Social theorists have explored the ways in which people are shaped
by, or subject to, the wider socio-historic context. With these associations,
subjectivity can refer to the sense and experience of ‘who I am’ which is
produced through a person’s position, experience and interactions as a
member of society. It has a similar reference to identity but emphasizes the
location of a person within larger social systems, subject to the workings
of power within those systems. The concept is associated with processes
by which ideas and value systems come to be part of people’s personal
beliefs and morality, that is, from an objective to subjective status. This is
sometimes contrasted with ‘identity’ as the objective, outsider view of the
same person, for example, as officially categorized by gender, age, occupa-
tion and so on.

Transcription
For the purposes of analysis, discourse researchers generally convert
spoken language and other ‘lived’ interactions into a written form. This
can be done by taking notes during the fieldwork or immediately after-
wards. Mostly commonly, however, it involves working from an audio- or
audiovisual recording to make a detailed record called a ‘transcript’. This
usually looks rather like the script of a play, listing each speaker and the
main words spoken. Other details will be included depending on the
researcher’s interest and the analytic approach; the same recording can
be transcribed in many different ways. Transcription is always time-
consuming. It is inevitably selective and is therefore inseparable from the
analytic process through which the researcher focuses on and interprets
particular features.

Trouble
The concept of ‘trouble’ is particularly associated with discourse analytic
research on identities. Discourse analysis necessarily involves the investiga-
tion of complexity, for example, because of the multiplicity of available
102 Glossary

discourses and discursive resources and associated subject positions. This

complexity gives rise to conflicts and contradictions, for example, between
how a speaker positions herself and is positioned by others, or between
a positive and negative valuing of the same identity. This kind of trouble
prompts discursive repair, for example, through explanations. Trouble in
discursive work is interesting for the analyst because it can draw attention
to different discursive resources in play and it can also signal the positive
and negative values attached to certain identities.
Bibliography

Abell, J., Condor, S., Lowe, R. D., Gibson, S. and Stevenson, C. (2007), ‘Who
ate all the pride? Patriotic sentiment and English national football
support’, Nations and Nationalism 13: 97–116.
Bell, K. (2010), ‘Cancer survivorship, mor(t)ality and lifestyle discourses on
cancer prevention’, Sociology of Health and Illness 32: 349–64.
Benwell, B. and Stokoe, E. (2006), Discourse and Identity, Edinburgh:
Edinburgh University Press.
Billig, M. (1987), Arguing and Thinking: A Rhetorical Approach to Social
Psychology, Cambridge: Cambridge University Press.
—(1995), Banal Nationalism, London: Sage.
—(1996), Arguing and Thinking: A Rhetorical Approach to Social
Psychology, Cambridge: Cambridge University Press.
—(1999), ‘Whose terms? Whose ordinariness? Rhetoric and ideology in
Conversation Analysis’, Discourse & Society 10: 543–82.
Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D. and Radley, A.
(1988), Ideological Dilemmas, London: Sage.
Bruner, J. (1990), Acts of Meaning, Cambridge, MA: Harvard University
Press.
Butler, J. (1990), Gender Trouble, London: Routledge.
Byford, J. (2006), ‘Distinguishing “anti-Judaism” from “antisemitism”:
Recent championing of Serbian Bishop Nikolaj Velimirović’, Religion,
State & Society 34(1) : 7–31.
Condor, S. (2011), ‘Rebranding Britain? Ideological dilemmas in political
appeals to British multiculturalism’, in M. Barrett, C. Flood and J. Eade
(eds), Nationalism, Ethnicity, Citizenship: Multidisciplinary Perspectives,
Newcastle: Cambridge Scholars.
Davies, B. and Harré, R. (1990), ‘Positioning: The discursive production of
selves’, Journal for the Theory of Social Behaviour 20(1): 43–63.

103
104 Bibliography

Dixon, J., Reicher, S. and Foster, D. (1997), ‘Ideology, geography, racial

exclusion: The squatter camp as “blot on the landscape”’, Text 17(3):
317–48.
Edley, N. (2001), ‘Analysing masculinity: Interpretative repertoires,
subject positions and ideological dilemmas’, in M. Wetherell, S. Taylor
and S. J. Yates (eds), Discourse as Data: A Guide for Analysis, London:
Sage.
Edley, N. and Wetherell, M. (1995), Men in Perspective: Practice, Power and
Identity, Hemel Hempstead: Prentice Hall/ Harvester Wheatsheaf.
—(1997), ‘Jockeying for position: The construction of masculine identities’,
Discourse & Society 8: 203–17.
Edwards, D. (1997), Discourse and Cognition, London: Sage.
—(1998), ‘The relevant thing about her: Social identity categories in use’ in
C. Antaki and S. Widdicombe (eds), Identities in Talk, London: Sage.
Edwards, D. and Middleton, D. (1986), ‘Joint remembering: Constructing
an account of shared experience through conversational discourse’,
Discourse Processes 9: 423–59.
Edwards, D. and Potter, J. (1992), Discursive Psychology, London: Sage.
Edwards, D., Ashmore, M. and Potter, J. (1995), ‘Death and furniture:
The rhetoric, politics, and theology of bottom line arguments against
relativism’, History of the Human Sciences 8(2), 25–49.
Fairclough, N. (2001a), Language and Power, 2nd edn, London: Longman.
—(2001b), Language and Ideology, Harlow: Pearson Education Limited.
Foucault, M. (2001), Madness and Civilization, London: Routledge.
Gee, J. (1992), The Social Mind: Language, Ideology and Social Practice, New
York: Bergin and Garvey.
Gergen, K. J. (1985), ‘The social constructionist movement in modern
psychology’, American Psychologist 40: 266–75.
Gilbert, G. and Mulkay, M. (1984), Opening Pandora’s Box: A Sociological
Analysis of Scientists’ Discourse, Cambridge: Cambridge University
Press.
Gill, R. (2008), ‘Empowerment/sexism: Figuring female sexual agency in
contemporary advertising’, Feminism & Psychology 18: 35–60.
Goodwin, C. (1994), ‘Professional vision’, American Anthropologist 96:
606–33.
Hall, S. (1992), ‘The west and the rest’, in S. Hall and B. Gieben (eds),
Formations of Modernity, Cambridge: Polity Press/The Open
University.
 Bibliography 105

Hammersley, M. (1998), ‘Get real! A defence of realism’, in P. Hodkinson

(ed.), The Nature of Educational Research: Realism, Relativism, or
Post-Modernism? Manchester: Manchester Metropolitan University.
—(2010), ‘Reproducing or constructing? Some questions about
transcription in social research’, Qualitative Research 10(5): 553–69.
Hammersley, M. and Atkinson, P. (2007), Ethnography: Principles in
Practice, London and New York: Routledge.
Harré, R. and Gillett, G. (1994), The Discursive Mind, Thousand Oaks,
California: Sage.
Have, P. t. (2004), Understanding Qualitative Research and
Ethnomethodology, London: Sage.
Herman, D. (2007), ‘Storytelling and the sciences of the mind: Cognitive
narratology, discursive psychology, and narratives in face-to-face
interaction’, Narrative 15: 306–34.
Hollway, W. (1998), ‘Gender differences and the production of subjectivity’
in J. Henriques, W. Hollway, C. Urwin, C. Venn and V. Walkerdine (eds),
Changing the Subject: Psychology, Social Regulation and Subjectivity,
London and New York: Routledge.
Hollway, W. and Jefferson, T. (2000), Doing Qualitative Research Differently:
Free Association, Narrative and the Interview Method, London: Sage.
Horton-Salway, M., Montague, J., Wiggins, S., and Seymour-Smith, S.
(2008), ‘Mapping the components of the telephone conference: An
analysis of tutorial talk at a distance learning institution’, Discourse
Studies 10(6): 737–58.
Jefferson, G. (1992), (ed.), Harvey Sacks: Lectures on Conversation,
Cambridge, MA: Blackwell.
Jones, R. L. (2003), ‘Older women talking about sex: A discursive analysis’,
PhD Thesis, Milton Keynes: The Open University.
Kress, G. (2001), ‘From Saussure to critical sociolinguistics: The turn
towards a social view of language’, in M. Wetherell, S. Taylor, and
S. Yates (eds), Discourse Theory and Practice, London: Sage.
Lawler, S. (2008), Identity: Sociological Perspectives, Cambridge: Polity.
McAvoy, J. (2009), ‘Negotiating constructions of success and failure:
Women in mid-life and formations of subject, subjectivity and identity’,
PhD thesis, Milton Keynes: The Open University.
Mulkay, M. (1985), The Word and the World: Explorations in the Form of
Sociological Analysis, London: Allen Unwin.
106 Bibliography

Nikander, P. (2012), ‘Interviews as discourse data’, in J. A. Holstein and

J. Gubrium (eds), Handbook of Interview Research, London: Sage.
Ochs, E. (1979), ‘Transcription as Theory’, in E. Ochs and B. Schieffelin (eds),
Developmental Pragmatics, London: Academic Press.
Onions, C. (ed.) (1965), The Shorter Oxford English Dictionary, Oxford:
Clarendon Press.
Phoenix, A. and Pattynama, P. (2006), ‘Intersectionality’, European Journal
of Women’s Studies, 13: 187–92.
Potter, J. and Hepburn, A. (2005), ‘Qualitative interviews in psychology:
Problems and possibilities’, Qualitative Research in Psychology 2:
281–307.
Potter, J. and Wetherell, M. (1987), Discourse and Social Psychology: Beyond
Attitudes and Behaviour, London: Sage.
Raghuram, P. (2009), ‘Migration: Changing, connecting and making places’,
in S. Taylor, S. Hinchliffe, J. Clarke and S. Bromley (eds), Exploring Social
Lives, Milton Keynes: The Open University.
Reynolds, J. (2006), ‘Patterns in the telling: Single women’s intimate
relationships with men’, Sociological Research Online 11(3).
—(2008), The Single Woman: A Discursive Investigation, Hove: Routledge.
Reynolds, J. and Taylor, S. (2005), ‘Narrating singleness: Life stories and
deficit identities’, Narrative Inquiry 15: 197–215.
Rose, N. (1989), Governing the Soul: The Shaping of the Private Self, London
and New York: Routledge.
—(1996), Inventing Our Selves: Psychology, Power and Personhood,
Cambridge: Cambridge University Press.
Scharff, C. (2008), ‘Doing class: A discursive and ethnomethodological
approach’, special issue: Class and discourse, Critical Discourse Studies
5(4), 331–43.
—(2011), ‘Towards a pluralist methodological approach: Combining
performativity theory, discursive psychology and theories of affect’,
Qualitative Research in Psychology 8(2): 210–21.
Schegloff, E. A. (1997), ‘Whose text? Whose context?’ Discourse & Society 8:
165–87.
Scully, M. (2009), ‘Plastic and Proud? Discourses of Authenticity among the
Second-Generation Irish in England’, Psychology & Society 2(2): 124–35.
—(2010), ‘Local identification and authenticity among the Irish diaspora
in England’, in L. De Pretto, G. Macri and C. Wong (eds), Diasporas:
Revisiting & Discovering, Oxford: Inter-Disciplinary Press.
 Bibliography 107

Seymour-Smith, S. (2008), ‘Illness as an occasion for storytelling: Social

influences in narrating the masculine self to an unseen audience’, in
C. Horrocks, K. Milnes, B. Roberts and D. Robinson (eds), Narrative,
Memory and Life Transitions, Huddersfield: Huddersfield University
Press.
Seymour-Smith, S., Wetherell, M. and Phoenix, A. (2002), ‘My wife ordered
me to come!: A discursive analysis of doctors’ and nurses’ accounts of
men’s use of general practitioners’, The Journal of Health Psychology
7(3): 253–68.
Smith, J. A., Hollway, W. and Mishler, E. G. (2005), ‘Commentaries on
Potter and Hepburn, “Qualitative Interviews in psychology: Problems
and possibilities’’’, Qualitative Research in Psychology 2: 309–18.
Speer, S. A. and Stokoe, E. (2011), Conversation and Gender, Cambridge:
Cambridge University Press.
Stokoe, E. H. (2003), ‘Mothers, Single Women and Sluts: Gender, Morality
and Membership Categorization in Neighbour Disputes’, Feminism &
Psychology 13: 317–44.
—(2010), ‘I’m not gonna hit a lady: Membership categorization and men’s
denials of violence towards women’, Discourse & Society 21: 59–82.
Taylor, S. (2001a), ‘Evaluating and applying discourse analytic research’, in
M. Wetherell, S. Taylor and S. J. Yates (eds), Discourse as Data: A Guide
for Analysis, London: Sage and The Open University.
—(2001b), ‘Locating and conducting discourse analytic research’, in
M. Wetherell, S. Taylor and S. J. Yates (eds), Discourse as Data: A Guide
for Analysis, London: Sage and The Open University.
—(2006), ‘Attitudes’, in D. Landridge and S. Taylor (eds), Critical Readings
in Social Psychology, London: Open University Press.
—(2007), ‘Narrative as construction and discursive resource’, in
M. Bamberg (ed.), Narrative – State of the Art, Amsterdam/
Philadelphia: John Benjamins.
—(2010), Narratives of Identity and Place, London: Routledge Psychology
Press.
—(2011), ‘Negotiating oppositions and uncertainties: Gendered conflicts
in creative identity work’, Feminism & Psychology 21(3): 354–71.
—(2012), ‘“One participant said . . .”: The implications of quotations from
biographical talk’, Qualitative Research 12(4): 388–401.
Taylor, S. and Littleton, K. (2006), ‘Biographies in talk: A
narrative-discursive research approach’, Qualitative Sociology Review II.
108 Bibliography

—(2008), Creative Careers and Non-Traditional Trajectories, National Arts

Learning Network.
—(2012), Contemporary Identities of Creativity and Creative Work,
Farnham: Ashgate.
Van Den Berg, H., Wetherell, M. and Houtkoop, H. (eds) (2003), Analyzing
Racist Discourse, Cambridge: Cambridge University Press.
van Dijk, T. (1998), ‘Principles of critical discourse analysis’, in J. Cheshire
and P. Trudgill (eds), The Sociolinguistics Reader Volume 2: Gender and
Discourse, London: Arnold.
Wertsch, J. (1990), Voices in the Mind: A Sociocultural Approach to
Mediated Action, London: Harvester Wheatsheaf.
Weatherall, A. (2002), Gender, Language and Discourse, Hove: Routledge.
Wetherell, M. (1998), ‘Positioning and interpretative repertoires:
Conversation analysis and post-structuralism in dialogue’, Discourse &
Society 9: 387–412.
—(2001), ‘Debates in discourse research’, in M. Wetherell, S. Taylor and S. J.
Yates, Discourse Theory and Practice, London: Sage.
—(2003), ‘Paranoia, ambivalence and discursive practices: Concepts of
position and positioning in psychoanalysis and discursive psychology’,
in R. Harré and F. Moghaddam (eds), The Self and Others: Positioning
Individuals and Groups in Personal, Political and Cultural Contexts,
Westport, CT: Praegar.
—(2012), Affect and Emotion: A New Social Science Understanding,
London: Sage.
Wetherell, M. and Potter, J. (1992), Mapping the Language of Racism,
Hemel Hempstead: Harvester Wheatsheaf.
Wiggins, S. (2004), ‘Good for ‘you’: Generic and individual healthy eating
advice in family mealtimes’, Journal of Health Psychology 9: 535–48.
Willig, C. (1999), Applied Discourse Analysis: Social and Psychological
Interventions, London: Open University Press.
Wodak, R. (2007), ‘What is critical discourse analysis? Ruth Wodak in
conversation with Gavin Kendall’, Forum: Qualitative Social Research
8(2): 29.
Wodak, R., Kwon. W. and Clarke, I. (2011), ‘“Getting people on board”:
Discursive leadership for consensus building in team meetings’,
Discourse & Society 22(5): 592–644.
 Bibliography 109

Wooffitt, R. (2005), Conversation Analysis and Discourse Analysis: A

Comparative and Critical Introduction, London: Sage.
Yates, S. (2001), ‘Researching internet interaction: Sociolinguistics and
corpus analysis’, in M. Wetherell, S. Taylor and S. Yates (eds), Discourse
as Data: A Guide for Analysis, London: Sage.
Index

advertisements  25, 58, 60, 77 Condor, Susan  18, 57, 94

anonymity  36, 64, 65, 71 consensus  5, 14, 43–8
anti-Semitism  34–8 construction  4, 22, 23, 25, 27, 34–8, 46,
apologizing  12 54, 55, 56, 62, 71, 85, 87
application  30, 36, 39, 43, 53, 83–5, 92 context  3, 4, 8, 9, 10, 12, 13, 14, 19, 23, 25,
argument  7, 74–5 37, 39, 40, 41, 43, 44, 46, 47, 54, 55,
attitude  23, 57 57, 58, 61, 74, 75, 78, 81, 82, 87, 96,
Austin, J.L.  12–13, 22 98, 99, 100, 101
Australia  44 conversation analysis  11–12, 39–43,
50, 60–1, 80, 81, 82–3, 88, 92, 93,
Bell, Kirsten  29–34, 48, 49, 50, 54, 56, 58, 95–6
70, 84 corpus linguistic analysis  45, 69
Benwell, Bethan  21 counselling  55
Billig, Michael  18, 38, 55, 58, 70, 94 counting  45, 82
bonding  46 Critical Discourse Analysis  14–15, 16, 19,
Bruner, Jerome  25 43–8, 50–1, 88, 92
business  43–8 critical discursive psychology  16, 26, 88,
Butler, Judith  22 96, 97
Byford, Jovan  32, 34–8, 45, 50, 72 critical linguistics  15, 17
culture  20–3, 30, 35, 37, 40, 78, 83
cancer  29–34
CAQDAS  68, 69 data collection  29, 30, 36–7, 41, 44, 49,
case study  36, 38 50, 54–5, 59–63, 81–2, 88
categories and categorisation  3, 9, 14, 20, data selection  41, 60, 88
29, 31, 35, 37, 38–43, 46, 48, 50, 55, dialect  14, 15
58, 60, 72, 74, 75, 82, 83, 88, 98 dilemma, ideological  18, 29, 60, 94, 99
Christian church, Orthodox  36, 38 directing  46
Clarke, Ian  43–8 discourse  7, 15–17, 48–9, 97
codes  13, 31, 70 discursive formation  9, 15
communication  3, 4, 7, 8, 12, 13–15, 17, discursive leadership strategies  43–8
18, 23, 27, 38, 49, 72, 77, 78, 79, 87, discursive psychology  17, 19, 23–6,
95, 97 38–43, 81–2, 85, 88, 94, 97
communication, three part model  distributed self  25
13–14 dominance  3, 14, 19, 83, 96
complaining  87 dominant discourses  70

111
112 Index

Edley, Nigel  16, 22, 26, 55, 93, 96 identity  3, 10, 17, 20–3, 24, 25, 26, 35,
Edwards, Derek  55, 78, 94, 97 39, 46, 55–6, 58, 67, 79, 83, 93,
emotion  8, 12, 13, 24, 26, 43, 77, 79, 80, 98–9
98, 101 image(s)  8, 25, 48, 49
encouraging  46 inconsistency  18, 34, 35, 99
ethics  58, 59, 61, 64, 65 indexicality  10, 14
ethnography  29–34, 43–8, 50, 60, 73, interaction  3, 11, 12, 22, 24, 34, 38–43,
85, 88 43–7, 54, 58, 61, 62, 64, 68, 70, 72,
ethomethodology  11, 12, 22, 26, 39, 54, 75, 80, 87, 95, 96, 99, 100, 101
92, 96, 97 interviews  3, 5, 12, 23, 30, 31, 32, 34, 36,
evidence  2, 3, 4, 7, 8, 10, 21, 24, 27, 30, 32, 43, 44, 45, 49, 53, 54, 56, 57, 58, 59,
39, 46, 54, 56, 57, 68, 71, 74, 75, 79, 60, 61–3, 65, 68, 70, 71–4, 80, 81–2,
83, 88, 95, 96, 97 88, 93, 94
expectations  20, 42, 62, 81, 82, 84, 99 investment  26

Fairclough, Norman  2, 16, 45, 77, 91, 96 Jefferson, Gail  11, 80, 93, 95
femininity  42 Jefferson, Tony  26, 56, 81, 97
fieldnote(s)  30, 31, 32, 44, 50, 58, 60, 71, 79
focus group  3, 33, 58, 59, 63 knowledge  17–20
formality  2, 8, 15 Kress, Gunther  14, 15, 91
Foucault, Michel  9–10, 15–16, 17, 34, Kwon, Winston  43–8
85, 92, 97
found data  5, 36, 38, 43, 53, 59, 60, 65, 88 language 
as constitutive  10, 24, 39, 56,
Garfinkel, Harold  11, 92 85, 100
Gee, James  16, 20 as functional  4, 12, 13, 19, 35, 39,
gender  2, 22, 23, 25, 32, 38–43, 49, 76, 83, 40, 42, 43, 45, 46, 56, 57, 72, 73,
93, 101 85, 87
gender identity  22, 39, 42, 54, 55, 83, 98 as referential  56
generalization  43, 49, 55, 68, 75 as transparent  56, 57
Gergen, Kenneth  19 laughter  70
Giddens, Anthony  26 lifestyle discourse(s)  29–48, 84
Gilbert, Nigel  18, 19 linguistics  1, 2, 3, 15, 17, 23, 44, 45, 65,
Gill, Rosalind  25, 58 92, 96
Goodwin, Charles  25 Littleton, Karen  56, 58, 70, 73, 80, 85,
governance  33, 34 94, 100
governmentality  17, 20, 34
grammar  3, 13, 15, 37, 46, 64–5, 79 masculinity  16, 55, 96
medical discourse(s)  30, 31, 32
Hammersley, Martyn  57, 63, 81, 93 meetings  5, 31, 43–8, 54, 58, 59, 60,
health  29–34, 48, 54, 78, 79, 84, 87, 88 73, 87
health promotion discourse(s)  30, 33, Membership Categorization
48, 88 Analysis  38–43, 50, 55, 60, 88
hegemony  19, 34 mental state  12, 19, 23, 24, 47, 48, 57,
Hollway, Wendy  26, 56, 81, 97 97
 Index 113

morality  5, 33, 38–43, 101 remembering  12, 24, 25, 65, 97

Mulkay, Michael  18, 19 repertoire  16, 24, 63, 71, 88, 96, 98, 99
research question  1, 59, 74
narrative  16, 22, 26, 56, 63, 70, 71, 85, 88, resource, discursive  4, 16, 17, 22, 24, 43,
94, 98, 99, 100 55, 56, 57, 62, 63, 70, 73, 79, 88, 96,
national identity  18, 21, 35, 36, 38, 54, 55, 97, 98, 99, 100, 102
56, 58, 83, 84, 98 Reynolds, Jill  22, 26, 56, 99
neighbours  43–8 rhetoric  5, 29, 34–8, 45, 48, 70, 72, 85, 87,
New Zealand  35 94, 100
news media  14, 19, 36, 40, 54, 58, 59, 60, rhetorical psychology  85
65, 77 Rose, Nikolas  20, 24, 33, 34

online material  58, 60 Sacks, Harvey  11, 40, 61, 80, 95

oral history  65 Schegloff, Emanuel  40, 82, 83, 95, 96
science  18
parenting  20, 98, 99 Scully, Marc  56
performance  12, 17, 22, 39, 44, 83 sentences  2, 65
political discourse(s)  5, 14, 18, 34–8, 40, Serbia  34–8
49, 54, 80, 88, 100 Seymour Smith, Sarah  26, 56, 96
positivism  81 social constructionism  25, 56,
Potter, Jonathan  23, 24, 35, 37, 81, 93, 94, 92, 100
96, 97 social psychology  1, 3, 7, 19, 23, 26, 34,
power  2, 9, 10, 14, 15, 17–20, 24, 25, 36, 38, 57, 81, 85, 87, 91, 93, 97, 98
26, 35, 37, 38, 42, 44, 48, 55, 77, 96, sociolinguistics  1, 17, 23, 44, 92
97, 101 sociology  1, 3, 44, 91
practice  3, 8, 9, 10, 11–13, 16, 17, 21, statistics  45, 68, 79, 82, 84
22, 23, 24, 25, 27, 29, 30, 35, 39, 47, Stokoe, Elizabeth  5, 11, 21, 32, 38–43,
48, 54, 55, 56, 75, 78, 80, 81, 85, 87, 44, 48, 49, 50, 55, 58, 59, 60, 64, 93,
97, 99 94, 97
prediction  84 subject position  21, 22, 24, 25, 26, 32, 33,
prejudice  5, 15, 34–8, 57, 70, 79, 83, 100 42, 60, 72, 79, 83, 98, 100, 102
prounciation  15 subjectification  24, 25, 33
psy disciplines  20 subjectivity  26, 56, 101
psychosocial research  26, 56, 79, 81, 85,
94, 97 team research  69
punctuation  79 texts  2, 15, 36, 45, 53, 54, 58, 77, 95, 96
transcript(s)  5, 31, 44, 45, 47, 50, 53, 58,
racism  5, 34–8, 54, 79, 83, 88, 100 59, 62, 63–6, 69, 71, 73, 80, 88, 93,
Raghuram, Parvati  82 96, 101
re/committing  47 transmission  13
recording  12, 31, 36, 40, 41, 44, 50, 58, trouble  102
59, 60, 61, 62, 63, 64, 65, 68, 73, truth  10, 17–20, 35, 57, 81, 82–3
101
register  14, 15, 18 van Dijk, Teun  14, 19, 35, 77, 96
reliability  68 Velimirović, Bishop Nikolaj  34–8
114 Index

version  18, 19, 35, 41, 62, 65, 70, 81, 95 Wetherell, Margaret  16, 22, 23, 24, 26, 35,
visual research methods  77, 94 37, 55, 57, 81, 91, 93, 96, 97, 99
vocabulary  141 Wodak, Ruth  5, 43–8, 49, 50, 59, 60, 69,
77, 92, 93, 96
web forums  58, 60 written language  2, 36, 38, 58, 63, 64, 65,
Wertsch, James  13–14 81, 87, 101