VISUAL-VERBAL INTERSEMIOTIC COMPLEMENTARITY

IN
THE ECONOMIST MAGAZINE

Terry Royce

A Thesis submitted in fulfilment of requirements for the degree of

PhD. in Linguistic Science

for

The University of Reading

Centre for Applied Language Studies
in the Faculty of Letters and Social Sciences

February 1999
 ABSTRACT
This research problematises the visual-verbal semantic interface in a representative
multimodal text from the Finance department of The Economist magazine. It proposes that the
social semiotic, metafunctional view of communication advanced by systemic functional
linguistics (SFL) is an effective theory for informing attempts to explicate the visual-verbal
interface in page-based multimodal texts.

This study shows that the SFL model has important and relevant applications to multimodal
forms of communication. It is particularly useful in elucidating the intersemiotic relations
between the linguistic and visual modes which commonly co-occur on page-based print
media. In addressing the question of the ways that the visual and verbal modes work together
in intersemiotic terms, this study tests the Hallidayan claim of the inter-relatedness of systems
of meaning. It also explores the proposition that both the visual and verbal modes, while
utilising the meaning-making features peculiar to their respective semiotic systems, work
together in the particular context described to realise a unified, coherent multimodal text.

The SFL model interprets communication in social terms as a system of meaning potential.
The speakers or writers make selections from the language system in response to various
contexts of situation which are also instantiations of the wider context of culture. In
accordance with the SFL view of context and language being in a dialectic relationship, the
sample multimodal text from The Economist magazine is interpreted as a particular contextual
configuration of the variables Field, Tenor and Mode which is situated in a particular context
of creation, and which possesses an intertextual history. General background information on
the context of creation is presented. This is concerned with The Economist magazine as an
institution producing economic and financial journalism, its institutional history, and its
editorial policy in terms of the ways that it aims to approach its readers through both verbal
and visual means. Further, the sample text is examined in terms of its intertextual history, or
the previously produced texts which contribute to its final form. These extra-visual and extra-
linguistic variables are interpreted as having an important bearing on the production and
subsequent interpretation of the sample text’s visual and verbal meanings.

Utilising this contextual information, it is demonstrated that in the representative Economist

magazine text the visual and verbal modes complement each other semantically to produce a
single textual phenomenon characterised by INTERSEMIOTIC COMPLEMENTARITY. A
descriptive analytical framework is developed, and it is proposed that intersemiotic
complementarity obtains when one or more of a range of criteria are met. These criteria are
identified as intersemiotic ideational, interpersonal, or compositional metafunctional
meanings. The results of this study demonstrate that the intersemiotic complementarity
between the visual and verbal modes is realised in the sample text through three
simultaneously-occurring intersemiotic resources. Firstly, the visual and verbal ideational
meanings are lexico-semantically related by means of the intersemiotic sense relations of
repetition, synonymy, antonymy, hyponymy, meronymy, and collocation. Secondly, the
visual and verbal interpersonal meanings are related through intersemiotic reinforcement of
address and intersemiotic attitudinal congruence. Thirdly, the visual-verbal compositional
meanings are integrated through the compositional relations of information value, salience,
and visual framing, as well as the influence of visual synonymy, and potential reading paths.
These results are examined in terms of their implications for further research in the
intersemiotic analysis of multimodal texts. Some educational implications are also discussed.
 LIST OF TABLES

CHAPTER ONE Page

1.1 A typology of economics discourse. 6

CHAPTER TWO

2.1 Micro-studies of economics discourse. 19

2.2 Macro-studies of economics discourse. 19
2.3 Educational studies of economics discourse. 19
CHAPTER FOUR

4.1 Representational visual structures (Ideational) adapted from 58

The Grammar of Visual Design (1996).

4.2 Interactive meanings (Interpersonal) adapted from The 71

Grammar of Visual Design (1996)..

4.3 Size of Frame and Social Distance. 75

4.4 Interrelated systems of compositional structuring principles. 92
4.5 Verticalised Ideal-Real meaning in multimodal texts. 96
4.6 Visual ‘clues’ for Compositional Salience. 99
4.7 Functions and systems in painting. 107
CHAPTER FIVE

5.1 Metafunctional views of visual communication. 125

5.2 First-order and Second-order context in multimodal text 129
5.3 Cohesive sense relation - hyponymy. 143
5.4 Cohesive sense relation - meronymy. 143
5.5 Ideational intersemiotic complementarity. 145
5.6 The principle types of Modal Adjunct. 160
5.7 Degrees [Values] in Modality. 162
5.8 Interpersonal intersemiotic complementarity. 166
5.9 Compositional intersemiotic complementarity. 182
5.10 Analytical framework for intersemiotic complementarity in 184
The Economist magazine.

CHAPTER SIX

6.1 Graphic design at The Economist - interview questions. 190

6.2 A description of the Mountains text’s context of situation. 205
6.3 Headlines setting the scene. 213
CHAPTER SEVEN

7.1 The sketch caricature Visual Message Elements (VMEs). 230

7.2 The line graph Visual Message Elements (VMEs). 234
7.3 (a) Lexico-semantic intersemiotic complementarity - sketch 238

v
 caricature

7.3 (b) Lexico-semantic intersemiotic complementarity - sketch 239

caricature

7.4 (a) Lexico-semantic intersemiotic complementarity - line 240

graphs

7.4 (b) Lexico-semantic intersemiotic complementarity - line 241

graphs

7.5 Intersemiotic sense relations in the Mountains text 243

7.6 (a) Addressing the reader/viewer - Interpersonal Intersemiotic 269
MOOD.

7.6 (b) Addressing the Reader/Viewer - Interpersonal 270

Intersemiotic MOOD.

7.7 Factual statements in the Mountains text 273

7.8 (a) Modality features of the sketch caricature 275
7.8 (b) Modality features of the line graphs 276
7.9 Attitudinal intersemiotic complementarity in the sketch 278
caricature

7.10 Discourse staging and visual narrative staging. 291

CHAPTER EIGHT

8.1 Schematic structure in the Mountains text. 313

8.2 Visual to verbal schematic structure intersemiotic 317
complementarity

8.3 Asking metafunctional questions of a visual 325

LIST OF FIGURES
CHAPTER FOUR Page

4.1 The frontal angle. 78

4.2 The oblique angle. 79
4.3 Cube views. 81
4.4 The dimensions of physical space. 98
CHAPTER FIVE

5.1 Levels of language and realisation in the SFL Model. 114

5.2 Levels of visual expression and realisation in the SFL Model. 118
5.3 A framework for visual-verbal intersemiotic complementarity 126
in page-based multimodal text.

5.4 Intersemiotic sense relation of repetition. 142

5.5 Intersemiotic sense relation of synonymy. 142

vi
 5.6 Intersemiotic sense relation of antonymy. 142
5.7 Intersemiotic sense relation of hyponymy. 143
5.8 Intersemiotic sense relation of collocation. 143
5.9 Intersemiotic sense relation of collocation. 144
5.10 Modality in the naturalistic visual continuum in The 154
Economist magazine

5.11 Mathematical visual modality in The Economist magazine 156

5.12 Information valuation on the page 178
CHAPTER SIX

6.1 The naturalistic visual continuum 195

CHAPTER SEVEN

7.1 Procedural steps in the INTERSEMIOTIC IDEATIONAL analysis 221

7.2 Reverse view of sketch caricature (left - right reading path) 226
7.3 Intersemiotic sense relations 237
7.4 Procedural steps in the INTERSEMIOTIC INTERPERSONAL 252
analysis

7.5 Procedural steps in the INTERSEMIOTIC COMPOSITIONAL 283

analysis

LIST OF VISUALS
CHAPTER FOUR Page

1 Unemployment - Little shop of horrors plant (The Economist, 61, 87

April 25th 1992, p57).

2 America’s government debt (The Economist, June 5th 1993, 66

p71)

3 America’s banking battles (The Economist, October 30th 68

1993, p79)

vii
 CONTENTS
CHAPTER ONE Introduction and Outline of the Research Page

1.0 Introduction 1
1.1 Aims of the Study 3
1.2 Data and Terminology 5
1.3 Hypotheses and Outcomes of the Research 9
1.4 Structure of the Study 12

CHAPTER TWO The Analysis of Economics Discourse by Economists and

Applied Linguists: A review

2.0 Introduction 16
2.1 The Analysis of Economics Discourse by Economists and Applied 16
Linguists

2.2 The Analysis of Economics Discourse and Visual Information 21

2.2.1 Visual Information as a Literary Device 21

2.2.2 Visual information and Applied Linguistic/Educational Studies 25
2.3 Conclusions 33

CHAPTER THREE Semiotics, Linguistics, and Visual Communication

3.0 Introduction 36

3.1 The Sign and Meaning in General Semiotic Theory 36

3.2 The Saussurean Semiotic Tradition 39

3.3 Summary 46

CHAPTER FOUR Hallidayan Interpretations of Visual Communication

4.0 Introduction 48
4.1 Social Semiotics 50
4.2 Reading Images 51
4.2.1 Representational Structures 57
4.2.1.1 Narrative Processes 59
4.2.1.2 Conceptual Processes 63
4.2.2 Interactive Meanings 70
4.2.2.1 Contact 72
4.2.2.2 Social Distance 74
4.2.2.3 Attitude 76
4.2.2.4 Modality 82
4.2.3 Compositional Meanings 90

i
 4.2.3.1 Information Value 93
4.2.3.2 Salience 100
4.2.3.3 Framing 101
4.2.3.4 The Importance of Reading Paths 103
4.3 The Language of Displayed Art 104
4.4 Summary and Conclusions 109

CHAPTER FIVE An Analytical Framework for the Analysis of Intersemiotic

Complementarity in The Economist Magazine

5.0 Introduction 113

5.1 The Systemic Functional Linguistic (SFL) Model 115

5.2 The SFL Model and Visual Semiosis - An Analytical Framework 117
5.3 The Context of Situation and Context of Culture 127

5.4 The Influence of Other Texts: Intertextual History 132

5.5 The Analysis of Intersemiotic Ideational Features 135

5.5.1 The Interpretation of Visual Ideational Features 136

5.5.2 The Interpretation of Verbal Ideational Features 138
5.5.3 The Analysis of Ideational Intersemiotic Complementarity 139
5.6 The Analysis of Intersemiotic Interpersonal Features 146

5.6.1 The Interpretation of Visual Interpersonal Features 147

5.6.1.1 The Analysis of Visual MOOD - Address (“Visual Speech Acts”) 148
5.6.1.2 The Analysis of Visual MOOD - Involvement 150
5.6.1.3 The Analysis of Visual MOOD - Power Relations 152
5.6.1.4 The Analysis of Visual MOOD - Degree of Social Distance 152
5.6.1.5 The Analysis of Modality in the Visual Component 153
5.6.2 The Interpretation of Verbal Interpersonal Features 158
5.6.2.1 The Analysis of MOOD in Verbal Propositions 158
5.6.2.2 The Analysis of Modality in Verbal Propositions 159
5.6.3 The Analysis of Interpersonal Intersemiotic Complementarity 164
5.7 The Analysis of Intersemiotic Compositional Features 165
5.7.1 The Interpretation of Visual Compositional Features 167
5.7.2 The Interpretation of Verbal Compositional Features 169
5.7.3 The Analysis of Compositional Intersemiotic Complementarity 175

5.8 The Theoretical Framework and Summary 183

CHAPTER SIX The Context of The Mountains Text

ii
 6.0 Introduction 186

6.1 The Economist Magazine and its Context of Creation 187

6.1.1 The Artistic Conventions of the Authors’ Community 187

6.1.1.1 The Institutional Background of The Economist Magazine 187
6.1.1.2 The Economist Style Book and Numbers Guide 188
6.1.1.3 The Economist Magazine’s In-house Graphic Design Policy 191
6.1.1.4 Visual and Verbal Typology at The Economist Magazine 193
6.1.2 The Authors’ Conception of their Audience 200
6.1.3 The Authors’ Individual Preoccupations 202
6.2 The Mountains Text in The Economist Magazine 203

6.2.1 The Context of Situation of the Mountains Text 206

6.2.2 The Intertextual History of the Mountains Text 206
6.2.2.1 The Subject Matter and the Issue (Field-related) 208
6.2.2.2 The Attitudes (Tenor-related) 214
6.2.2.3 Visual and Verbal Typology (Mode-related) 217
6.3 Summary 219

CHAPTER SEVEN The Analysis of Intersemiotic Complementarity in The Mountains

Text

7.0 Introduction 220

7.1 Analytical Procedures 220

7.2 An Intersemiotic Ideational Analysis of the Mountains Text 222
7.2.1 Visual Message Elements in The Sketch Caricature 223
7.2.2 Visual Message Elements in the Line Graphs 231
7.2.3 An Intersemiotic Ideational Analysis of the Mountains Text 235
7.2.3.1 The Sketch Caricature 242
7.2.3.2 The Line Graphs 248
7.3 An Intersemiotic Interpersonal Analysis of the Mountains Text 253
7.3.1 Visual Address (MOOD): Offers and Demands 254
7.3.2 Visual Point of View: Involvement 255
7.3.3 Visual Point of View: Power Relations 257
7.3.4 Visual Point of View: Social Distance 258
7.3.5 Visual Modality: Viewer Attitudes 259
7.3.5.1 The Attitudes in the Sketch Caricature 259
7.3.5.2 The Attitudes in the Line Graphs 263
7.3.6 The Mountains Text: An Interpersonal Intersemiotic 267
Complementarity Analysis
7.3.6.1 The Analysis of Intersemiotic Address in the Mountains Text 268
7.3.6.2 The Analysis of Intersemiotic Attitudinal Congruence and 272
Attitudinal Dissonance in the Mountains Text (Visual-Verbal

iii
 Modality)

7.3.6.3 The Analysis of Intersemiotic Attitudinal Congruence and 279

Attitudinal Dissonance in the Mountains Text (Attitudinal Attributes
and Adjectives).

7.4 An Intersemiotic Compositional Analysis of the Mountains Text 282

7.4.1 Intertextuality and Composition in the Mountains Text 284

7.4.2 Intersemiotic Compositional Features in the Mountains Text 285

7.5 Discussion of Results and Summary 293

CHAPTER EIGHT Conclusion

8.0 Concluding Summary 305

8.1 Directions for Future Research 309

8.1.1 Analyses Focussing on Specific Intersemiotic Metafunctions 309

8.1.2 Intersemiotic Complementarity between the Visuals and the Verbal 312
Discourse Staging

8.1.3 Intra-visual Compositional Intersemiotic Complementarity in the 318

Mountains Text

8.2 Educational Applications of this Study 321

8.2.1 A Suggested Methodology for Using Visuals in Educational 324

Contexts

8.3 Final Comments 328

BIBLIOGRAPHY 330

APPENDICES
A1 - The Mountains Text
A2 - Sentence Division of the Mountains Text
A3 - The SFL Model and Economics Discourse

iv
CHAPTER 1 - Introduction and Outline of the Research

1.0 Introduction
One of the most outstanding features of this age is that it is an age of signs. There
is a proliferation of signs. Not just signs in the usual sense of “Beware of the Dog”
or “Keep off the grass” signs, but in the semiotic sense. There are orthographic or
written, and verbal or spoken signs, the subject of modern linguistic science.
There are visual signs, as in human non-verbal behaviours (gestures),
photography, architectural design, cinema, theatre, television, sculpture, painting,
fashion design, the dance, advertising images, icons on computer screens, and a
multiplicity of non-specific images which operate to guide, to instruct, to constrain
or to allow, to protect or to warn. There are also the aural signs, as in classical
music. These linguistic, visual and aural sign types can occur in isolation, or may
be produced in combination as in the aural, spoken and visual in opera, the aural
and visual in ballet, or the aural, visual and orthographic in musical Christmas
cards. This potential for a multiplicity and variety of signs, in combination with
the exponential growth of the Internet, multimedia GUIs (Graphical-User-
Interfaces), and their role in what has been characterised as the “information
revolution”, has in recent times raised consciousness of the visual sign as a
conveyor of meaning, so much so that there is an increasing interest amongst
educationists of a possible need for developing students’ abilities in visual
literacy, rather than simply oracy or literacy.

The co-occurrence of and interrelationship between visual and linguistic signs is

the issue that is addressed in this study. In Western culture, and indeed most
developed cultures, there is no question that the linguistic, spoken and written
forms are generally viewed as the dominant, and indeed, the superior mode of
communication. The visual sign is generally seen as being subordinate to the
spoken/written sign, and even more so is the aural sign. The dominance of the
linguistic, and the dominance of the written over the spoken as the more
“advanced” mode is all-pervasive, and is inculcated across various cultures in and
through educational practices and attendant value systems (McLuhan 1964; Ong
1982). The rate and degree of change that has been brought about by the

1
increasingly computerised and multimedia-based modes of communication in this
age of signs may well form a challenge to the traditional dominance of the
linguistic over the visual age, but it is still an embryonic challenge. The fact
remains that the linguistic dominates; it dominates in the educational systems, and
it dominates in the print media, whether it be academic or journalistic, whether the
texts therein utilise only the linguistic or written mode, or whether they combine it
with some kind of visual representation.

The motivation for the present study lies in a question arising out of this writer’s
involvement, over more than 20 years, in various educational contexts relating to
the discipline of economics. This involvement covers an undergraduate major in
economics, the teaching of economics principles courses in senior high schools,
and most recently research into the pedagogical difficulties inherent in economics
discourse for non-native speakers of English (Royce 1993, 1994). The question
relates to a perception that whilst the various forms of published economics
discourse all utilise some form of visual communication, they do so differentially.
Any reader of economics discourse will appreciate that the primary channel for
the presentation and discussion of the content is the written, orthographic medium
(hereafter referred to as the verbal mode), and that this is consistently
accompanied by some form of visual communication. What is notable, however, is
that whilst the professional, academically-oriented literature as a matter of course
directs the readers to refer to the various graphs, tables, charts and diagrams
utilised (as in “Consider the relationship shown in Fig.3.1, …. ”, or more
indirectly through “This relationship can be represented diagrammatically … ”),
the journalistic print media dealing with economic issues do so only rarely.
Business magazines such as The Far East Economic Review and financial
newspapers such as The Financial Times include many of the same types of visual
forms of communication found in the professional, academically-oriented
publications, but they do not specifically direct or ask the reader to interact in
some way with the graphs, tables, diagrams and charts they utilise. Both the verbal
and visual modes seem to simply co-occur. Given the importance of visual modes
of communication in professional, academically oriented economics discourse, the
questions arise: what then is the function of the visual mode when used in

2
conjunction with the verbal in economically-oriented journalistic print media, and
what connections are construed when these two page-based modes co-occur?

1.1 Aims of the Study

While the central focus of the discipline of linguistics throughout the twentieth
century has been on the study of natural language, either in structural,
psycholinguistic or functional terms, some of the trends of the latter decades of the
century have reflected an increasing interest in forms of communication other than
linguistic and the ways that they project and organise their meanings. This interest
has been fueled not only by the continual development and enrichment of semiotic
theory, but also by the already-mentioned rapid changes in the modes of
communication brought on by the revolution in communication technology. The
result is an increase in the publication, for example, of studies dealing with
meaning projections in digital imagery (e.g. Marchese 1995), discussions on the
development of automated presentation design, intelligent multimedia interfaces
and associated architectural and modelling issues (e.g. Maybury 1993), and an
expanding debate about the ‘visual’ character of contemporary culture (e.g. Jenks
1995). It has also been recognised that developments in general linguistic theory
could inform the interpretation of other communication modes besides language.
Researchers utilising concepts from linguistic theory to examine non-linguistic
modes have included the Prague School of the 1930’s and 1940’s, the Paris School
of the 1960’s and 1970’s (see Nöth 1990 for a good overview), and more recently
what has been loosely characterised by Kress and van Leeuwen as “the fledgling
movement” of “Social Semiotics” (1996:5). This latter paradigm utilises Systemic
Functional Linguistic Theory, developed by M.A.K. Halliday (1978, 1985, 1994).
This theory interprets language as a socially-based semiotic system, and
applications of it to non-linguistic forms of communication have in recent times
been gaining prominence through research on the visual semiotics of displayed art
forms such as sculpture, architecture and painting by O’Toole (1994, 1995), and
the proposal of a ‘grammar’ of visual design in images in general as well as in
educational contexts by Kress and van Leeuwen (1990, 1996).

3
In order to address the question of the function of the visual vis a vis the verbal in
economics discourse, this study problematises the visual-verbal interface in
multimodal text (which is here defined as any text which utilises verbal and more
than one other semiotic code to project its meanings), which entails a theoretically
motivated investigation of the ways that the different semiotic modes project
meanings, and of the intersemiotic relationships between them. It will thus involve
an exploration of the proposition that both the verbal and visual modes, while
utilising the meaning-making features peculiar to their respective semiotic
systems, work together in various contexts to project a unified, coherent message
to their viewers/readers. The subject of this exploration will be a multimodal text
extracted from the economically-oriented journalistic print media, specifically
from The Economist magazine. To that end, the analysis of intersemiotic relations
in this study will take as its theoretical framework a Hallidayan view of
communication, and will utilise and seek to test the applicability of Halliday's
Systemic Functional Linguistics theory (hereafter SFL) to multimodal meaning.

According to the Hallidayan SFL theory, language is viewed in a "social semiotic

perspective", where “social” refers firstly to the social system (which is
synonymous with culture) and secondly to the fact that language is to be
interpreted in terms of its relationship to social structure (Halliday 1978, 1985:3-
4). For Halliday, semiotics is not simply “the general study of signs” (as per
Saussure et. al.), but is “the study of sign systems …… the study of meaning in its
most general sense” (Halliday and Hasan 1985:3-4). The discipline of linguistics
can be seen as one “kind of semiotics” which studies the ways that language
organises meaning. However, there are “other ways of meaning, other than
through language ….. there are many other modes of meaning, in any culture,
which are outside the realm of language” (op.cit:4). These other modes of
meaning may comprise
both art forms such as painting, sculpture, music, the dance, and so
forth, and other modes of cultural behaviour that are not classified
under the heading of forms of art, such as modes of exchange,
modes of dress, structures of the family, and so forth. These are all
bearers of meaning in the culture. Indeed we can define a culture as
a set of semiotic systems, as a set of systems of meaning, all of
which interrelate. (loc.cit. - my emphasis).

4
The assumption that semiotic systems interrelate seems to be an established one.
Aspects of this have been explored in some depth for example by semioticians like
Schapiro (1973:9-11), who studied medieval images derived from religious
narratives, and Barthes (1977:38-41), who in his famous essay “The Rhetoric of
the Image” examined the various ways that verbal text and image interact in press
photographs and advertisements, asserting inter alia that the relation between
them is one of either anchorage (image-text dependency) or relay (image-text co-
operation). The question therefore arises: if it is assumed that different semiotic
systems can and do work together semantically, what evidence is there for it, and
how can it be accounted for? Or put in another way, what is the function of the
visual vis a vis the verbal mode, and vice versa?

This study will therefore seek to test the claim of the inter-relatedness of systems
of meaning, in this case, the interface between the visual and the verbal semiotic
systems in a multimodal text extracted from one instance of economically-oriented
journalistic print media, The Economist magazine.

1.2 Data and Terminology

The data which constitutes the focus of this study is a multimodal text drawn from
The Economist magazine, specifically from a section which is commonly referred
to as the Finance department. This text has been extracted from a March 1993
issue, bearing the title heading “Mountains still to climb”, and is reproduced in
Appendix One (The Economist, March 27th, 1993:77-78). Two questions need to
be addressed in relation to the institutional discourse focus of this study and the
multimodal text chosen for analysis. The first is why The Economist magazine has
been selected from the range of possible types of economic discourse, and the
second is why this particular text.

For the purposes of this study, the text extracted from The Economist will be
viewed as a sample of written, page-based economics discourse, one of a range of
potential discourse forms within the disciplinary culture of economics (Becher
1981:109, 1987:261). The various types within this disciplinary culture range
from academically-oriented professional economics journal articles through to

5
journalistically-oriented newspaper reports on economics topics. A typology and
some well-known examples of these discourse forms is given in Table 1.1. The
Economist magazine has been chosen as the focus of this study because it uses
almost all the visual forms that can potentially be used across the range of
identified types of written economics discourse shown. The other forms of
economic discourse such as professional journals, textbooks and finance industry
reports tend to use specific types of visual forms and not others. The Economist
magazine does not use the specialised mathematical symbolism of algebra and
statistics commonly used by the professional literature, but it does use almost all
the other visual forms drawn from mathematic semiotic systems such as graphs,
charts, and tables. The Economist magazine also uses sketch drawings (mostly
caricatures) and photographs (mostly portraits and scenes) common to media
discourse types such as magazines and newspapers. It would seem therefore that
an examination of a multimodal text drawn from The Economist magazine, one
which could potentially include the full range of possible visual types used in
economics discourse, would be a more useful text to focus on in terms of the
applicability of its results to other disciplines in the social and general sciences
which often utilise the same range and type of visuals.

Discourse Type Examples Most common visual forms

Professional economics Economics, Journal of International Line graphs, tables, bar graphs, pie charts,
journals Economics, Journal of Economic schematic diagrams.
Literature, Review of Economics and
Statistics.
Economics textbooks Introductory economics texts, university Photographs (author portrait); line graphs,
texts on theoretical aspects of macro and tables, bar graphs, pie charts, schematic
microeconomics. diagrams, sketches.
Reports by economic OECD reports, Barclays Bank economic Photographs (portrait), line graphs, tables,
organisations reports, stock market reports. bar graphs, pie charts, diagrams; flow charts.
Reports and analyses of The Economist Magazine, The Financial Photographs (portrait and scene), line
economic issues by Times, The Wall Street Journal, The Far graphs, tables, bar graphs, pie charts,
various journalistically- East Economic Review. diagrams, flow charts; sketch caricatures.
oriented print media.

Table 1.1 - A typology of economics discourse

For the purposes of this study, visual information will be interpreted as those
features which in a traditional linguistic sense are generally not analysed as part of
"text", such as the various drawings, diagrams, graphs, tables, and charts that are
6
most commonly used in the discourse types outlined in Table 1.1. Mathematically-
based algebraic equations (e.g. the Keynesian four-sector model of the economy
expressed in the algebraic equation Y=C+I+G+[X-M]) and statistical formulae
(e.g. the analysis of covariance) will not be interpreted as visual information here,
firstly because as already mentioned above, this type of symbolic communication
is rarely, if ever used in The Economist magazine, and secondly because algebraic
and statistical formulae differ fundamentally from the ways that the traditional
forms of visual representation (pictures, graphs, diagrams etc.) are considered to
project their meanings — they are often viewed as another semiotic code (see
Nöth 1990:217-218 on numerical codes), or as within the linguistic system, as
Lemke (1990) points out in his discussion of science as a subject in educational
contexts and its use of mathematics:
mathematics itself, that is the use of mathematical expressions, is part of
language, not something different from or alternative to language. You
can, quite literally, talk mathematics, either by reading the symbols, or by
converting them into conventional words and phrases of the language
(register) of mathematical English. Most mathematical expression is a
form of written language, and mathematics makes use of its own
specialised written genres, such as the Proof (op.cit:159).

It is important at this point also to clarify just what is meant by text in the term
multimodal text. Without providing an exposition of the various definitions of text
typically and traditionally applied by theoretical and applied linguistic researchers
(see for example Stubbs 1983, Brown and Yule 1983, Hoey 1991, van Dijk 1977,
Halliday & Hasan 1976 etc.), text in this work will accord with Halliday and
Hasan's view that it is "a semantic unit: not of form, but of meaning" (1976:1-2),
and that it is functional, or “language that is doing some job in some context”
(1985:10). It will also accord with their view of cohesion, which is also viewed in
terms of meaning, being the "relations of meanings that exist within the text, and
that define it as a text" (1976:4). Although Halliday and Hasan are referring
primarily to language in its spoken or written forms, they are cognisant of the fact
that their view of text as contextualised meaning and function permits the
consideration of other modes of meaning-making. Thus, a text “may be either
spoken or written, or indeed any other medium of expression that we may like to
think of.” (1985:10 my emphasis). In this study, the multimodal text extracted

7
from The Economist magazine (hereafter referred to as the Mountains text), will
therefore be considered as a contextualised, cohesive unit of meaning, with the
understanding that the ways the various meaning(s) are projected can be
potentially realised via differing coding or semiotic systems. Accordingly, in this
study the use of the term text will be made with the understanding that this means
multimodal text, not simply text in the traditional linguistic (spoken or written)
sense.

The Mountains text has been selected for analysis because it is typical of the
Finance department leader articles which can be found in any issue of The
Economist magazine between early 1992 and late 1994 (in 1995 the department
changed its heading to Economics and Finance and started including coloured
graphics). Over this period a corpus of over one hundred texts from the Finance
department was compiled, and an analysis was conducted to identify and
characterise the texts according to the most common text-type. The results of this
analysis revealed:
• None of the leader articles occurred as verbal text only. They all included some
kind of graphic or image, the term ‘image’ incorporating both photographs and
sketched drawings, and the term “graphic” incorporating mathematically-
derived tables, charts and graphs, of which there are various types (line graphs,
pie charts, bar graphs/charts, tables etc.). Thus all the texts were multimodal.

• All the texts contained some form of an image, the most common being what
may be characterised as sketch caricatures (approximately 97%), the other
being head and shoulder (portrait) photographs of a person, people, or people in
a scene (approximately 3%).

• Many of these verbal plus image texts were combined with some form of a
graphic, the most common being a line graph or a bar graph/chart.

• A small number included more than one of each type of visual mode - for
example some included a sketch caricature, a photograph, and a line graph,
while others contained a sketch caricature and two photographs only.

From this analysis it can be seen that the most typical multimodal text type which
could reasonably be viewed as representative of the types that occurred between
the period early 1992 and late 1994 would be one that is characterised by the co-
occurrence of verbal language, a sketch caricature, and either a line graph or a bar
graph/chart (Preferably a line graph, since most of the bar graphs in the corpus
8
express the same rate of change relationship as a line graph — the remainder of
the bar charts simply presenting comparative proportions, a point treated in greater
depth in Chapter Six).

In terms of the terminology used, the standard SFL conventions as used by

Halliday (1994) and Matthiessen (1995) are adopted in this study. Briefly, these
are as follows:
• the metafunctions written in lower case: ideational, interpersonal, and textual.
• the names of systems written in all capitals: MOOD, TRANSITIVITY,
SPEECH FUNCTION, THEME etc.
• the names of structural functions in the clause are written with initial capitals:
Actor, Process, Goal, Mood, Residue, Finite, Predicator, Theme, Rheme etc.

1.3 Hypotheses and Outcomes of the Research

As already indicated, this study problematises the visual-verbal semantic interface
in multimodal text by questioning the functioning of the visual vis a vis the verbal
modes in economically-oriented journalistic print media. In doing so it aims to test
the Hallidayan claim of the inter-relatedness of systems of meaning, in this case
the visual and verbal semiotic systems in a multimodal text extracted from The
Economist magazine. It will therefore involve an exploration of the proposition
that both the visual and verbal modes, while utilising the meaning-making features
peculiar to the respective semiotic systems, work together in this particular
context to realise a unified, coherent text for potential viewers/readers. That is, it
is proposed that the visual and verbal modes semantically complement each other
to produce a single textual phenomenon in a relationship which can be referred to
as intersemiotic complementarity, and that this relationship will obtain when one
or more of the following occurs:
• when the ideational meanings in both modes are related lexico-semantically
through intersemiotic sense relations of repetition, synonymy, antonymy,
hyponymy, meronymy, and collocation.

• when the interpersonal meanings in both modes are related through

intersemiotic reinforcement of address, and through intersemiotic attitudinal
congruence and attitudinal dissonance (modality) relations.

• when the compositional meanings are integrated by the compositional relations

of information value, salience, visual framing, visual synonymy, and potential
reading paths.

9
The antithesis of this proposition is that while the verbal and visual modes utilise
the meaning-making features peculiar to their respective semiotic systems, they
simply co-occur and do not work in concert to project a unified, coherent text. In
other words there is no intersemiotic complementarity between the visual and
verbal modes, and the intersemiotic ideational, interpersonal and compositional
meanings do not occur.

The analysis which follows in this study will examine the above proposition, and
will present a framework developed to explore it. It is assumed that the realisation
of intersemiotic complementarity in multimodal text does not imply that the visual
semiotic and verbal semiotic simply co-occur on the page space and have a
relationship of conjunction. Rather, the implication is that the relationship is
synergistic in nature, a concept which describes the ability of elements, in the act
of combining, to produce a total effect that is greater than the sum of the
individual elements or contributions (Random House 1992). The sense of the
usage of the term complementarity in this study therefore differs substantively
from Greimas’ usage, that of a logical relation between semantic oppositions as
represented in his “semiotic square” (see Nöth 1990:318-319). It also differs to its
usage in research on the nature of the relationship between speech and writing
which suggests that it is potentially one of heteronomy, or of autonomy, or of
partial to full interdependence or complementarity — in the latter area of research
complementarity means that writing and speech can influence each other in
linguistic communication, and that this interdependence can be revealed in
evolutionary, psycholinguistic and cultural terms (op.cit:262-263).

Further, the question of whether one mode is more dominant than the other, or
whether a particular verbal text can be understood with or without the visual
mode, will not be addressed in this study. Aspects of these questions are explored
in some depth by Barthes (1977), who looks at the various ways that text and
image interact in advertising. His work is discussed more fully in Chapter Three.
A discussion of the processes involved in the production of Economist magazine
texts and the general reasons why various graphic design decisions are made is

10
outlined in Chapter Six, and this will help to clarify the views of the magazine
itself on the question regarding which mode is dominant. The concern of this
study however is with the verity of multimodal text in The Economist magazine,
with the analysis of a text produced by an organisation which has chosen for
contextually-driven reasons to include both visual and verbal modes in its articles,
and the semantic evidence within the text to support the assertion that they are
interrelated.

Intersemiotic complementarity as used here implies further that while a

multimodal text with one of the modes removed would still produce a coherent
visual or verbal text, it would be one which would somehow be diminished in its
communicative power. It is the aim of this study therefore to take a first step
towards the explanation of how this intersemiotic synergism is realised, and to test
the claim that intersemiotic complementarity occurs in ideational, interpersonal
and compositional terms.

There are three outcomes expected of this study. It is expected that:

1. The visual and the verbal modes in The Economist magazine work together in
intersemiotic semantic complementarity as defined above to produce a coherent
multimodal text, and that the nature of the intersemiotic semantic resources
which can be deployed to realise that complementarity will be clarified.

2. The results will provide evidentiary support for Halliday and Hasan’s
(1985:4,10) assertion of the interrelatedness of systems of meaning, and will
extend the Systemic Functional Linguistic concept of metafunctions to
multimodality in text.

3. The results will show that some of the concepts and terminology of existing
linguistic theory and their attendant analytical tools can be productive in
explaining how semiotic systems interact. These include the concept of sense
relations in lexical cohesion as outlined by Halliday (1994:330-334) and
Halliday and Hasan (1985:80-82), and the analytical tool of cohesive chains
which is commonly used to illustrate the ways that lexical items in text are
semantically related to each other (84).

This research is therefore a qualitative and descriptive study which aims to extend
theory by means of its application.

11
1.4 Structure of the Study
In order to establish the need for this study, Chapters Two, Three, and Four will
review the literature dealing with its general focus, which is visual communication
in economics discourse. In Chapter Two the review of the literature on the
analysis of economics discourse by applied linguists and economists will firstly
highlight the nature, focus and findings of the various studies, and secondly
evaluate the treatment of visual modes of communication where it is mentioned or
acknowledged in some way. It will be shown from this review that the published
research by applied linguists and economists has not attempted or been able in any
rigorous way to describe, explain or account for the role of the visual mode in
economics discourse.

Chapter Three will overview and review relevant work by semioticians and
linguists who have attempted to analyse various visual modes of communication.
This will initially involve a brief overview of the conception of the sign and
meaning in general semiotic theory, and then a survey of the two main schools of
semiotic theory which have attempted to examine other modes of communication
utilising linguistic insights originally derived from Saussure’s treatise. This
chapter will show that throughout the twentieth century there has been consistent
interest in explaining the characteristics of various visual modes, but that most of
this work however has concentrated on single-mode analysis, with only some
attempts to examine intersemiotic relations. The two main schools of semiotic
analysis examined in this chapter are the Prague School and the Paris School, with
a concentration on the work of the most influential of the Paris School’s theorists,
Roland Barthes. Barthes’ attempts to build upon Saussure’s semiological program
contributed much to text semiotic studies of myth, theology, literature and
narrative, as well as to ideas about the relationship between image and text in
advertising. This chapter will thus act as a bridging chapter to Chapter Four by
providing the background to the development of semiotic theory as derived from
Saussure.

Chapter Four will outline and review the recent work of those who have used
insights derived from the general theory of language and communication

12
developed by M.A.K. Halliday (1978, 1985) to examine non-linguistic modes.
This work has recently been referred to as the Hallidayan school of “social
semiotics” (Kress and van Leeuwen 1996:5), whose adherents have been active in
applying Halliday’s principles to various modes of visual communication. M.A.K.
Halliday’s Systemic Functional Linguistics (SFL), along with selected aspects of
the work recently carried out by others using its principles, will form the
theoretical framework for the analysis of the text extracted from The Economist
magazine. This review will show that a significant amount of work has been
carried out by those working from a Hallidayan interpretation of meaning-making
in visual semiotic modes, and that it is concerned mainly with two areas: various
types of images such as photographs, diagrams and drawings, and displayed art
such as painting, sculpture and architecture. Common to this work is the
reinterpretation of Halliday's metafunctions of language in order to examine the
main processes used by visual modes such as photographs, drawings, diagrams,
maps and charts, as in Kress and van Leeuwen's (1990, 1996) 'grammar of
images', and those used by paintings, sculpture and architecture, as in O'Toole's
(1994, 1995) 'language of displayed art'. This review will also demonstrate that
although there is an expanding body of work examining these different types of
visual modes, within this particular school there is a dearth of research into the
nature of the intersemiotic semantic relationship between the visual and verbal
modes in multimodal text, and none at all in relation to economics discourse. At
this point the need for this study will therefore be clearly established.

The next chapter, Chapter Five, will present the theoretical foundations for the
analysis of the sample Economist magazine text. This chapter discusses the
fundamental assumptions of the SFL model as a social semiotic theory of
communication, covering theoretical principles such as context of culture, context
of situation, and metafunctions, and then applies these principles to the ways that
the two modes, the linguistic (verbal) and the visual realise meanings. A
discussion of SFL theory applied to the visual mode and the core ways that it
projects its meanings is given (this is complemented by a discussion in Appendix
Three of the verbal mode in SFL theory utilising examples drawn from economics
discourse). Based on the discussion in Chapter Four, Chapter Five will present the

13
parameters of an analytical framework which will be used to examine
intersemiotic complementarity in The Economist magazine text i.e. the possible
ways in which the SFL model can be extended to apply not just to language or to
visual modes, but to the semantic intersemiotic relationships between them.

Chapters Six and Seven form the heart of this study. Chapter Six, in line with the
context-driven nature of the SFL model, presents background information on The
Economist magazine, as an institution of economic and financial journalism, its
institutional history, and its editorial policy in terms of the ways that it aims to
approach its readers through both verbal and visual means. Much of this
information is derived from The Economist magazine Style Book (1990) and
Numbers Guide (1991), and the main points arising from an interview which was
conducted in 1994 with Ms Penny Garrett, the Chief Editor of Graphic Design at
the Head Office of The Economist magazine in London. This chapter also
provides a visual typology based on an analysis of the range and varieties of visual
communication commonly utilised in issues of The Economist between early 1992
and late 1994, an analysis which provides important background information on
the ways that the most frequently occurring visual types, sketch caricatures and
line graphs, project their meanings. The features of the context of situation of the
sample Economist magazine text will also be examined, which will involve a
discussion of its intertextual history and context of creation. This is concerned
with the ways that the preceding texts deal with the subject matter and issue being
addressed, the attitudes that are expressed towards this subject matter/issue, and
the ways that the magazine composes them in visual and verbal terms on the page.

Chapter Seven will apply the analytical framework presented and outlined in
Chapter Five in order to examine and analyse the intersemiotic complementarity
between the verbal and visual modes in the sample Economist magazine text. The
application of this analytical framework to this multimodal text will utilise the
range of informational insights gathered from the contextual analysis of Chapter
Six — these include the relevant features discussed concerning The Economist
magazine as a commercial publishing institution, its stylistic and graphic design
conventions, and the various contextual variables which make up the text’s

14
intertextual history. This chapter examines the representative text extracted from
the Finance Department of The Economist magazine. The text analysed includes
both a sketch caricature and a mathematical visual (two line graphs), and is
analysed to ascertain the ways that intersemiotic complementarity is realised in the
intersemiotic ideational, interpersonal, and compositional features between these
modes. Specifically, the analysis will attempt to test the applicability of the
concept of intersemiotic complementarity for demonstrating how the visual and
verbal modes work together on the page in The Economist magazine. It is
proposed that intersemiotic complementarity obtains when the ideational
meanings in both modes are related lexico-semantically through intersemiotic
sense relations of repetition, synonymy, antonymy, hyponymy, meronymy, and
collocation, when the interpersonal meanings in both modes are related through
intersemiotic reinforcement of address, and through intersemiotic attitudinal
congruence and attitudinal dissonance relations, and when the compositional
meanings are integrated by the compositional relations of information value,
salience, visual framing, visual synonymy, and potential reading paths.

Chapter Eight discusses the results of the analyses in Chapters Six and Seven in
terms of the central hypotheses, and draws some conclusions based on the work. It
examines the results of the study in terms of its implications for further research in
the intersemiotic analysis of multimodal texts, as well as applications and teaching
activities relevant to both first and second language pedagogy.

15
CHAPTER 2 - The Analysis of Economics Discourse by
Economists and Applied Linguists: A review

2.0 Introduction
The overall aim of this chapter is to review the literature on the analysis of
economics discourse by economists and applied linguists from two perspectives.
The first is to ascertain the nature, the focus and the findings of the various
studies, and the second is to review them in terms of their treatment, or perceived
lack of treatment of visual modes of communication. The specific aim of this
chapter is to show that despite an increasing awareness of the nature of economics
discourse by both applied linguists and economists, the published research has not
attempted or been able to in any significant way to describe, explain or account for
the visual mode in economics discourse. The research by both economists and
applied linguists is varied, but it tends to cover the full range of the available
varieties of economics writing that are published within the field in both
academically and commercially-oriented areas.

In recent years there has been a steady increase in interest and research into
economics discourse by both economists and applied linguists which has spawned
an expanding body of work. The nature of this work in part reflects not only the
varied academic backgrounds of the writers, but also the evolutionary
development of applied linguistics in general and its sub-discipline of discourse
analysis in particular. This body of work is not only succeeding in clarifying many
of the ways that economists use language to express themselves, but has also
proven to be very useful for educators attempting to find solutions to the
difficulties that students often have with the specialised language of economics at
both the lexicogrammatical and discourse levels.

2.1 Analysis of Economics Discourse by Economists and Applied Linguists

Since the early 1980s, the discussion of various controversial issues in the
economics discourse community has led to increasing debate among concerned
economists about the ways that they communicate with each other, as well as with
non-economists. This debate has been vigorous, and has also influenced the
direction and nature of the research into economics discourse by applied linguists.

16
Before reviewing how visual communication is handled in the research on
economics discourse, a brief exploration of these discussions by economists about
their own discourse as well as an overview of the main areas of applied linguistic
research would be instructive.

Economists’ assessments of their own discourse has contributed to a growing

awareness by many that the ways they communicate their ideas in economics do
not accurately correspond to the ways they actually “do” economics. The major
figure amongst those economists who advocate that fellow economists should
examine the ways they use their own discourse is Donald McCloskey, an
economic historian and economist, whose range of publications dealing with the
"rhetoric of economics" (1983, 1984, 1986, 1990), has attracted academic interest
from not only his economics colleagues, but also from applied linguists.
McCloskey's thesis is to argue for a new methodology of economics. He asserts
that economists have two attitudes to their discourse, termed the "official and
unofficial, the explicit and implicit" (1986:5), and that the official, explicit attitude
(and therefore rhetoric) reflects a scientific methodology which is "modernist", a
modernism which consists of "an amalgam of logical positivism, behaviourism,
operationalism, and the hypothetico-deductive model of science" (1983:484).
McCloskey argues that economists in practice don't follow the rules as laid down
by this official methodology (which he suggests may be a good thing), but in
reality argue using the unofficial, implicit rhetoric of economics. He therefore
believes that the rhetoric of economics should be examined by those economists
who use it, suggesting that the quality of their argument would be at a more
sophisticated level if they were more aware of the grounds on which they were
arguing, because they:
claim to be arguing on grounds of certain limited matters of statistical
inference, on grounds of positive economics..... and other positivistic
enthusiasms of the 1930's and 1940's. They believe that these are the only
grounds for science. But in their actual scientific work they argue about the
aptness of economic metaphors, the relevance of historical precedents, the
persuasiveness of introspections, the power of authority, the charm of
symmetry, the claims of morality (op.cit:482).

17
The responses to McCloskey's work have generally been quite positive (e.g.
Button 1988; Pearce 1991:25), prompting even McCloskey himself to remark on
their warmth (1984:579). Criticism of his ideas, however, has covered a range of
areas, and it has varied in its intensity. Among economists who express strong
reservations about, or directly oppose the usefulness or underlying assumptions of
McCloskey's arguments, are Caldwell and Coats (1984:577). They agree with
McCloskey's claim that positivism is dead, but disagree that it is the same
situation with economic methodology. Others include Maki (1993), who refutes
McCloskey's treatment of the nature of "Truth" in relation to a rhetoric of
economics by distinguishing between two philosophies within that rhetoric, and
Gerrard (1993), who suggests that hermeneutics can provide insights into the
analysis of economics discourse.

Other economists besides McCloskey have also been actively examining aspects
of economics discourse. Henderson (1986) for example, offers a seemingly
parallel, but unconnected (there is no reference to McCloskey's work) examination
of the various ways that metaphor in economics can be investigated, stating that
they are very common both in economics as a science, and in discussions
involving economic policy. Like McCloskey, he examines metaphor as a series of
tropes (as in metaphor, simile, and analogy), and states that what he terms as
"living" and "dead" metaphors are an integral part of the economics lexicon, and
are in fact inter-woven into the concept-structure of introductory economics
textbooks. Although McCloskey is an economist who examines the language of
economics, his work here is not specifically addressed to his economist
colleagues, but more towards those working in the education and applied
linguistics discourse communities. As Tables 2.1, 2.2, and 2.3 on the analysis of
economics discourse by applied linguists will show, Henderson’s collaborations
with various colleagues in that field (especially with Dudley-Evans 1990, 1991,
1993; and with Hewings 1987a, 1987b, 1990) have helped fuel the interest shown
by applied linguists in the study of economic discourse.

Additional studies of interest which examine or respond to the implications of the

18
Lexical Studies Bramki and Williams (1984); importance of lexical familiarisation strategies to facilitate
Henderson and Hewings (1987b). reading comprehension, and students' difficulties.
Syntactic Studies Spencer (1975/6) specialised vocabulary: noun/verb combinations in
textbooks from biology, chemistry, economics and law.
Johns (1980) "constellations" of cohesive devices in business discourse.
Mead and Henderson (1983) the relationship between conditional form and meaning in
economics writing.
Mason (1990, 1991) the linguistic mechanisms used by economists in developing
abstract concepts
Hewings (1989); Hewings (1990) modelling in economics textbooks, and aspects of lexical
realisation
Henderson and Hewings (1990) the use of key words used to build economic models
Tadros (1979, 1981, 1985) linguistic (as opposed to economic) prediction in economics
text
Cross-Cultural Mauranen (1993) metatext in the economics writing of native and non-native
Aspects speakers

Table 2.1 Micro-studies of economics discourse

Genre Analysis Pindi and Bloor (1987); hedging and schematic structure within the economic survey
Pindi (1988) and forecasting text genres
Dudley-Evans and Henderson genre analysis of introductions to economics articles on
(1990a) taxation in The Economic Journal
Politeness Strategies Bloor and Bloor (1991) the ways that economists modify their propositions through
hedging in economics writing
Myers (1991) the ways economics researchers rewrite grant proposals to
two research councils
Dudley-Evans (1991) politeness strategies in economics writing

Table 2.2 Macro-studies of economics discourse

Teaching Techniques Mead and Lilley (1975) the preparation and presentation of English teaching
materials for students of economics at the University of
Libya
Tadros (1977), principles for selecting reading passages and developing
relevant teaching/learning activities, preparing teaching
materials for paragraph and essay writing
Jordan (1990) the problems that undergraduate students have with
economics essays and theses
Syllabus Design Hewings (1989) the major reading problem areas for non-native speakers of
English studying economics
Hewings and Henderson (1987) the teaching of economics to adults in a part-time social
science degree course at the University of Birmingham
Fisher (1990) the relevance of recent research on economics discourse to
ESP teachers and economics, economics/ESP textbooks
Houghton and King (1990) an analysis of classroom discourse aimed at studying the
difficulties experienced by students of development
economics in the University of Birmingham
Allen and Pholsward (1988) the reading difficulties experienced by students of
economics in English in Thailand
Course Descriptions De Escorcia (1984) report on subject-specialist, team-teaching economics
program for first-year students in a university in Chile
Jordan (1978, 1984) special classes organised for overseas post-graduate students
in the University of Manchester
Schleppegrell (1985) an ESP program for Egyptian university students
Royce (1984, 1993, 1994) a program for senior secondary school students in Australia
(1984); a reading/discussion course for economics graduates
and businessmen in Japan (1993); an economics readings
course for liberal arts students in a bilingual (Japanese and
English) university (1994).
Mparutsa (1991) an analysis of the oral reports delivered by Zimbabwean
economics students
Mparutsa, Love, and Morrison the use of concordancing programs to assist students in
(1991) dealing with unfamiliar discourse in three academic
subjects, one of which is economics

Table 2.3 Educational studies of economics discourse

19
debate about the rhetoric of economics include those in the history of economic
thought, by researchers who have been examining the writings of eminent
economists such as Adam Smith (Bazerman 1993; Brown 1993), Francis
Edgeworth and Alfred Marshall (Henderson 1993), and John Maynard Keynes
(Rotheim 1988; Anuatti 1991; Favretti 1991). Others have been analysing the
more recent debates between the Chicago School's Milton Friedman and his
Keynesian critics such as James Tobin (Backhouse 1993). Related work by
Mirowski (1991) traces mathematical discourse in economics and its historical
influence, suggesting that mathematical expression in economics discourse was
neither inexorable nor unhindered, but revealed a rather disjointed narrative due to
the fact that "in the context of the development of the evolution of economic
thought, the participants were far from convinced that the subject matter
intrinsically demanded mathematical expression" (op.cit:146). This debate is
supported and enhanced by discussions relating to theory versus empiricism in
academic economics (Morgan 1988), laboratory experimentation as opposed to
theory-building for the study of economics (Smith 1989), and the rhetorical role of
statistical testing and econometrics in economic proofs (Darnell 1991).

Although McCloskey's and his detractors' work is interesting for the insights it
provides regarding economists' current views of their own discourse, the research
dealing with the analysis of economics text by applied linguists represents an
expanding body of work which in recent times has begun to intersect with the
discussions held by economists, and reflects a growing awareness that
opportunities have been created for establishing common ground between
language specialists and economists. Recent reviews and overviews of this
research include Dudley-Evans and Henderson (1990), Henderson, Dudley-Evans
and Backhouse (1993), and Royce (1995). The work by Royce divides the
research into the three categories of Micro-studies, where the main focus is on the
analysis of lexical or syntactic aspects of economics discourse, Macro-studies,
where the main focus is on economics discourse beyond the sentence including
text analyses, and Educational Studies, in which the main focus is on the analysis
of economics discourse for pedagogical purposes (op.cit:138). The most
significant studies and their focus are summarised in Tables 2.1, 2.2, and 2.3.

20
2.2 The Analysis of Economics Discourse and Visual Information
The previous section examining the literature on the analysis of economics
discourse by economists and applied linguists was a brief overview aimed at
identifying the important research contributions, clarifying interrelationships
between them, and demonstrating the main areas of analysis. This section attempts
to focus specifically on selected studies in terms of how they deal with visual
information in economics discourse, especially those which have made some
reference to it in their analysis. It seeks firstly to clarify just what is said about
visual information, and then to critically assess this treatment in terms of the aims
of this study. As already outlined in Section 1.2, visual information in economics
discourse refers to the various drawings, diagrams, graphs, tables, and charts that
are used across the spectrum of economics discourse types (see Table 1.1).
However, some of the studies which do consider visual information in some way
often incorporate it with a discussion of the importance of mathematical or
algebraic modes, which are generally considered as non or extra-linguistic.
Accordingly, this review will not only comment on references to drawings,
diagrams, graphs, tables, and charts in the literature, but also on various
interpretations of mathematically-based algebraic equations and statistical
formulae and their importance in economics discourse.

It will be seen from the following appraisal that although visual communication is
often acknowledged incidentally as being functionally related to the verbal text
(commonly in terms of co-text by the studies concerned with language education),
rarely is this relationship explored or analysed in any rigorous or extensive way.

2.2.1 Visual Information as a Literary Device

Two authors from the previous survey in section 2.1 on the nature of the
discussions by economists about their own discourse draw upon principles derived
from literary criticism to consider visual information in economics discourse,
notably McCloskey (1986) and Henderson (1986). Both discuss visuals in terms
of their use as metaphor.

21
The discussion by McCloskey (1986) about rhetoric in economics provides some
relatively useful insights into the role and importance of visual information in
written economics discourse. His treatment of visual information occurs within the
context of his discussion about how economists use literary devices in their
"conversations" with each other, thus forming part of the discipline's rhetorical
method. Linking the development of this current rhetorical form to the growth of
modernism in economic methodology, he argues that although economics
“conversation” has been lucid in the past, the most eloquent “conversationalists”
have been the economists who have used mathematical (meaning the use of
descriptive techniques derived from algebra, geometry, calculus etc.) techniques.
They have drawn on model-building procedures and econometric techniques that
have been derived from mathematics and/or statistics, and these include not only
linear, mathematically-based methods, but also visual forms of representation such
as graphs, tables and flow charts.

To support this claim McCloskey compares articles from the American Economic
Review drawn from the period 1931-33 with those drawn from the period 1981-3.
He asserts that the economists of the earlier period were naive mathematically and
unskilled in conversing in "curvey metaphors", relying mostly on tabular statistics.
Since the period 1931-33 he claims, there has been a clear growth in the influence
and use of mathematical metaphors. As he states of his examination of the papers
from the 1980's, "only six [authors] used words alone and only four added to their
words tabular statistics alone, the one formal device common in 1931-33. The
techniques of mathematics, statistics, diagrams, and explicit simulation, which
economists had viewed once as useless and arcane, had become routine."
(op.cit:4)

McCloskey reinforces this view through the examples he gives in his analysis of a
famous text, Robert Solow's essay on the production function and productivity
change (op.cit:83-86). This is an economic model which is often expressed in both
algebraic and graphic forms. Although the main emphasis in McCloskey's book is
on the role of metaphor, he points to the fact that the master tropes of metaphor,
metonymy, synecdoche and irony are common in Solow's analysis. The

22
production function as either an algebraic equation or a function curve (a line
graph) is the metaphor for decision-making in production. The use of the letters
“K” and “L” in his algebraic equation as the symbols representing capital and
labour are treated as an instance of metonymy, and the use of the multiplicative
factor A{t} (the part) which is to be identified with technical change (the whole) is
an example of synecdoche. The example of irony he gives is the only literary
device which is not associated with an instance of visual communication.

McCloskey is thus working with a form of rhetorical analysis (he doesn't use this
term, but uses simply "rhetoric") that has been derived from a recent American
tradition of rhetoric studies, philosophy and literary criticism. McCloskey is not
an applied linguist, so in some areas he is not explicit about the use and derivation
of some of the analytical devices he uses (see Dudley-Evans & Henderson
1990:5), and he also relies heavily on his specialist subject knowledge
(McCloskey 1986:5-6). Furthermore, McCloskey's treatment of visual information
in terms of literary devices basically precludes him from examining in depth any
possible intersemiotic functional relationship between the visual and verbal
modes. This is of course a more linguistic task that he has not set out to do.
However, the insights he has provided about the role of visual information in
economics are a very instructive confirmation of its importance for the discipline,
and provides an awareness of the need to develop the analytical tools to conduct
further research in that area.

Henderson's (1986) treatment of visual information also occurs within the context
of his discussion of metaphor in economics. Like McCloskey, he draws on aspects
of the application of literary criticism, and provides some background on the use
of metaphor in the debate over economics methodology and the degree to which
economics is a predictive science. He examines metaphor, simile and analogy as a
series of tropes under the label of metaphor, (although perhaps the classifier
"figurative language" would have been more appropriate to avoid possible
confusion). It is through his discussion of metaphor, the various examples he
gives, and the comments he makes about them that his treatment of visual
information is clarified. Metaphor, he claims, contains an implicit "as if" notion,

23
and economic models (which of course are often expressed in a variety of forms of
visual information) are also "as if" statements about the world. Henderson gives
examples of this in a macroeconomics passage where the economy is seen in a
variety of metaphorical ways, one of which is as a set of graphs, the others being
as something physical with gaps that need plugging, a ship at sea, a fire, a
machine, or a person (op.cit:114).

Henderson's suggestion that metaphors belong to one of two basic metaphoric

traditions in economics, the mechanistic and the organic, also points to the
metaphorical role and importance of visual information. The mechanistic
metaphoric tradition, according to the examples that Henderson uses, seems to be
the one that has facilitated the growth of the use of visual information techniques
in economics. This can be seen in his discussion of the predictive model of price,
where the
metaphorical (mechanistic) foundation of price theory is accepted ...... not
because we think the economy is a machine but because treating it as if it
were a machine has led to the development of a consistent and predictive
theory of price by application first of the diagrammatic and later by the
mathematical method already implicit in diagrams. (op.cit:115)

Here the link between metaphor and visual information is clearly established.

Henderson also discusses the difficulties that readers often have in reading and
understanding diagrams. He states that people unfamiliar with economics writing
will often find the diagrams of supply and demand (line graphs) difficult to
understand. They may look reasonably decipherable, but the difficulty for the
uninitiated lies in the fact that "....elementary supply and demand diagrams are in
fact iconic metaphors i.e. they are like a map of a town that does not exist....or
they are like maps of all possible towns of a certain type” (loc.cit.). The degree to
which the language is literal can also cause some difficulty if the economics
student is required to extend his or her understanding beyond the given example.
Thus,
Price in the elementary diagram is arrived at by the balance of the
influences of both demand and supply. To understand and use the
diagram for the analysis of real world markets, the notion of price
itself must be seen as a metaphor and applied to all sorts of

24
 phenomena not normally seen as a price e.g. income and wages as a
price for labour. ..…. There is a deceptive simplicity of diagrams
combined sometimes with a spurious concreteness that makes
understanding difficult and memorisation self-defeating with
respect to the move from markets in general to the study of a
particular market. (loc.cit.)

Again, this treatment of metaphor in economics discourse (and the visual

information which is a fundamental part of it), is a useful application of various
understandings from the techniques of literary criticism. However, it does not
provide any real analytical insights for the linguist attempting to explain the
semantic, intersemiotic relationships between the visual and verbal modes. Both
Henderson and McCloskey are not linguists but economists, so this criticism
should be modulated by that fact; further, their aims were clearly not to address
the members of the linguistics community but their fellow economists, in the hope
of improving both economics methodology and communication between
economists as a result.

2.2.2 Visual information and Applied Linguistic/Educational Studies

This section will examine the treatment of visual communication from three
perspectives. The first will review those studies which have primarily an applied
linguistic focus, studies which aim to elucidate aspects of economics discourse to
advance researchers’ knowledge of how the discourse is structured and organised.
The second will review those studies which focus on an analysis of economics
discourse to assist the development of effective teaching programs or syllabuses,
while the third examines studies which focus on the development of teaching
materials and pedagogy.

Applied Linguistics. Hewings (1990) examines the problems that students have
with the ways that introductory economics textbooks switch between the real
world and degrees of abstraction in their explanation of fundamental theories and
notions, but concentrates primarily on the ways that this is carried out
syntactically in the verbal aspect of the text (op.cit:31). Despite the relevance of
visual information in this switching process and the problems students have with
it, Hewings doesn't directly mention the visual mode - she deals mainly with

25
Sinclair’s (1986) general notions of fictional/non-fictional worlds, and more
specific sentential aspects of Winter's (1977) work on clause relations. In the latter
case, where she deals with lexical realisation (which she glosses as “specifying the
unspecifics”), Hewings recommends that
Authors of pedagogic texts, if they are efficiently and effectively to
inform their readers, must employ devices which indicate something
about the phrase, clause, sentence or groups of sentences that
follows and how it relates back to what has been said, or forward to
what is coming (1990:34).

This is an important point in that it applies equally to the visual mode. Hewings
doesn't include visual information as a device or tool to “efficiently and
effectively … inform their readers”, which is surprising considering the fact that
the authors of introductory pedagogic texts frequently and increasingly make use
of different forms of visual communication. They do this explicitly through
linguistic devices directing the readers, and implicitly through various semantic
devices which require readers to refer forward and backwards between the modes.
This is illustrated in the following extracts from the economics textbook Hewings
analyses, one dealing with elasticity of supply, the other with the effect of changes
in price of other goods on demand.
Now, suppose instead we have a horizontal supply curve. Here a
tiny cut in P will cause Q to become zero and the slightest rise in P
will coax out an indefinitely large supply …. (op.cit:36).
And,
A rise in the price of a commodity complementary to X will shift
the demand curve for X to the left, indicating that … (op.cit:38).

Both of the above quotes implicitly refer to and assume that the reader can at least
see a line graph somewhere in the textbook. The kinds of graphic visuals (supply
curve/demand curve) and their component parts (P/Q, and X) which implicitly
relate to these linguistic devices (termed nominal forms by Hewings) often follow
or precede them on the page, and are in fact an integral part of any explication of
an important theoretical model such as “The Market Mechanism”, which is the
theoretical focus of the above extracts.

These criticisms of Hewings (whose work on economics discourse is both seminal

and extensive) should be mediated by the fact that they can also be applied to

26
much of the research covering the analysis of lexical, syntactic and generic aspects
of economic discourse. In most mainstream and traditional applied linguistic
analysis as well, visual modes of communication are sometimes acknowledged,
but rarely analysed as relevant in relation to their connection to the verbal mode.
They are often just simply omitted from the original texts, as in the syntactically-
focused study by Spencer (1975/76). The sampling technique in this study
involved randomly selecting five pages from some introductory economics,
biology, law and chemistry textbooks - all were pre-university or first-year
university level, “but no page containing a drawing or diagram was included”
(op.cit:31). The primary purpose here was to analyse semantic combinations,
reflecting the applied linguist’s traditional concern with verbal, orthographic text
(loc.cit.).

Henderson and Hewings (1990) continue in this vein, where they use examples
which apparently did include visual information in the original extracts, which
was intentionally edited out and acknowledged accordingly. The examples they
quote in their analysis have, as an integral part of their message, the use of,
referral to, and assumed awareness on the part of the reader of a diagram
illustrating demand and supply curves, and the effect of changes on them
(op.cit:44). Henderson and Hewings are not aiming to examine visual information,
but they are aware of its connection with the verbal mode, and they briefly
integrate that awareness into their discussion, as in
Textbook writers frequently illustrate movements of curves and associated
concepts in terms of demand changes and give less treatment to supply
suggesting that the reader transfer ideas across from the way in which
demand has been handled. At the same time supply, and what is to be
found behind the supply curve, is normally given more detailed treatment
later in the textbook. Since demand and supply diagrams are concerned
with the analysis of change, it comes as no surprise that ‘change’ is the
next most frequently used word (op.cit:48).

Henderson and Hewings also briefly refer to the importance of illustration in an

earlier paper (1987b), where they focus on lexical familiarisation techniques used
by economics writers. They find, that in the order of frequency of use, illustration
as a lexical familiarisation technique in their sample made up 2% of the items they
tabulated. As they point out, illustration "is an additional non-verbal aid to

27
comprehension. Graphs, tables, diagrams etc. are usually supported by references
from within the main body of the text" (op.cit:125).

Other researchers who have touched on, but not focused specifically on visual
information in economics discourse are Mauranen (1993), Tadros (1985), Mead
and Henderson (1983), Mason (1990, 1991), Allen and Pholsward (1988), and
Cameron (1991). Mauranen's (1993) study on the writing of Finnish economists
composing in English focuses on metatext in its role as a text organiser. She
identifies four types of metatextual organisers termed as connectors, reviews,
previews and action markers, which, although they relate primarily to verbal text,
are also relevant to visual information, because some of the actual examples she
quotes are used to refer back or forwards to visual information, or suggest an
action which can relate to visual information. For instance, Mauranen gives an
example of previews such as "We show below....", which can be used to direct the
reader to a visual, as can the examples of action markers, such as "to illustrate the
size of this distortion..." (op.cit:10).

Tadros' (1985) research focuses on linguistic prediction in economics text by

examining a corpus derived from an introductory economics textbook. She
identifies six major categories of prediction. One of these, advance labelling,
occurs where the writer refers to acts of discourse in advance and is therefore
committed to carrying them out. This category is a notable explanation of the
predictive relationship between the verbal and the visual text (referred to as
‘linear’ and ‘non-linear’ text by Tadros). Tadros identifies three types of advance
labelling, the first of which is realised by linear text (although visual information
may co-occur but isn't predicted). The second type labels an act which must
obligatorily be realised by non-linear text, while the third act predicts two acts that
may be linear or non-linear (op.cit:24-28). The characteristics of each of these
three types will not be explained in depth here, but it is important to note that this
analysis, within the context of an examination of linguistic prediction in
economics discourse, suggests an analytical structure or framework which may be
useful to adapt for use when considering other features/types of economics
writing. That statement must however be qualified by the fact that Tadros' study is

28
aimed at an analysis of linguistic prediction only, so it may not be suitable to
account for the other instances where non-linear text co-occurs but is not overtly
predicted (such as a caricature in The Economist magazine), as is the case in the
first type briefly mentioned above, and which Tadros herself identifies (op.cit:24).
It is this type which is most relevant to the focus of this study.

Mead and Henderson's (1983) analysis of the use of conditionals in economics

writing to realise economic predictions complements Tadros' work. The difference
here is that they are working with economic prediction as opposed to linguistic
prediction (op.cit:141). The authors show through their analysis that economic
predictions are expressed by various conditional and non-conditional forms, and
that these forms generate a range of functions or meanings. They examine the
conditional forms found in introductory economics textbooks, and distinguish the
various meanings that are carried by these forms. Of these forms (the predictive,
the illustrative, the defining, and the directive conditional), the directive is the one
which relates most closely to visual information. This functional form instructs the
reader on how to interact with the economics content, requiring him or her to look
elsewhere in the text, or to do something with the content being considered
(op.cit:153-155). This may potentially take the form of directing the reader's
attention to a graphic, or it may ask him or her to interpret a graphic. Mead and
Henderson's closing comments about economic text as a result of their analysis
include some conclusions about book structure and why some students experience
difficulties with visual information. They suggest that the textbook they used as a
source for their study has a non-linear structure (in terms of sequencing of
information). They also suggest that problems may be caused for the students by
the other ways that the textbook is non-linear, in situations where "reference is
made to graphs and mathematical equations which may themselves be derived
from earlier graphs and equations" (op.cit:140).

Mason's (1990,1991) work on abstract versus concrete language in economics is

an attempt in part to deal with visual information linguistically and to comment on
its role in economics text. She does this by referring to its importance as a source
for writers to use where they can draw upon other semiotic systems, especially

29
mathematical ones to present information and "do" economics. She asserts that
economics, in a similar way to other scientific disciplines, makes use of
mathematical systems for modelling reality. The features of these systems can be
referred to, and discussed in relation to other features in terms of their usefulness
and importance, but they can really be only fully understood mathematically
(1990:27). She also points out that even though economists may use mathematical
language to varying degrees, they have to at some point express their conclusions
in natural language, referring to mathematical systems as much as their natural
language will allow (1991; see also Lemke 1990:159). This is a clear
acknowledgement that there is an important relationship between verbal and
visual text, as well as a suggestion of the primacy of verbal over visual modes in
economics discourse.

Teaching Programs and Syllabus Design. The most notable feature about the
treatment of visual information in studies dealing with economics discourse and
teaching syllabuses is that they include visual information in the organised
classroom activities in a variety of ways. This suggests an acknowledgement on
the part of educators and course designers that visual information is an important,
integral part of economics discourse, and is an area of concern in relation to
specific student problems. Thus, instructors and course planners have identified a
need to focus on these areas if they are going to effectively help students to
develop communicative skills in economics.

Examples of papers and reports which incorporate work on visual information in

their EAP teaching programs are numerous. Hewings and Henderson (1987) for
example, use various methods in training their students to read bank articles, such
as asking them to refer to graphs and tables, and using flow diagrams that give an
overview of the information structure of the reading passages (op.cit:167 & 171-
2). Schleppegrell (1985) emphasises the use of authentic materials in her course,
providing examples of classroom materials that require the students to work with
tabular information as stimuli for both reading and listening passages. De Escorcia
(1984) also uses visual information as an aid to reading comprehension, requiring
students to transfer information from reading passages to schematic form. There is

30
no other reference to visual information in her paper, however. Mparutsa et.al.
(1991b), in their study of the usefulness of concordancing in the classroom,
mention visual information only in their introduction, and then only in a
subsequent reference to allowing the students to treat text examples in a non-linear
fashion.

The same can be said for Houghton and King's (1990) report on the types of
questions asked by students and lecturers in development economics in the
University of Birmingham's economics education and ESP/EAP courses. They
suggest through their examples that students are required to read, interpret and
respond to numerical and tabular information in their tasks, and thus are required
to develop not only linguistic competency, but competency with the mathematical
semiotic system as well (op.cit:98). Fisher (1990) also refers to this requirement
when she examines the relevance of recent research on economics discourse for
ESP teachers and uses her findings to evaluate current economics and
ESP/economics textbooks. She does this with reference to the problem areas of
target students, vocabulary, discourse, culture and numeracy skills. With regard to
numeracy skills, she concludes that students often have difficulty in seeing the
integration between text and diagram (op.cit:91). Citing Henderson & Hewings'
(1987a) finding that a major problem for students is that they react to visual
information in their rhetorical roles of exemplifying and elucidating (as they have
often been taught), when in fact they may need to view visual information in the
rhetorical acts of presenting a problem or demonstrating an ideal state, Fisher
comments that students often have difficulty in figuring out which of these
rhetorical acts is being used (1990:91). Further, students sometimes have
difficulties with simple numeracy, and are often mostly concerned with that, rather
than with identifying whether the instance is real or hypothetical (loc.cit.).

Allen and Pholsward (1988), in their paper on the development of a diagnostic

reading test for EAP economics students, provide a linguistic analysis which
briefly considers the importance of visual information. In an attempt to confirm
and extend the work of previous researchers, they overview the results of previous
work on the main problem areas for students reading in economics; these seem to

31
be grammatical structures (conditionals, passives, relative clauses, logical
connectives), rhetorical functions (assumption, hypothesis, prediction,
explanation, generalisation), and the use of visual data (statistical tables, maps and
graphs). Their own analysis of economics text in terms of lexis, syntax and
discourse confirms and extends the results of their overview. They mention visual
information within their section on discourse, but only briefly, since they
concentrate heavily on confirming the frequency of rhetorical acts. What they do
is include or pair the incidence and role of visual information with rhetorical
functions in the organisation and development of the discourse structure. They
identify certain regular patterns of discourse development as in statement of
concept/theory/argument, explanation with examples, illustration with graphics,
and then a summary or restatement of the concept/theory or argument (op.cit:61-
62). The mention of visual information within this context is quite brief, mainly
because of the concentration on rhetorical acts. This is significant, because the
authors intuitively seem to include diagrams as part of discourse structure, and
definitely include them in the test items they develop (op.cit:65). Again, then,
there is an acknowledgement of the importance of visual information in
economics text, but little in the way of explanation as to how it relates to the
verbal.

Teaching Materials and pedagogy. In reports detailing the structure and scope of
teaching materials and pedagogical techniques to assist students with the discourse
of economics, brief mention is made of visual information in the context of the
discussions of specialised vocabulary and rhetorical functions as potential problem
areas.
Two articles by Jordan (1978,1984) are of note here. In the earlier of two reports,
Jordan analyses the main language difficulties that students seem to have in
coping with economics. He describes the program developed at the University of
Manchester, focussing on the general difficulties they have with the subject, and
on the main linguistic problems. In his listing of these language problems, he cites
the students' need to understand figures as being a problem area. By this he means
the use of mathematical symbols, segmental diagrams and tabular information
(1978:181). His later article reports on the development of a teaching program for

32
overseas graduate students of economics, the methods and materials arising from
an attempt to meet their needs, and the resulting higher student motivation. It
gives a description of the program with examples of the materials and methods,
which include work on visual information. This is in the speaking part of the
curriculum, where various activities involve small groups and pairs working on
role-plays, problem-solving, and an activity which Jordan refers to as “describe
and draw”, which allows the students to work with tables, charts, and graphs
(1984:84). The work on visual information is also integrated with the other
activities, as in the small group discussions, where a role play involves a pairwork
activity which provides the students with two sets of information, one of them
having "statistical information (e.g., tables of economic indicators for population,
per capita income, structure of imports" (op.cit:86). The students are required to
treat visual information as an important part of economic text, and therefore need
to develop the linguistic skills to be able to read and talk about it. Similar
pedagogical techniques and points of view were expressed by Royce in various
programs set up to meet the needs of university-level students attempting to cope
with the demands of reading introductory textbooks and journalistic writing in
economics (1984, 1993, 1994).

2.3 Conclusions
The review in this chapter has given a brief overview of the various significant
contributions to the analysis of economics discourse by applied linguists over the
last decade, by grouping them into three broad categories, Micro-studies, Macro-
studies and Educational Studies. It has also attempted to show that this interest in
economics discourse has not been confined to applied linguists, but that members
of the economics discourse community have also been concerned with the nature
of their own means of communication. It has been an area of lively debate, as the
reactions of economists and applied linguists to Donald McCloskey's The Rhetoric
of Economics (1986) illustrate.

The central focus of this chapter, however, has been an effort to show that
although there is clear evidence of a recognition of the importance of visual forms
of communication in economics discourse in the literature by applied linguists,

33
there is also clear evidence of a general lack of a rigorous treatment, a lack which
betrays a need. Some of the studies reviewed here have treated visual information
linguistically in a limited way, such as Tadros' (1985) analysis of advance
labelling in linguistic prediction in economics textbooks. Many others have
touched on visual information only briefly, well within the confines of the purpose
of their studies, whether they be an exploration of some syntactic aspect of
economics discourse, a discussion of student problems, or a report on a syllabus.
The overall picture, however, is one of a need to examine and account for visual
information in a more rigorous way, to ascertain just how the various visual modes
utilised make and project their meanings, and more importantly, to describe,
explain and account for the intersemiotic connections between them and their
verbal co-text.

This raises the question as to why researchers in applied linguistics have not to
any significant extent applied their analytical tools to this intersemiotic
relationship, given its importance to the social sciences in disciplines such as
economics or geography, and many other disciplines which often draw upon what
has been characterised as scientific/positivist methodological techniques. When
one considers, however, the cultural and intellectual primacy of the verbal over
the visual mode since the development of writing systems in human history, the
special place accorded to linguistics as the master-pattern (or patron général) in
semiology (semiotics) by Ferdinand de Saussure (1916/66:68), and the directions
that applied linguistics has taken with text and discourse analysis over the last
twenty to thirty years, this lack of exploration of a possible visual-verbal
intersemiotic relationship is understandable. However, given the already-
mentioned changes in information technology and the increasing interdependency
of the different communication modes in the information revolution, there is an
increasing need for applied linguists to broaden their focus and to develop the
analytical tools and frameworks which can examine and account for the ways that
the visual modes not only project their meanings, but also relate intersemiotically
to the verbal mode. This broader consideration of economics discourse and text,
along with any insights into this intersemiotic relationship could then perhaps flow
through to what seems to be an important area of research for many applied

34
linguists, that of syllabus development and pedagogical techniques appropriate to
students' needs, whether they be first or second language learners.

35
CHAPTER 3 - Semiotics, Linguistics, and Visual Communication

3.0 Introduction
Having established the notable lack of analysis of visual communication and its
potential interrelationship with the verbal mode in economics discourse by
economists and applied linguists, this chapter will now examine and review
relevant work by semioticians and linguists working with other forms of
communication to analyse the visual mode. This will initially involve a brief
overview of the conception of the sign and meaning in general semiotic theory,
and then a survey of the main schools of semiotic theory which have attempted to
examine other forms of communication utilising linguistic insights originally
derived from Saussure’s treatise. This chapter will act as a bridging chapter to
Chapter Four, which will examine and review the work of those who have
attempted to investigate various kinds of visual modes from within the SFL
theoretical framework which, as already stated in Chapter One, will form the
theoretical foundation for the analysis of the Mountains text. This review will
demonstrate that within this particular theoretical framework there is also limited
research into the nature of the intersemiotic semantic relationship between the
visual and verbal modes in text, and none at all in relation to economics discourse.
The culmination of the reviews in Chapters Two, Three and Four will therefore be
an established need for an application of the Hallidayan social semiotic theoretical
framework to multimodal texts.

3.1 The Sign and Meaning in General Semiotic Theory

Semiotics has been variously defined and subdivided into pure, descriptive and
applied areas, and there have been any number of definitions (see Nöth 1990),
depending on the particular tradition and the particular nuances given to the
meaning of the notion of semiotics. Across the sweep of these various views of
semiotics as a theory, a science or even as a fashionable movement, the prevalent
feature is that it is concerned with the study of the sign. Some of the more typical
definitions reflecting this variation include semiotics as “the antique doctrine of
signs” (Sebeok 1994:5), “the general science of signs and meanings” (Danesi and
Santeramo 1992:vii), “the science or doctrine that studies signs” (Danesi
1993:280), “the study of signs and symbols as elements of communicative

36
behaviour” (Random House Electronic Dictionary 1992), “the study of sign
systems” (Halliday and Hasan 1985:4), and most commonly Saussure’s definition,
from which most others are derived, as “a science that studies the life of signs
within society” (1916/1966:16).

In general, semiotics examines signs as the vehicles which allow human beings to
represent their world, to present information, to influence, or as Eco suggests,
even to deceive and obfuscate (1976:7). A broad spectrum of signs and systems of
signs in communities have been studied, whether the subject of that study is
verbal, visual or some composite blend of both. These studies have included the
semiotic analysis of speech, drama, novels, mime, comedy, paintings, architecture,
sculpture, myth, folktales, comics, news stories and advertisements, and more
recently, the previously-mentioned study of multimedia modalities. Semiotics has
also been applied to the study of signalling behaviour in and across animal species
in zoosemiotics, a branch of semiotics which attempts to account for the
“corresponding designative processes among the speechless creatures” (Sebeok
1972, 1994:19).

In the history of semiotics from the mid to late nineteenth and throughout the
twentieth century, a number of people are prominent in terms of their role in the
development of the modern form of the discipline, and its major branches.
According to Nöth (1990), these figures and their major contributions include
Peirce (philosophy, classification of signs), Morris (the scope of semiotic theory),
Saussure (semiology and linguistics), Hjelmslev (theory of glossematics) and
Jakobson (poetics and linguistics). Of these, it is generally recognised that both
Charles Sanders Peirce [1839-1914], who is the major figure in the philosophical
branch, and Ferdinand de Saussure [1857-1913], who is commonly referred to as
the founder of semiology (semiotics) and the “father” of modern linguistics, are
the pre-eminent figures in the development of semiotic theory in this century, and
that their differences, inter alia, are often characterised largely via their
interpretation of the sign. The fundamental concept of the sign in its broadest
sense as “a natural or conventional semiotic entity consisting of a sign vehicle
connected with meaning” (op.cit:79) is clarified further by Larsen:

37
 a sign is any object which represents another object. Meaning is the
representation of an object in or by another object. The sign or the
representing object can have any material manifestation as long as
it can fulfil the representational function: a word, a novel, a
gesture, a reaction in the brain, a city etc. On the status of the
represented object nothing is made explicit by this definition. It
may be material or mental, fictitious or factual, fantasised or real,
natural or artificial. From this it follows that something which is a
sign in one context may be an object in another and vice versa.
Signs do not constitute a class of objects. A sign is a ‘functional’
unit (1994:3824).

The principal functional status of a sign means that its existence is related to its
integration in a concrete process of meaning-making, a coded process involving
the production and comprehension of signs which is commonly defined as
semiosis (Danesi 1994:280; Danesi and Santeramo 1992; Nöth 1990:42; Eco
1976). This process means that “one infers something from a phenomenon one
thus considers a sign, concerning something else, the object”, and that “through
this inference, the relation between sign and object is specified according to a
code” (Larsen 1994:3824).

The views of Saussure and Peirce as to the nature of a sign and the process of
semiosis have been discussed and developed in this century to the point that most
of the semiotic literature seems to identify itself with having originated from what
has been characterised as either semiology or semiotics. Sebeok (1989:63) alludes
to these two terms being associated with the “two traditions” of semiotics, which
Nöth explains as the “linguistics tradition from Saussure to Hjelmslev and Barthes
[which was] usually defined as semiology” and the “the general theory of signs in
the tradition of Peirce and Morris [which] was called semiotics” (1990:13). In
more recent times (from 1969) the International Association of Semiotic Studies
has decided to drop this terminological distinction and to adopt the term semiotics
to cover all the research published from both the traditions of semiology and
general semiotics (op.cit:14).

Notwithstanding this terminological rapprochement, the semiotic literature can

still be interpreted in terms of its association with these two major traditions, and
it is within the Saussurean linguistic tradition which this study will be situated. It

38
has been recognised by many in this branch that semiotic research and analysis
deriving from linguistic theory could inform the interpretation of other
communication modes besides language. The rest of this chapter section will form
a brief overview of the work of those following the Saussurean linguistic tradition,
and a brief review of the work of those researchers utilising various concepts
derived from linguistic theory to examine non-linguistic modes. These can be
conveniently referred to as ‘schools’, and indeed have been described as so by
Nöth (1990). They include the Prague School of the 1930’s and 1940’s, the Paris
School of the 1960’s and 1970’s, and more recently that which could be loosely
described as the Hallidayan School, which as already mentioned in Section 1:1,
will provide the theoretical framework for this study.

3.2 The Saussurean Semiotic Tradition

It is generally recognised that the study of language has served as the model for
much work in the field of semiotics, and that the nature of this work derives
initially from Saussure’s “semiological program”, Hjelmslev’s extension of
Saussure’s ideas in his glossematics, and the further application of these ideas by
various schools of semiotics in their examination of other non-linguistic modes of
communication.

In Saussure’s semiological program, the central semiotic notion is the sign, which
is defined as a two-fold entity consisting of the signifier and signified
(1916/66:67). The signifier is defined as the material vehicle, or the “physical part
of the sign, the actual substance of which it is composed (sound waves, alphabet
characters etc.)” (Danesi 1993:24). The signified however is defined as the
meaning or mental concept to which the signifier refers. This Saussurean
dichotomy is to be contrasted with the Peircean triadic notion of the sign, which is
comprised of the representamen or sign, the object or that to which the
representamen refers, and the interpretant or the individual’s comprehension of,
and reaction to, the representamen/object association. The representamen is
synonymous with Saussure’s signifier, identifying the material or present part of
the sign, while the object and interpretant are Saussure’s signified in two parts.
The Peircean sign model is thus a “triple connection of sign, thing signified,

39
cognition produced in the mind” (C. S. Peirce, quoted in Nöth 1990:42). Both
Saussure and Peirce developed their conceptions of the sign at the same time but
independently, and both have been adopted and utilised in subsequent semiotic
studies by various semioticians. Indeed, and as already mentioned, much of the
semiotic literature can be interpreted as being derived from either a Saussurean or
Peircean perspective. On this point Leeds-Hurwitz (1993:23) argues that the two
conceptions of the sign should not be viewed as contradictory, but that the
Peircean triadic model should be viewed as an elaboration of the Saussurean
dyadic conception, which is therefore “the more basic and essential”. This is also
the view adopted for this study.

Saussure (1916:65-70) elaborated his model of the sign only to the extent required
to explore the nature of the linguistic sign, but as already mentioned, many
following the Saussurean semiological tradition transferred his ideas to non-
linguistic signs. Two schools are of note in that regard — the Prague School of the
1930’s and 1940’s, and the Paris School of the 1960’s and 1970’s.

The Prague School was characterised by an elaboration of the work of the Russian
formalists, who attempted to develop a scientific approach to literature and art
through their notions of automatisation and deautomatisation (or foregrounding),
and the application of selected aspects of Edmund Husserl’s [1859-1938]
phenomenology by such luminaries as Jan Mukarovsky [1891-1975], Karl Bühler
[1879-1963], and Roman Jakobson [1896-1982] (Larsen 1994:3829). The notion
of ‘foregrounding’ was applied to language (i.e. phonological or syntactic forms
were deviated from the expected, standard forms for artistic purposes), as well as
to a range of art forms. Mukarovsky (the arts in general), Jakobson (aesthetics,
music, cinema), and Bogatyrev (folklore, theatre, clothing and fashion) were the
most significant figures in that area (Nöth 1990; Larsen 1994).
The Paris School is distinguished by its direct relationship to and development of
Saussure’s semiological conception of the sign, and the influence of Hjelmslevian
linguistic theory on its central figures. This school is often characterised as being
derived from linguistic structuralism (often referred to as the French Structuralist
School), and its members are noted for their work in text analysis (see the Greimas

40
school of structural semantics), as well as for their semiotic analyses of a range of
non-linguistic modes of communication. These analyses draw upon Hjelmslev’s
theory of connotation and glossematics, in which he insisted on locating language
within a broader semiotic framework — where language is defined as comprising
both “linguistic” and “non-linguistic languages” (Larsen 1994:1586). Some of the
more prominent semioticians developing these ideas include Metz (semiotics of
film), Nattiez (semiotics of music), Fresnault-Deruelle (graphic representation in
comics), Lindekins (images in photography), and finally Barthes whose work on
the photographic image, fashion, architecture, painting, film, and advertising has
made him the most prominent and cited member of this school (Nöth 1990).
Barthes’ contribution bears some consideration, since his analysis of image-text
relations is directly relevant to the focus of this study.

Roland Barthes [1915-1980], as a follower of Saussure’s semiological program

and a leading structuralist, is recognised for his contribution to text semiotic
studies of myth, theology, literature and narrative. He became famous also for his
contribution to the semiotics of various types of visual communication, and for his
systematic model of signification, a Saussurean term which Barthes defined as “a
process: it is the act which binds the signifier and signified, an act whose product
is the sign” (1964/67:48). Barthes set up two orders of signification, which he
referred to as denotation and connotation. His particular usage of these terms
derives directly from Saussure’s dyadic (signifier-signified) model of the sign, and
from a “greatly simplified version of the glossematic sign model” developed by
Hjelmslev (Nöth 1990:310). They refer to the first and second levels or systems of
meaning in a sign. Defined, denotation is the literal (or informational) meaning of
a sign, in terms of what is “objectively” observable or which is easily recognised
and identified. For example, an image of an eagle denotes a bird, or type of bird.
Connotation, on the other hand involves meanings which are dependent on the
denotative level. It is the symbolic level. A sign is interpreted by a viewer or
reader (and an audience in the case of music) connotatively when the
interpretation goes beyond the literal (denotative) interpretation via the activation
of context-dependent conventions or codes. An image of an eagle (a symbol),
depending on the context, can connote a football team (Manly Sea Eagles rugby

41
club), danger or a need to warn or protect (as in advising tourists/campers in a
national park), or spiritual superiority (as in many ancient religions and cultures).

Barthes explains the operation of these concepts by stating that “the first system
becomes the plane of expression or signifier of the second system …… the
signifiers of connotation …… are made up of signs of the denoted system”
(1967:91). Viewers of signs can make sense of or integrate these two levels only
by making reference to their background knowledge of the culturally-based codes
and various associative meanings; the system of connotation is dependent on this
background knowledge for its existence. Thus, in Saussurean terms the eagle
example above is a signified which can become the signifier (the carrier or
vehicle) of a further signified at another level.

The importance of context, and therefore ideology here cannot be understated, for
when readers or viewers engage with verbal or visual signs, their interpretation is
the result of the background cultural codes they introduce to unify the signifier
and signified. Barthes clarifies these concepts in his treatment of the semiotics of
publicity images, where he discusses the levels of meanings in advertising images.
The denotative level of meaning in advertising images is referred to as a non-
coded iconic message, while the connotative level is termed a coded iconic, or
symbolic message, which is grounded on underlying pragmatic, cultural, patriotic,
historic or aesthetic background knowledge. For Barthes, the advertising
photograph (and image) exists in an ideological world, a view which Dyer
(1982:129-130) elaborates:
Ads as a means of representation and meaning construct ideology
within themselves through the intervention of external codes which
are located within society. The ad will use images, notions,
concepts, myths, etc. already available in the culture. An ad does
not simply reflect ideology; it reworks it, thus producing new
meanings. It uses objects which are signifieds of ideological
systems and thought that already exist and then makes them
signifiers of another structure (the ad). Its connotational process
depends on our knowledge of the forms of ideology that
advertisements employ.

42
There is a third level of meaning, besides the denotative and connotative, that
Barthes (1977) discusses in his study of press photographs and advertisements.
This relates to the way that ideology operates through the linguistic message that
may or may not accompany the image, and he raises questions about the nature of
the functions of the linguistic message with regard to the denotative and
connotative aspects of the iconic message. Barthes states that there are two:
anchorage and relay, a dichotomy developed by Barthes in an attempt to address
the question of the contextual relationship between images and verbal text
(op.cit:38).

In explaining anchorage, Barthes envisages a need for the meaning of images to

be always related to, or in some way dependent on verbal text. Images without
verbal support are seen as being too open to a variety of meanings, having an
inherent indefiniteness which necessitates a reliance on verbal language, inter
alia, to provide a more substantive interpretation. The rationale is that
all images are polysemous; they imply, underlining their signifiers,
a 'floating chain' of signifieds, the reader able to choose some and
ignore others. …… Hence in every society various techniques are
developed intended to fix the floating chain of signifieds in such a
way as to counter the terror of uncertain signs; the linguistic
message is one of these techniques (op.cit:38-39).

The function of the linguistic message then, whether it be in the form of a caption,
heading, headline or accompanying reportage or prose, is to fix or ‘anchor’ the
various possible meanings, directing the reader’s interpretations and settling
possible visual ambiguities and contradictions. The verbal text does this at the
denotative level by answering “the question: what is it? The text helps to identify
purely and simply the elements of the scene and the scene itself; it is a matter of a
denoted description of the image.” (op.cit:39). The verbal text also does this at the
connotative, symbolic level where ideology is operating in a significant way. Here
the [verbal] text directs the reader through the signifieds of the
image, causing him to avoid some and receive others …… it
remote-controls him towards a meaning chosen in advance. In all
these cases of anchorage, language clearly has a function of
elucidation, but this elucidation is selective, a metalanguage
applied not to the totality of the iconic message but only to certain
of its signs. The text is indeed the creator’s (and hence society’s)

43
 right of inspection over the image; anchorage is a control, bearing a
responsibility …… for the use of the message. (op.cit:40)

In terms of an interpretation of the total message of the verbal/visual text,

anchorage therefore involves a situation of visual-verbal dependency, whereby
the verbal text links the image and the situation in space and time, a relation which
cannot be established through purely visual means of expression. Further, the
verbal text selects for the viewer/reader one of several possible interpretations of
the image, thereby fixing “the floating chain of signifieds” (op.cit:39).

Barthes’ view of relay however is that “text and image stand in a complementary
relationship; the words, in the same way as the images, are fragments of a more
general syntagm and the unity of the message is realised at a higher level”
(op.cit:41). Thus relay, in terms of an interpretation of the total message of the
verbal/visual text, involves a complementary relationship between both the verbal
and visual (image) constituents — it denotes a reciprocal association, such that
each mode contributes to the overall message projected. Barthes claims that this is
rare in fixed images, but most common in modes such as in the speech balloons
used in comics, and those involving moving images as in TV or film where the
dialogue works in concert with the images.

Vestergaard and Schrøder, in a discussion of Barthes’ work, assert that the two
text-picture relations should not be viewed as mutually exclusive, for there are
obviously multimodal texts containing both images and verbal text where either
the primary relationship is one of anchorage, with some element of relay, or one of
relay with some element of anchorage (1985:33-36). Nöth (1990:454) develops
this point in a brief, general overview of research into picture-text relationships in
which the anchorage-relay dichotomy does not obtain, or in which anchorage
and/or relay could be considered as sub-types of other higher-order relations.
These relationships are listed below with some examples, and it can be suggested
that these may apply equally well to other forms of visuals such as mathematical
visuals like graphs, tables and charts, as well as the photographic image or picture:

44
1. ILLUSTRATION: Here the visual is subordinated in terms of its semiotic
function to the verbal text which it illustrates. This is exemplified most
commonly in the illustrations used in fiction - some editions of literary works
are produced with illustrations, while others of the same work are not, as in for
example a high school edition of Hemingway’s “The Old Man and the Sea”,
compared to an edition sold to the public.

2. PICTORIAL EXEMPLIFICATION: This is a sub-type of relay, and unlike

illustrations, the visuals in pictorial exemplification “retain a higher degree of
informational autonomy in relation to the verbal message they exemplify
visually” (454). A common example would include the photographs provided
in encyclopaedias to exemplify the type of phenomena being described or
discussed.

3. LABELING: This involves the affixing of a label to a picture, and may be an

instance of anchorage (as in the title/name of a portrait or work of art), or as a
type of relay (as in the name given for a person in a portrait or photograph).

4. MUTUAL DETERMINATION: This is a mixture of relay and anchorage; the

interpretation of the picture is dependent upon the verbal text, but the picture
is equally as necessary for the interpretation of the verbal aspect. This is very
often seen in advertisements, where the two modes are combined via a
verbally posed question (“Can YOUR graphics board do all this for $369?”),
and a set of four pictures connoting what the board can do in terms of speed,
true colour, video, and real-time 3D (from PC Magazine, May 16 1995:265).

5. CONTRADICTION: This is rare, but involves the presentation of opposite or

contrasting messages by both the language and the picture. This has perhaps
become more common in recent times with radical graphic design techniques
becoming more in vogue, as evidenced by magazines which are constructed in
ways which purposely breach compositional conventions for the effects it can
create. For example the pop magazine RAYGUN refers to itself as:
“THE BIBLE OF MUSIC + STYLEANDTHE END OF PRINT” (sic).

It can be seen from the above that Barthes’ was concerned with answering the
question: “Does the image duplicate certain of the informations given in the text
by a phenomenon of redundancy or does the text add a fresh information to the
image?” (1977:38). However, this is a simplification which Nöth suggests does
not capture the fact that “the juxtaposition of picture and word usually results in a
new holistic interpretation of the scripto-pictorial or the audio-visual message”
(1990:453). Here it is not simply a question of a text-image relation of addition
and duplication, but of the ways in which both the modes work together to
produce a unified, coherent, multimodal text, a text that in Halliday and Hasan's
terms is "a semantic unit: not of form, but of meaning" (1976:1-2). It is thus a

45
question of the nature of intersemiotic complementarity. As already stated, it is the
focus of this study to attempt to describe and account for the ways that this
complementarity is realised in the Mountains text.

3.3 Summary
This chapter has briefly discussed and reviewed relevant work by semioticians and
linguists working with other modes of communication to analyse the visual mode.
A brief overview of the conception of the sign and meaning in general semiotic
theory, and a survey of the main schools of semiotic theory which have attempted
to examine other modes of communication utilising linguistic insights originally
derived from Saussure’s treatise shows that throughout the twentieth century there
has been consistent interest in explaining the characteristics of various visual
modes. Most of this work however has concentrated on single-mode analysis
however, although some have attempted to examine intersemiotic aspects. Perhaps
the most influential of these theorists has been Roland Barthes. Like many other
semioticians Barthes attempted to build upon Saussure’s semiological program,
and he contributed much to text semiotic studies of myth, theology, literature and
narrative. Of specific relevance to the focus of this study are firstly his systematic
model of signification and the two orders of signification, which he referred to as
denotation and connotation, and secondly his recognition that relay involves a
complementary relationship between both the verbal and visual (image)
constituents. What is clear from his work however is that although he recognises
that this relationship can occur between image and text, his work does not attempt
to demonstrate the semantic resources which may be utilised to realise it.

The next chapter will examine and review the work of those who have utilised the
SFL theoretical framework to investigate various kinds of visual modes. This
review will demonstrate that within this particular theoretical framework there is
also limited research into the nature of the intersemiotic semantic relationship
between the visual and verbal modes in text, and none at all in relation to
economics discourse. The culmination of the reviews in this, the previous and next
chapter will therefore be an established need for an application of the Hallidayan
SFL theoretical framework to multimodal forms of economic discourse.

46
CHAPTER 4 - Hallidayan Interpretations of Visual
Communication

4.0 Introduction
The previous chapter suggested that developments in general linguistic theory
have informed the interpretation of other communication modes besides language.
It was noted that there have been groups of researchers, usually characterised as
the Prague School and the Paris School, which have utilised linguistic concepts
derived originally from Saussure’s principles to examine non-linguistic modes.
This chapter will now concentrate on another, more recent group of researchers
who have attempted to use linguistically-derived insights to examine non-
linguistic modes. This school, which could rather loosely be referred to as the
‘Hallidayan School’, draws on the general theory of language and communication
developed by M.A.K. Halliday (1978, 1985) which is generally referred to as
Systemic Functional Linguistics (SFL), and was outlined in the previous chapter.
This school also interprets language as a socially-based semiotic system, and
drawing on the Saussurean conception of the sign and Hjelmslevian ideas
regarding the positioning of language within a broader semiotic framework, has
attempted analyses of such areas as the ‘grammar’ of images in educational
contexts, and the visual semiotics of displayed art (sculpture, architecture and
painting).

The most significant amount of work carried out by those working on a

Hallidayan interpretation of meaning-making in visual semiotic modes is
concerned with two areas: various types of images such as photographs, drawings
and diagrams, and displayed art in painting, sculpture and architecture. With the
analysis of images, two sets of publications by Gunther Kress and his co-authors,
Robert Hodge and Theo van Leeuwen are of note. The first set, by Kress and
Hodge, comprises Language as Ideology (1979) and Social Semiotics (1988), and
the second set, by Kress and van Leeuwen, comprises Reading Images (1990) and
Reading Images: The Grammar of Visual Design (1996). These four could in a
sense be read as the serial unfolding of a “social semiotic theory of
representation”, and Kress and van Leeuwen in fact refer to their ideas as being

48
part of a “fledgling movement” of “social semiotics” (op.cit:5), which they posit
as being one of three schools of semiotics, the other two being the already
discussed Prague and Paris Schools (loc.cit.). In the first set of books, Social
Semiotics is the source of much of the theoretical foundation of the two
subsequent Reading Images publications, of which Reading Images: The
Grammar of Visual Design (1996) is largely a reiteration and extension of its
predecessor, Reading Images (1990). Accordingly, this review will focus on the
latter of the two books and draw attention to differences where appropriate.

The language of displayed art is examined by O’Toole in two recent publications

(1994, 1995), wherein he reinterprets Halliday's metafunctions of language to
examine the ways that the visual modes of sculpture, architecture, and classical art
project their meanings. He applies the Hallidayan linguistic principle of Rank
Scale in constituent structure analysis (Halliday 1994:23, 35) to the interpretation
of displayed art, where it is viewed and interpreted in terms of a hierarchy of
meaningful units, and also attempts to relate these meanings to the contexts in
which they are situated. O'Toole's work will be discussed in the final section of
this chapter.

Since this study takes a Hallidayan view of communication in an attempt to

account for the ways that intersemiotic complementarity is realised in page-based
multimodal text, and seeks to test the assumption that different semiotic systems
can and do work together semantically (Halliday and Hasan 1985:4), the work by
Kress and van Leeuwen on visual grammar and O’Toole on language of displayed
art would seem to be germane. Accordingly, this chapter will review and briefly
explicate those aspects of their models/analyses which are potentially the most
applicable to the interpretation of the kinds of visuals commonly used in The
Economist magazine. These key points will be used for the analysis of the sample
text drawn from the Finance Department of The Economist, and where appropriate
to assist the reader’s understanding of how visuals are interpreted in these
approaches, some illustrative visual examples drawn from The Economist will be
analysed.

49
This chapter will consist of four main sections, the first dealing with Kress’ early
collaboration with Hodge in developing their ‘social semiotics’, the second with
Kress and van Leeuwen’s grammar of images, and the third with O’Toole’s
analysis of the language of displayed art. The fourth and final section will bring
together those aspects of the models/analyses which are deemed most applicable
to the interpretation of The Economist magazine visuals.

4.1 Social Semiotics

The first book by Kress and Hodge, Language as Ideology (1979), is a discussion
of the possible ways that language can be linked in ideological terms not only to
its social uses as a tool for communication, but also to its role as a means of
political control. Ideology in language here is viewed in two ways: in scientific
and metaphysical terms, and in terms of various kinds of political ideologies
(op.cit:6). This work only focuses exclusively on the linguistic mode however, in
an attempt to illuminate verbal language as a social phenomenon, and to provide
analytical tools for various disciplines to use in their explorations of how social
and political forces and processes act on and through discourse.

Hodge and Kress suggest that their perceptions of the limitations inherent in their
approach in Language as Ideology provided the impetus for the subsequent and
second work of the set, Social Semiotics (Hodge and Kress 1988). The aim of this
work is to produce a ‘usable linguistics’, or a ‘critical linguistics’ (op.cit:vii). The
new approach is based on two premises: the first is “the primacy of the social
dimension in understanding language structures and processes”, and the second is
that “no single code can be successfully studied or fully understood in isolation”
(op.cit:viii). This recognises that meaning is not restricted only to the linguistic
code, but “resides so strongly and pervasively in other systems of meaning, in a
multiplicity of visual, aural, behavioural and other codes, that a concentration on
words alone is not enough.” (op.cit:vii). Thus, Social Semiotics is an attempt
consider the ways that meanings are projected via a range of modes, such as
through language, images, comics and television. Kress and Hodge set up and
outline their views of a theory of communication and society, views which they
acknowledge in this and the previous publication as being derived directly from or

50
greatly influenced by M.A.K. Halliday’s work, especially his Language as Social
Semiotic (1978). These views are summarised by Hodge and Kress in the
following:
We see communication essentially as a process, not as a
disembodied set of meanings or texts. Meaning is produced and
reproduced under specific social conditions, through specific
material forms and agencies. It exists in relation to concrete
subjects and objects, and is inexplicable except in terms of this set
of relationships. Society is typically constituted by structures and
relations of power, exercised or resisted; it is characterised by
conflict as well as cohesion, so that the structures of meaning at all
levels, from dominant ideological forms to local acts of meaning
will show traces of contradiction, ambiguity, polysemy in various
proportions, by various means. So for us, texts and contexts, agents
and objects of meaning, social structures and forces and their
complex interrelationships together constitute the minimal and
irreducible object of semiotic analysis (1988:viii).

Hodge and Kress analyse some linguistic and non-linguistic modes in an attempt
to develop and explain the application of these principles, analyses which form the
foundation of the approach taken by Kress and van Leeuwen in their set of
publications. These analyses are interesting for the insights they provide in terms
of the ways that ideology is projected through different visual and verbal modes.
However, there is little to inform the focus of this study, which is to examine the
ways that these visual and verbal modes complement each other to project
meaning multimodally. This is because their analyses, like those of many other
researchers working in this area, are mode-specific — they examine how each
mode projects its meanings separately, and do not attempt to clarify what happens
intersemiotically when two modes co-occur.

4.2 Reading Images

The interest shown in the visual mode by Kress in Social Semiotics is extended in
the next set of two books co-written with Theo van Leeuwen, Reading Images
(1990) and Reading Images: The Grammar of Visual Design (1996), both of
which make the same assumptions as Social Semiotics about the nature of
communication, and which utilise similar approaches in interpreting visual and
verbal meanings in multimodal texts. The aim of both these publications is to
develop a ‘grammar’ of images, the first exploring this aim in an analysis and

51
discussion of images in children’s educational literature, and the second
continuing and extending this exploration to more generalised images drawn from
public media sources such as advertisements, magazine articles, maps, art images
and various kinds of diagrams. In both works the influence of the Hallidayan
linguistic model is much more pronounced and systematically applied than in
Social Semiotics, and there is a continuation of the application of principles
derived from critical analysis, though in a seemingly less overt way (Kress’
previous co-author Hodge is referenced in the first Reading Images and
acknowledged as an important influence in the Preface of the second). Since
Reading Images: The Grammar of Visual Design (1996) is largely a reiteration
and extension of its predecessor, this review will focus on the propositions it
presents exclusively and draw attention to differences between the two where
necessary, with the understanding that reference to Reading Images: The
Grammar of Visual Design also implies that the same points are presented in
Reading Images. For ease of identification and where it is necessary to
differentiate the two titles, the first book will be referred to as Reading Images
(1990) and the second as The Grammar of Visual Design (1996).

Kress and van Leeuwen start out in The Grammar of Visual Design by stating that
their aim in producing a ‘grammar of visual design’ is to present a socially-based
theory of visual representation (1996:1,5). In doing so they draw an analogy with
language, noting that others working in visual semiotics before them have tended
to concentrate on what could be described as the ‘lexis’ rather than the ‘grammar’
of images, in that they have concentrated on the meaning projected by the
individuals, scenes and objects portrayed within images rather than the connected
meanings (op.cit:1). The use of the term ‘grammar’ therefore implies that they will
attempt to examine the ways in which what is depicted in images is combined into
a coherent, meaningful whole, in much the same way that discourse analysts
examine how words are combined into clauses, sentences and whole texts. This is
a kind of visual discourse approach which aims, in common with many functional
linguistic models, to link form with meaning, and where linguistic and visual
“grammatical forms [are seen] as resources for encoding interpretations of
experience and forms of social (inter)action” (op.cit:1). Thus, culture and ideology

52
are also important in both the verbal and visual grammars, a point which Kress
and van Leeuwen highlight in quoting Halliday’s assertion that “grammar goes
beyond formal rules of correctness. It is a means of representing patterns of
experience … It enables human beings to present a mental picture of reality, to
make sense of their experience of what goes on around them and inside them”
(1985:101).

In both books, Kress and van Leeuwen attempt to describe the structures that
visuals use to realise a variety of different kinds of meanings, and they link their
analyses with discussions of visual literacy and the implications for education.
They do this because of their perception of the overwhelming importance of visual
communication in the modern world, the dominance of the verbal over the visual
in educational systems, and their view that there is a " staggering inability on all
our parts to talk and think in any serious way about what is actually
communicated by means of images and visual design" (1990:3, 1996:16). This last
view is one that could be readily taken issue with, especially considering that
Kress and van Leeuwen do draw very heavily on selected publications from a
considerable body of work by researchers in such areas as communication and
media studies (Dondis 1973; Dyer 1982; Fiske 1982; et.al.), studies on the
psychology of visual perception (Arnheim 1969, 1974, 1982; Gombrich 1960
et.al.), information design (Tufte 1983 et.al.) and visual semiotics (Barthes 1967,
1977; Eco 1976; Saint-Martin 1987 et.al.), in their interpretations of the ways that
images do project their meanings. What is new and interesting in Kress and van
Leeuwen’s approach to the interpretation of visual meaning however, is the
application and adaptation of linguistic insights from the socially-based SFL
model in an attempt to link the visual meanings in an image to the producers of
that image and their particular social contexts.

In their rationale, Kress and van Leeuwen place themselves at variance with
Barthes' (1977) view of the meaning of the image, and do not totally accept his
notion of dependency between image and verbal text, where he suggests that the
meaning of images (as well as other codes such as food, dress etc.) are related to
and mostly dependent on language for ‘fixing’ their meanings. They argue that

53
while Barthes' essay on image-text relations explains elements of the
communicative relationship between the two codes, it fails to recognise that "the
visual component of a text is an independently organised and structured message
— connected with verbal text, but in no way dependent on it. And similarly the
other way round." (Kress and van Leeuwen 1996:17). Their aim therefore, is to
utilise underlying principles in the 'grammar' of the verbal to explicate the
'grammar' of the visual, taking the point of view that "language and visual
communication both realise the same more fundamental and far-reaching systems
of meaning that constitute our culture, each by its own specific forms, and
independently" (loc.cit.). The implication here is that both verbal and visual codes
are seen to express the same kinds of meanings, but by different methods which
draw on different semiotic systems. In any particular cultural context (Western
European, for example), there may be a considerable degree of congruence
between the two codes. There may also be areas of difference, areas where the
verbal can express itself and the visual cannot, and vice versa. The two semiotic
systems and their potential meanings are therefore neither fully conflated, nor are
they wholly opposed in their respective codes (op.cit:17-18). Further, both these
modes are realisations of social semiotic systems, wherein the meanings which all
communicators (whatever the code) choose to express are seen to be social in
nature, and arise out of the culture in which they are situated.

As already mentioned, Kress and van Leeuwen utilise Halliday’s (1978, 1985)
SFL theory to provide an analogy for the development of a visual grammar and to
outline the kinds of categories which they regard as essential to the analysis of the
visual semiotic. They suggest that
The visual, like all semiotic modes, has to serve several
communicational (and representational) requirements, in order to
function as a full system of communication. We have adopted the
theoretical notion of ‘metafunction’ from the work of Michael
Halliday for the purpose of dealing with this factor. The three
metafunctions which he posits are the ideational, the interpersonal,
and the textual (1996:40).

Kress and van Leeuwen then posit that reading (or viewing) a visual involves two
kinds of participants: the interactive participants, and the represented participants

54
(op.cit:45-46). The former are the participants who are interacting with each other
in the act of reading a visual, one being the graphic designer/photographer/artist,
and the other the viewer. This category represents the social relations between the
viewer and the visual. The latter is all the elements or entities that are actually
present in the visual, whether animate or inanimate, elements which represent the
situation shown, the current world-view, or states of being in the world. Kress and
van Leeuwen also assert that reading (or viewing) a visual involves reading a
structurally coherent arrangement of elements which combines and integrates
these two kinds of participants, thus representing the structuring of the current
world-view (op.cit:40-41). By applying Halliday's concept of metafunctions to
other modes beside the linguistic, Kress and van Leeuwen are clearly assuming
that the visual mode draws upon the same semantic system as does language and
that "everything [which can be] said about the semiotic code of language can be
said, in terms specific to it, about the semiotic code of pictures" (1990:19). Also,
while visuals do differ in terms of the choices from the cultural semantic system
that they can realise and in the ways in which these choices are realised, still “the
semiotic code of language and the semiotic code of pictures each have their own
quite particular means of realising what in the end are perhaps quite similar
semantic relations” (1996:44).

The three elements, the interactive and represented participants, and the coherent
structural elements of a visual, are correlated with Halliday's three metafunctions
and may be summarised briefly as:
• IDEATIONAL: a range of ways of semantically relating represented
participants.
• INTERPERSONAL: a range of ways of semantically relating interactive
participants.
• TEXTUAL: a range of ways of semantically relating the elements on a page to
each other.

The categories of the visual grammar proposed by Kress and van Leeuwen are
summarised in Tables 4.1, 4.2 and 4.4 following. What is of immediate notice is
that the metafunctional terminology posited by Halliday to describe meanings at
the semantic level in his SFL model have been changed from ideational to
Representational meanings, from interpersonal to Interactive meanings, and from

55
textual to Composition(al) meanings. Why this was done is not made clear, but
Kress and van Leeuwen do adopt much of the lexicogrammatical terminology
utilised by Halliday as the following discussion will show. This includes the
terminology used in the grammatical system of TRANSITIVITY, which
“construes the world of experience into a set of manageable process types”
(Halliday 1994:106), the grammatical system of MOOD where the clause “is also
organised as an interactive event involving a speaker, or writer, and audience”
(op.cit:68), and THEMATIC STRUCTURE, which “gives the clause its character
as a message” (op.cit:36). Kress and van Leeuwen also attempt to take account of
the paradigmatic features of Halliday’s systemic grammar by representing the
various aspects of their visual grammar in terms of system networks, or networks
of potential choices from which the people who create a visual can make
selections. This is an attempt to capture the SFL interpretation of language as
meaning potential, where interlocutors make choices from various meaning
systems in the act of communicating in various contexts.

It should be pointed out at this juncture that the analysis and interpretation of
visual forms of communication involves examining them from different points of
view, and that in focussing on one point of view in particular, for example
Representational meaning and its sub-systems, it is easy to lose sight of the fact
that a visual is the result of “the convergence of many different signifying
systems” (Kress and van Leeuwen 1996:265). Thus it is necessary to keep in mind
that the visual systems of Representational (ideational), Interactive (interpersonal)
and Compositional (textual) meanings occur and project their meanings
simultaneously, and that they are multidimensional structures. Like sentences,
which can be simple (only one clause or process), or complex (several clauses,
each with their own process, and hypotactically and paratactically related to each
other), visuals can also be simple or complex (op.cit:112).

The next part of this chapter will overview and assess Kress and van Leeuwen’s
visual grammar in terms of those aspects which may be applicable to the analysis
of The Economist magazine visuals, and which could be utilised in the framework

56
developed for the analysis of intersemiotic complementarity. A small sample of
visuals from The Economist are reproduced and presented where appropriate.

4.2.1 Representational Structures

This section will discuss and evaluate Kress and van Leeuwen’s description of the
ways that various kinds of visuals organise and represent their meanings
representationally (ideational). The two sections following this one will then
discuss and review the ways that visuals attempt to address their potential viewers
in interpersonal terms, and the ways that visuals organise their meanings on the
page, in compositional or layout terms.

In their discussion of representational structures in the visual transitivity system,

or the visual resources for representing interactions and conceptual relations
between people, places and things in visuals, Kress and van Leeuwen recognise
two major processes: Conceptual and Narrative processes (1996:56). Both these
processes represent patterns of experience and phenomena in terms of sequences
of process configurations, and configurations of processes, participants, and
circumstances, objects, qualities, and quantities. Conceptual processes are seen to
“represent participants in terms of their generalised and more or less stable and
timeless essence” (op.cit:56), and “in terms of class, or structure, or meaning”
(op.cit:79), while Narrative processes or patterns (which were formerly referred to
as Presentational processes in Reading Images 1990) “serve to present unfolding
actions and events, processes of change, transitory spatial arrangements”
(op.cit:56). Narrative processes therefore deal with depicted actions and events,
rather than depicted states of being which have the essence of constancy, and their
spatial arrangements are in a sense transitory, rather than being concerned with a
fixed and constant spatial order, as in the conceptual processes. Kress and van
Leeuwen approximate the conceptual processes with the existential and relational
processes of Halliday's TRANSITIVITY system, and the narrative processes with

57
 · Action (Actor + Goal)
· Reactional (Reactor + Phenomena)
Processes · Speech & Mental
· Conversion
Narrative · Geometrical symbolism
Representations

· Setting
Circumstances · Means
· Accompaniment

Representational Classificational · Covert

Structures Processes · Overt (Single or multi-leveled)
(ideational)

· Unstructured
Conceptual · Structured
Representations 1. Temporal
Analytical Processes 2. Exhaustive and inclusive
3. Conjoined & compounded exhaustive
structures
4. Topographical and topological processes
5. Dimensional and quantitative topography
6. Spatio-temporal

Symbolic · Attributive
Processes · Suggestive

Table 4.1 Representational visual structures (Ideational) adapted from The Grammar of
Visual Design (1996).

58
the material, behavioural and in a limited way mental and verbal processes, again
making the point that they recognise that “while visual structures and verbal
structures can be used to express meanings from a common cultural source, ……
the two media are not simply alternative means for representing ‘the same thing’”
(op.cit:75-76). These visual representational processes are summarised in Table
4.1.

4.2.1.1 Narrative Processes

Within Narrative processes in visuals Kress and van Leeuwen recognise two
major kinds: Actional and Reactional processes. Actional processes are those in
which some kind of physical action relates the represented participants. They can
be non-transactional, where there is only one participant and therefore no action
directed towards anyone or anything, or transactional, where there are two or
more participants and something is in a sense exchanged. In this situation there is
an actor, as well as a goal and a beneficiary. Both transactional and non-
transactional action relations approximate the transitive and intransitive verb
distinction in language, and are realised by a vector, a line that can be visually
projected from the actor (the most prominent of the participants), and extended
towards the other participant(s) or some goal (1990:82, 1996:61-64). The actor’s
prominence can be realised by relative size, place in the composition, contrast
against a background, colour saturation or prominence, sharpness of focus, and the
'psychological salience' which the human face has for viewers (1996:64). There
may also be more than one actional process going on in one image which are
secondary transactional processes (termed 'minor' processes). These are an
instance of what Kress and van Leeuwen refer to as Embedding in images. This
difference between 'major' and 'minor' processes and therefore the difference
between 'major' and 'minor' participants can be realised via a range of visual
techniques, such as relative size and relative conspicuousness of the elements
portrayed (op.cit:113).

Reactional processes are processes in which the represented participants are

characterised by a reaction, which is realised by the direction of the glance of one
of the participants, the reacter. Since a reaction to something is necessarily a trait

59
of living beings, the reacter should be human, or a human-like animal (with visible
eyes and the ability to produce facial expressions). The direction or focus of this
reaction (a look or gaze) follows a vector to the receiving participant or whole
process, termed the phenomenon, and the particular nature of this reaction is
encoded in the way that the reacter is looking at the phenomenon (1990:83-84,
1996:64-67).

In The Grammar of Visual Design (1996) Kress and van Leeuwen add three more
visual narrative processes which were not given in Reading Images (1990). These
three processes, which are rare or do not occur at all in the kind of visuals used in
The Economist magazine, are speech and mental processes, conversion processes,
and geometrical symbolism. The speech and mental processes approximate
Halliday’s ‘projection’ (1994:219), and refer to the vectors which can be drawn
from speakers or thinkers to their thoughts, or dialogue speaking balloons in
comic strips, and more recently speaking images on computer display screens.
These are transactional, and connect a living being with attendant speech or
thoughts (Kress and van Leeuwen 1996:67). Conversion processes refer to the
kinds of processes found in diagrams portraying chained processes and cycles of
action (as in a Water Cycle diagram), where there is “a participant which is a goal
with respect to one participant and the Actor with respect to another” (op.cit:68).
Geometrical symbolism refers to the kinds of diagrams where there are no
participants, but the symbolic nature of the diagram suggests an action which can
be identified by a vector, as in a diagram representing the spiral action of a
‘twister’, or a simple arrow, which simply indicates a direction.

Kress and van Leeuwen (op.cit:71-73) also recognise that there are secondary
participants in many images, identified as Circumstances, which are defined as
“participants which could be left out without affecting the basic proposition
realised by the narrative pattern, even though their deletion would of course entail
a loss of information” (op.cit:71). Following Halliday (1994:149ff), Kress and van
Leeuwen recognise three forms. These are Locative Circumstances, Circumstances
of Accompaniment, and Circumstances of Means. These circumstantial
participants

60
Visual 1: Unemployment - Little shop of horrors plant (The Economist, April 25th 1992, p57).

61
are not related to each other by vectors, but via aspects of location (thus the
Setting becomes a participant), or by simply co-occurring together in the same
visual with no obvious relationship other than being with one another in the frame
(a relation of Accompaniment), or finally by being used by a participant in some
action (a relation of Means) where there may be an absence of a clear vector
between the user and the tool, but which may often assist in the formation of
vectors between the most salient Actor and the Goal (1996:71-73).

As already mentioned, visuals are multidimensional in that they simultaneously

utilise a range of different visual processes in order to project their meanings to
their viewers. A good illustration of this kind of complexity can be seen in Visual
1, which contains a number of embedded major and minor processes. Using Kress
and van Leeuwen’s terminology, this visual reveals:
1. A major narrative transactional process, that of the threat posed by an
overhanging “Little Shop of Horrors” plant (as actor) towards the young shoots
(as goal). This is realised firstly by the vector formed from the plant’s mouth
towards the plants — notice that the gardeners are not under threat from this
mouth, as there is no vector which can be drawn from it towards them — and
the relative size of the overhanging plant, which makes it visually more salient
than the other represented participants.

2. A major narrative transactional process, which is perhaps competing in terms

of its ‘visual force’ with the threatening process of the plant, is that of the
gardening/tending action of the three other represented participants (actors)
towards the young shoots (goal). This transactional relation is realised by
vectors which can be drawn from the eyes of the three figures directed towards
the young shoots.

3. A minor narrative transactional process enacted by the kneeling man (actor)

who is touching (tending to) the young shoots (goal).

4. Participants acting as circumstances of means: the implements which the three

‘gardeners’ are holding work to support the gardening/tending process
interpretation as they also form vectors towards the goal, the young shoots, and
they represent circumstances of means to that process.

From this multiple analysis it can be seen that in terms of Representational

meanings only (and not Interactional and Compositional), a visual can be a very
complex multidimensional configuration.

4.2.1.2 Conceptual Processes

62
Within Conceptual processes in visuals Kress and van Leeuwen recognise three
major kinds: Classificational, Analytical, and Symbolic Processes (the first two
Conceptual processes were originally described as Descriptive Classificational,
and Descriptive Analytical in Reading Images, and an extra set of processes,
Genealogical Processes, has been omitted in The Grammar of Visual Design). In
Conceptual processes the participants are represented in generalised stable, and
timeless terms, and in terms of class, or structure, or meaning (1996:79). They
therefore have the sense of projecting states of being which have the essence of
constancy, rather than being concerned with a fixed and constant spatial order.

Classificational processes are those which relate the participants in a 'kind of'
relation, in terms of a taxonomy of types of things, or classes of things (op.cit:79-
88). Each of the participants is presented as a typical (Subordinate) member of a
particular overarching (Superordinate) category or class of things, and this can be
either a Covert Taxonomy or an Overt Taxonomy. The Covert taxonomy is realised
by a symmetrical spatial arrangement of the participants, such that any similarity
between them as members of a particular Superordinate class is enhanced by the
symmetrical arrangement in their placement on the page. This is uncommon in
The Economist magazine, but often seen in photographic advertising, where the
aim is not to show every single detail clearly, but to emphasise the relevant,
salient details to attract the viewer (potential buyer) to them. In educational
diagrams (in for example a Geography text) there may be an emphasis on the
depiction of generic items, items with common characteristics, rather than the
features of specific items from the same class. Overt taxonomies however, include
the Superordinate participant in the frame. This process is most commonly
represented as a tree diagram, a series of photographs joined in the same tree
organisation and linked together by drawn lines, or some kind of a pyramid
structure. All these structures can be single levelled or multi-levelled in terms of
the layers of subordinate classes. In The Economist magazine these kinds of visual
structures are not common, typical instances being more likely to occur in
pedagogical texts.

63
Analytical processes, on the other hand, relate the participants in an image in
terms of part/whole relations, where one participant represents the ‘whole’,
referred to as the Carrier, and the other participants (of any number) are the
‘parts’, termed the Possessive Attributes. This relation may be Structured, where
the Carrier and the Possessive Attributes are shown, or they may be Unstructured,
where only the Possessive Attributes are depicted, and the viewer is left to deduce
the nature of the Carrier. This latter type does not occur in The Economist
magazine, but a typical instance would be a dress-making pattern diagram which
shows the parts of a dress but not the finished item (op.cit:94). Within the
Structured analytical processes however, Kress and van Leeuwen identify six
other categories of processes which can be spatially or temporally organised, the
final two of which are certainly relevant to The Economist. These six categories
are:
(1) the Temporal,
(2) the Exhaustive and Inclusive,
(3) the Conjoined and Compounded Exhaustive,
(4) the Topographical and Topological,
(5) the Dimensional and Quantitative Topography, and
(6) the Spatio-temporal (op.cit:95-108).
The two relevant processes referred to as Dimensional and Quantitative
Topography typically involve pie charts and bar charts that are drawn to scale, but
the scale is not based on the actual physical dimensions of the participants but “on
the quantity or frequency of aggregates of participants that are taken to be
identical” (op.cit:103). Pie charts and bar graphs both involve visual
representations of relative size, where for example each numerically or
proportionally-based sector in a pie chart is shown in relation to the other
numerically or proportionally-based sectors. In Kress and van Leeuwen’s sense
both pie and bar charts are also dimensional, in this case one-dimensional, in that
they show only one carrier with its Possessive Attributes, which are quantitative in
nature (op.cit:104). A very common instance of this from The Economist
magazine is Visual 2, where the pie chart and bar graphs both divide their Carriers
into quantities or aggregates; the pie chart’s carrier (redemption periods for
securities) is divided up into the Possessive Attributes based on time to maturity,

64
while the bar graphs’ carrier (also redemption periods for securities) is divided up
by country into another series of aggregates for comparison purposes. In both
these cases quantitative figures are translated in visual representations of relative
size — for example in the pie chart each sector shows proportionally the different
times to maturity in relation to the other times; this means that it is not showing
that the 2-10 years redemption figure is actually located next to the 1-2 years
figure, but how the number or percentage of the 2-10 years time period stands in
relation to the number or percentage of 1-2 years time period.

In Spatio-temporal processes there is a conjunction between a set of spatially

oriented participants (as in a line graph representing volume, amount, frequency,
and proportion in the same way as a bar chart can) and a time line, giving an
impression more of rates of change than relative proportion over time. In Kress
and van Leeuwen’s sense this is two-dimensional (1996:105-106). A very
common visual of this type in The Economist magazine, and therefore highly
relevant to this study, is the simple two-dimensional line graph on the left-hand
side of Visual 2 (following page), which shows the behaviour of some data
(maturity periods in months) changing in relation to the passage of time (in years).
This graph has a specific year (1945) as a point of origin, and the discrete
chronological stages are actually individual points of intersection between years
and average maturity periods. These are bound so tightly together into a single
line however that the pre-eminent impression is one of a dynamic process of
change, growth and fluctuation. One would not initially read the graph by
thinking: “in 1946 the average maturity was 125 months, but in 1975 it was about
30 months” and so on. The initial reading would be one which absorbs holistically
the impression of fluctuations and change, and not in a piecemeal, fragmented
fashion (which is not to say it couldn’t be read in that way and may in fact occur
in a more detailed analysis).

The final set of processes in Conceptual Representations are Symbolic processes.

These are about what a participant means or is (op.cit:108). In other words, these
processes are connected to the symbolism or messages conveyed by the

65
participant relations portrayed. There can be either two participants in the relation
or only one.

Visual 2: America’s government debt (The Economist, June 5th 1993, p71)

66
Where there are two participants, the participant whose meaning is established in
the relation is the Carrier, and the participant which represents the meaning or
identity itself is the Symbolic Attribute. This is the Symbolic Attributive process,
where objects in images (such as in artworks) are made significant or prominent
by such things as being foregrounded, having exaggerated size, being well-lit, or
containing strong colours. They are also pointed out by some kind of gesture or
posture which can only be interpreted by the viewer as the action of pointing out
the attribute to the viewer. They can also look slightly incongruent (or out of
place) in terms of the whole of the visual, and they may have some kind of
conventional symbolic value, as in a wire fence as a conventional symbol of
imprisonment, or barrier to entry (loc.cit.).

This Symbolic Attributive process and the kind of visual representation used to
project it is very common in The Economist magazine, where well-known, public
figures are often portrayed in such a way that some symbolic attributes are
strongly attached to them. In The Economist sketches are commonly used, a
medium which can be more easily manipulated to portray various meanings
(photographs are rarely if ever used to do this, since they are not so easily
manipulated as sketches). A typical example is Visual 3 (following page), which
is a sketch portraying a caricatured human Carrier (Lloyd Bentsen) holding
upright (and therefore displaying) two implements which are the Symbolic
Attributes that confer on him the meaning of cleaner or protector/exterminator.
The interpretation of him as cleaner is supported by the symbolic meaning
attached to the cleaning brush (to be used on domestic banks), which is reinforced
by the proximity of a verbally-labelled cleaning bucket and the fact that he is
wearing an apron, a universal symbol of house-cleaning and domestic functions.
The Symbolic Attribute of protector/defender is realised by the spiked club, which
is a universal symbol of defence and of the classical warrior who must defend his
‘castle’ from the enemies without (identified verbally as the foreign banks).
Further, this visual is posed for the viewer; there is no action to speak of except
the fact that he is displaying these two implements for the viewer to notice and to
‘get the symbolic meaning’, a meaning which is also perhaps reinforced by the

67
differences in colour between the

68
Visual 3: America’s banking battles (The Economist, October 30th 1993, p79)

69
two clubs — one is white (for the good guys?) and one is black (for the bad
guys?). It could be argued also that this ‘display’ is reinforced by the gaze of the
main represented participant directed towards one of the symbolic attributes,
which is in a way saying ‘look at this’ to the viewer, and as a corollary is pointing
out its symbolic meaning.

Where there is only one participant, also termed the Carrier, its symbolic meaning
can be established in another way which suggests an interpretation. This is the
Symbolic Suggestive process, which is characterised by a de-emphasising of image
detail in favour of 'mood' or 'atmosphere'. In this the focus is off the main
participant as being the only primary participant, and there is a sense of an
indirect, suggestive attempt to give to the carrier some attribute. This is very
common in advertising images and calendars which attempt to give the viewers
some sense of the ‘atmosphere’ of the scene, but it is relatively uncommon in The
Economist magazine.

To sum up then, in their discussion of representational structures in the visual

transitivity system, or the visual resources for representing interactions and
conceptual relations between people, places and things in visuals, Kress and van
Leeuwen have recognised two major visual processes they refer to as Conceptual
and Narrative processes (1996:56). Both these processes are used to represent
patterns of experience and phenomena in terms of sequences of process
configurations, and configurations of processes, participants, and circumstances,
objects, qualities, and quantities in the visual mode. They appear, therefore, to
adapt the conceptual meaning of the clause constituents in the SFL Transitivity
system and to utilise them for the visual Transitivity system. These functional
constituents of the clause are referred to as Participant, Process and Circumstance,
as well as Goal (Material Processes) and Attribute (Relational Processes), and
would seem to be terminologically useful for the description of multimodal
meanings. However, it would seem that the specific terms used by Kress and van
Leeuwen to identify these major visual processes may not be of immediate
terminological use in this study, because they have no real general terminological
counterparts in the SFL model and they relate specifically to the visual mode (this

70
is discussed more fully in Chapter Five). Their underlying conceptual basis is
however applicable and relevant in terms of the kinds of visuals commonly used
in The Economist magazine, because they approximate the most common forms
used in the magazine, visuals which often represent these kinds of processes, i.e.
sketches and graphs/charts. Their underlying conceptual elements also are relevant
to what Kress and van Leeuwen refer to as visual ‘coding orientations’, a term
which will be explained more fully in the next section on Interactive meanings and
applied to visuals which display naturalistic and quantitative meanings.

4.2.2 Interactive Meanings

Up to this point this review has discussed Kress and van Leeuwen’s description of
the ways that various kinds of visuals organise and represent their meanings
representationally (ideational). This section deals with a discussion of the ways
that visuals attempt to address their potential viewers in interactional
(interpersonal) terms in their visual grammar. In their discussion of interactive
meanings in their visual grammar, Kress and van Leeuwen recognise that visual
forms of communication also utilise resources which both constitute and maintain
interaction between the producer(s) and viewer(s) of a visual. As pointed out
earlier, Kress and van Leeuwen posit that reading (or viewing) a visual involves
two kinds of participants, the interactive participants, and the represented
participants (1996:45-46). The former are the participants who communicate with
each other via visual means (the photographer and the viewer(s) for example), and
the latter are what is actually depicted in a visual (the people, places and things
shown). As it is between speaker/writers and listener/readers, the interpersonal
metafunction in relation to visual communication is also concerned with the
representation of social relations, in this case between the visual and the viewer.
This is important because the placement of the viewer and the visual socially will
have a significant influence on how the visual is read and used, or in other words,
their relative social placement affects what may be represented, the ways that it is
represented, as well as how it may be read and put to use.

71
 Image Act · Offer (Information)
Contact · Demand (goods/services)
Gaze · Direct (degrees of Engagement)
· Indirect (degrees of Disengagement)
· Close (Intimate/Personal)
Social Distance Size of Frame · Medium (Social)
· Long (Impersonal)
· Horizontal angle (degrees of Involvement &
Subjective Image Detachment)
Attitude · Vertical angle (degrees of Power to the viewer, to the
represented participants, or a relation of equality)
Objective Image · Action Orientation (frontal angle)
· Knowledge Orientation (top-down angle)
· Colour saturation
Interactive Colour · Colour differentiation
Meanings · Colour modulation
(interpersonal) Contextualisation · Absence of background
· Full detail
Representation · Maximum abstraction
· Maximum representation
Depth · Absence of depth
Modality · Maximally deep perspective
Illumination · Full representation of light and shade
· Absence of light and shade
Brightness · Maximum brightness
· Black and white or shades of light grey and dark grey
· Technological
· Sensory
Coding Orientation · Abstract
· Naturalistic

Table 4.2 Interactive meanings (Interpersonal) adapted from The Grammar of Visual
Design (1996).

72
Thus, viewing a visual involves "being located in a particular social way by and in
relation to the image" (Kress and van Leeuwen 1990:23). This happens despite the
fact that the producer of the visual is more often than not absent physically from
the communicative situation. Even though there is this disjunction between the
context of production and the context of reception, the “two do have elements in
common: the image itself, and a knowledge of the communicative resources which
allow its articulation and understanding, a knowledge of the way social
interactions and social relations can be encoded in images” (Kress and van
Leeuwen 1996:120). The visual resources for creating or maintaining this social
engagement between the viewers and the image are summarised in Table 4.2 and
will be overviewed in the following sections. Kress and van Leeuwen emphasise
again that these visual dimensions should be considered as “simultaneous
systems” in that “any image must either be a ‘demand’ or an ‘offer’ and select a
certain size of frame and select a certain attitude” (op.cit:153). As was done in the
preceding section, only those aspects of Kress and van Leeuwen’s framework
which are relevant to the analysis of the kinds of visuals commonly used in The
Economist magazine and which could be utilised in the framework developed for
the analysis of intersemiotic complementarity will be overviewed and assessed. A
small sample of visuals from The Economist are reproduced and presented where
appropriate.

4.2.2.1 Contact
The visual resources used to establish and maintain contact between the image
and the viewer are concerned with the ways that images directly or indirectly
address their viewers, and what they simultaneously require them to do
(op.cit:119-121). The aspect of importance here is the relation between the sets of
participants involved in the viewing of the visuals, the interactive participants. In
conventional spoken communication, the interactive participants are usually seen
as the speaker-listeners (who can in turn reverse roles). There are also the
participants who are represented participants by virtue of being spoken or written
about. They may not physically be there (a typical situation), or they could be the
speakers and listeners themselves. With visuals however, there is no physical
speaker or drawer actually there; the viewer is alone with the visual, and there is

73
no opportunity for turn-taking as there is in normal speech-based communication.
The situation with the interactive participants in viewing visuals is not dissimilar
to that of the writer-reader interactive relationship, where communication is
carried out orthographically. The interactive relationship which is common in
visuals is one where there is an interactive participant (viewer or viewers) and one
or more represented participants which have been drawn or produced for the
viewer(s).

Taking Halliday's notion of speech functions and the speech acts they can realise,
Kress and van Leeuwen draw a distinction between two kinds of image acts.
These are demands and offers. In speech, one can offer information (make a
statement), which can then be agreed with or contradicted, or one can offer goods
and services (make an offer), which can then be accepted or rejected. Also, in
speech one can demand information (ask a question), which can then be answered
or disclaimed, or one can demand goods and services (give a command), which
can then be obeyed or refused (op.cit:127-129). When images offer however, they
most commonly offer information, and when they demand, they most often
demand the particular goods and services which would realise a particular social
relation, or some kind of response from the viewer. Thus, as Kress and van
Leeuwen assert, in Western cultures, visuals generally perform only these two
image acts, and not the full range that is possible with the four primary speech
functions.

The realisation of a visual demand is determined by the presence or absence of a

gaze, which indicates a form of direct or indirect address to the viewer (op.cit:121-
130). The producer is thus using the image to act on or do something to the
viewer. The gaze always takes the form of a vector formed by the glance of one or
more of the animate represented participants outwards to the viewer of the visual.
This may often be supported by some kind of physical gesture. The animate
represented participant(s), which could be human or animal (mostly human), will
demand something via one or more pairs of eyes looking directly at the viewer,
and what is demanded by the represented participant(s) in the image depends
entirely on how the look is conveyed. There may be a smile (suggesting social

74
affinity), a stare (suggesting disdain), or a pout (suggesting a sexual offer). Each
of these actions requires some kind of response from the viewer in terms of
entering into some kind of social relation, which in this case is to accede to or
deny the demand (op.cit:122-123). In The Economist magazine instances of visual
demands are quite rare, except perhaps in some of the advertising in each issue.
Within The Economist’s journalistic articles however, visual demands occur only
rarely in the form of posed portrait photographs which look directly at the viewers
and basically require an acceptance of the message they are projecting about
themselves.

In visual offers, on the other hand, there is not the requirement of the viewer to
enter into some kind of imaginary social relation with the represented participants.
The viewer is however asked to examine or look at the represented participants as
objects for uninvolved viewing, objects which really have no socio-relational call
on the viewer. In visuals where offers are being made, the represented participants
are always looking away from the viewer - there is no look being projected
directly at the viewer. This kind of image is an offer because “it ‘offers’ the
represented participants to the viewer as items of information, objects of
contemplation, impersonally, as though they were specimens in a display case”
(op.cit:124).

4.2.2.2 Social Distance

Visually-based interactive resources are also used to determine how much social
distance there is between the viewers and the represented participants. Social
distance is concerned with the kinds of social relations between interactants,
whether they be long-term types in which they may be considered as intimates,
friends, colleagues, acquaintances, total strangers, or even aliens (in a cultural
sense), or short-term relationships which last only as long as the actual
communication act, and are largely context-dependent. Kress and van Leeuwen
suggest that the use of size of frame, in a similar way to the ways it is used in
cinematography, can be used to convey a sense to the viewer of his or her social
closeness in relation to the represented participants (op.cit:130-135). These social
relations are realised by varying the size of frame through the use of very close-up

75
shots, close-up shots, medium shots, and long or distance shots, all techniques
which derive from cinematography and photography. These techniques are used
even when the participants represented may be complete strangers to the viewers.
In most cases (especially in a magazine like The Economist), the represented
participants are complete strangers to the viewers, although they may well be
famous people who are in a popular media sense well known to them (as in pop
stars, politicians, sports stars, etc.). Whether they are known to the viewers or not
in this populist sense however is not important in determining the choice of frame
size. In visuals, the social
relation between the human participants represented in images and
the viewer is once again an imaginary relation. People are
portrayed as though they are friends, or as though they are
strangers. Images allow us to imaginarily come as close to public
figures as though they were our friends and neighbours - or to look
at people like ourselves as strangers, ‘others’ (op.cit:132).

The ways these differences in social relation are realised in visuals portraying
human represented participants is realised by the choice of a type of shot from a
continuum of shots. These are summarised in Table 4.3.

FRAME SIZE CHARACTERISTICS SOCIAL RELATION

very close up less than head and shoulders of subject intimate

close shot head and shoulders of subject friendly or personal

medium close cuts off subject approximately at waist social or ' one of us'

medium shot cuts off subject approximately at knee level 'familiar' social

medium long shows full figure general social

long shot human figure fills half image height public, largely impersonal

very long shot and any thing beyond (wider) than half height little or no social connection
Table 4.3 Size of Frame and Social Distance (op.cit:130)

Kress and van Leeuwen suggest further that the methods used to create a sense of
social distance can also be applied to the representation of objects and of the
environment, a suggestion not given in Reading Images (1990). They tentatively
express the idea that
at least three significant distances can be distinguished, and that
there are correspondences between these distances and our
everyday experience of objects and of the environment, in other

76
 words, that size of frame can also suggest social relations between
the viewer and objects, buildings and landscapes (Kress and van
Leeuwen 1996:134).
Thus, a close-up shot implies engagement, that the object can almost be touched
or used, a middle distance shot gives a sense of more distance and less
engagement, while a long distance shot suggests that the object is there for the
viewer’s contemplation only, and that there is no possibility of engagement
beyond this. This is an area that is explored more fully by O'Toole's (1994) work
on the language of displayed art, which will be discussed in section 4.3 in this
chapter.

4.2.2.3 Attitude
In their discussion of the ways that certain visual resources can be used to express
an attitude or 'point of view' towards the represented participants in visuals, Kress
and van Leeuwen examine the importance of the system of (central) perspective,
and how it is used to express subjective attitudes. Perspective is defined as a
graphic design technique used for depicting volumes and spatial relationships on a
flat surface, in such a way that the represented objects appear to the eye as a
typical visible scene with respect to the viewed objects’ relative positions and
distance (Random House 1992). The system of perspective was developed during
the Renaissance, a period when individual expression of one's point of view and
subjectivity were inculcated as worthwhile social values. Kress and van Leeuwen
assert that since the Renaissance there have been two kinds of visuals in Western
cultures. There have been the objective (without perspective - no built-in point of
view), and the subjective (with central perspective - carrying a built-in point of
view).

Subjective visuals project a very clear distinction between the represented world,
which requires most of the viewer's attention, and the frame or physical space in
which the image is viewed. In these types of visuals there is an attempt to create a
division between the visual and its environment, to emphasise its main features as
opposed to the features of its background (Kress and van Leeuwen 1996:136). In
objective visuals however, taking account of the viewer is ignored and there is a
disregard of any reference to the viewer in terms of who or where he or she is, or

77
the actual time setting. The designer of the subjective visual has selected the point
of view for the viewer, and consequently there is a kind of designer-to-viewer
relationship set up whereby the point of view of the designer, via the imposed
point of view in the represented participants in the visual, is 'forced' onto the
viewer. As Kress and van Leeuwen state, the "perspectival image compels the
viewer to become, together with the institution that has produced the image, an 'us'
with respect to a 'them' or an 'it' - that is, with respect to the participants
represented in the image" (Kress and van Leeuwen 1990:32).

Kress and van Leeuwen examine aspects of the subjective features of visuals in
depth, dividing them into two possible simultaneous choices which they refer to as
degrees of involvement and power. In the former there can be either involvement
or detachment, and in the latter there can be viewer superiority, viewer equality, or
viewer inferiority. Both the categories of involvement and power are realised
through various choices of visual horizontal and vertical angle respectively. The
horizontal angle relates to the interaction between the frontal plane of the visual
designer or producer (as in a photographer) and the frontal plane of the
represented participants. The are a number of possibilities here. They can have
parallel alignment, and thus have a frontal point of view, or they can be at an
angle, and thus have an oblique point of view. This frontal to oblique distinction
may be seen as a continuum of degrees of obliqueness, as opposed to a strict
either/or dichotomy, and is related to the relative placement of vanishing points.
Vanishing points are the points at which receding parallel lines appear to
converge, and can be derived when lines are drawn through the visual along the
perceived linear arrangements of the participants. They are important in that they
control the direction of view (Porter and Greenstreet 1980:76). This is illustrated
in Figures 4.1 and 4.2.

A frontal angle occurs when the vanishing point(s) can be placed within the
vertical boundaries of the visual, as shown in Figure 4.1, whereas an oblique angle
occurs when the vanishing point(s) can be placed outside the vertical boundaries
of the image, as shown in Figure 4.2. The range of oblique and frontal angles
presents different degrees of involvement. The horizontal angle of a visual

78
therefore encodes whether or not the graphic designer (and concomitantly the
viewer) are in some

79
Figure 4.1 - The frontal angle (Montague 1993:8)

80
Figure 4.2 - The oblique angle (Dondis 1973:49)

81
kind of relationship with the represented participants or not. The frontal angle
projects a message of inclusion, suggesting that the represented participants are
part of the viewer's and visual designer's world, while the oblique angle projects a
message of exclusion, suggesting that the represented participants or the scene is
not part of their shared world, and as a result there is a lack of involvement (Kress
and van Leeuwen 1996:142-143).

The vertical angle is an important element in interpersonal meanings in visuals in

that it allows for the establishment of power relations between the viewer and the
represented participants. These relations are also important in cinematography,
and operate in the ways that visuals attempt to project meanings. The major
difference is that the viewer is the subject of having or not having the power,
rather than the participants in a film or video. Thus, if the represented participant
is viewed by the viewer from a high angle, or from 'above' as it were, then the
interactive participant (the viewer) is deemed to have a more powerful position
relative to the represented participant. The relation is reversed when the angle is
one of the represented participant being in a lower position. Further, if the
represented participant is at the same level as the interactive participant, then the
relation is one of equality, or of neutral power. As in the horizontal angle and
degrees of involvement, the power relation is a continuum, or a matter of degrees
of power (op.cit:146-148).

In many visuals there is no immediate, obvious point of view or social relation

from the interactive viewer's perspective to be ascertained, but a couple of likely
possibilities. In The Economist magazine there are many examples of a medium-
long to long shot of a scene taken from a relatively high vertical angle and an
oblique horizontal angle which do not suggest a clear role for the viewer. In these
visuals the vertical angle may suggest a position of power over the represented
participants, but the oblique horizontal angle may mediate that effect by
suggesting a lack of involvement with the participants depicted, or an exclusion of
that situation. Potential social roles for the viewer could cover a range of
possibilities, ranging from some position of power to that of an observer or

82
reporter who observes the scene from a superior, yet dispassionate stance
(op.cit:148).
In the older forms of objective visuals which Kress and van Leeuwen discuss,
there is no built-in point of view but a desire to present all the information
necessary to the viewer, even at the expense of deforming the laws of naturalistic
depiction (op.cit:136). The more modern examples of objective visuals include
scientific and technically-based visuals, as in technical graphics and drawings,
maps, and charts. Many of these encode an objective attitude through the use of
either a directly frontal, or perpendicular top-down angle, which “do suggest
viewer positions, but special and privileged ones, which neutralise the distortions
that usually come with perspective, because they neutralise perspective itself”
(op.cit:149). As can be seen in Figure 4.3, a cube viewed from the directly frontal
angle looks simply like a square, and the same cube viewed from a perpendicular
top-down angle looks the same.

(1) From the front. (2) From above. (3) From frontal-isometric
perspective
Figure 4.3 Cube views. (Montague 1994:33,35; Kress and van Leeuwen 1996:149,152)

However, viewing this same cube from a perspectival point of view, the sides and
top can be seen, and because of perspective there will be some degree of
distortion, the level of which will depend on the encoded angle of the viewer
(loc.cit.). The graphs, charts, and tables used in The Economist magazine are all
presented from the directly frontal angle, and the occasional maps used from a
perpendicular top-down angle.

These directly frontal and perpendicular top-down angles do not project their
sense of objectivity in the same manner, however. The directly frontal angle is
“the angle of maximum involvement” which in many other contexts is “the angle
of ‘this is how it works’ ‘this is how you use it’, ‘this is how you do it’” (loc.cit.).

83
It is an angle commonly used in schematic drawings for example, which are
concerned with showing, instructing and informing its viewers about situations,
processes and actions (as in a picture showing the parts of a video deck, or
instructions on how to make bread for a bread-making machine). In the case of
The Economist magazine, it is also the angle used to say ‘this is how it is’, or
‘these are the facts’.

4.2.2.4 Modality
The final aspect that Kress and van Leeuwen examine in their discussion of
interactive meanings in images is that of visual modality. Drawing once again on a
linguistic interpretation, in which modality is defined as the resource used by
speakers or writers to express judgements about the truth or credibility of
propositions in verbal communication, they assert that various forms of visual
communication also utilise a continuum of techniques to express modality. Thus,
“visuals can represent people, places and things as though they are real, as though
they actually exist in this way, or as though they do not — as though they are
imaginings, fantasies, caricatures etc.” (op.cit:161). An important point too is that
like the linguistic, visual modality is considered to be an aspect of interactional
and not ideational meanings, because what one group or culture considers credible
may not be judged the same way by another group or culture; and so, “modality
judgements are social, dependent on what is considered real (or true, sacred) in the
social group for which the representation is primarily intended” (loc.cit.). This can
be seen in the differences between the ways that photographs as opposed to
diagrams are considered as ‘real’ by different audiences. This difference is an
indication of the fact that “reality is in the eye of the beholder; or rather, what is
regarded as real depends on how reality is defined by a particular social group”
(op.cit:163).

To clarify this point, Kress and van Leeuwen consider the differences between
naturalistic and scientific realism. In the former, which Kress and van Leeuwen
claim is the dominant standard or criterion for what is considered ‘real’ in western
society and according to current technologies for representation and production,
reality is defined on the basis of how accurately what is portrayed approximates

84
what would be seen with the naked eye. Thus, reality is defined by the existing
technology provided by the capacity of 35mm photography (loc.cit.). In the latter,
in scientific realism, reality is defined based on the ‘generic’ nature or ‘regularity’
of the characteristics of represented objects. What can be simply observed with the
naked eye is the beginning point — in scientific realism there is an attempt to
represent things beyond what can be seen in reality, to represent objects according
to what they do in terms of processes (as in a Water Cycle diagram in a
Geography textbook or more abstractly, a line graph in The Economist), or
internal structures (as in an architectural schematic). Each of these realisms, the
naturalistic and the scientific, has its own views of what is real and what can be
considered as not real, but the dominant one is the natural: “although different
realisms exist side by side in our society, the dominant standard by which we
judge visual realism and hence visual modality, remains for the moment,
naturalism as conventionally understood, ‘photo-realism’” (loc.cit.).

Kress and van Leeuwen examine naturalistic modality in terms of a series of

modality marker continuums or scales (op.cit:165). All are relevant and applicable
to the interpretation of the kinds of visuals used in The Economist, and cover such
areas as:
1. the use of colour variation
2. contextualisation
3. representation
4. depth
5. illumination
6. brightness

The use of colour variation is an important method for expressing visual modality
in that the more that colour is reduced, the lower the modality. One continuum of
visual modality in colour is colour saturation, which relates to common standards
of photographic naturalism and runs from full colour saturation to a complete
absence of colour (as in black and white visuals where the only variation in the
colours is in terms of brightness) (op.cit:164-165). Another scale is colour
differentiation, which is a scale ranging from full colour differentiation in the
visual to what may be termed a 'reduced palette', and then to monochrome
(op.cit:165). There is also colour modulation, a continuum running from

85
modulated (using many different shades of a single colour) to a single plain,
unmodulated or flat colour. The principle for all scales is the same: the more the
colour is abstracted from naturalistic presentation, the lower is the modality.

Contextualisation refers to the degree to which a setting is presented in a visual

(loc.cit.). When a represented participant is shown with no setting, then it is in a
sense in a void, and is thus presented generically as a typical example of its type,
rather than as something with an individuality which is peculiar in a contextual
sense. In naturalistic images, the absence of context lowers the modality and there
is a continuum from the presentation of a full setting to no setting. Within this
continuum there are degrees of contextualisation ranging from a full contextual
representation to varying degrees of de-focused settings, or settings which are
under or over-exposed, thus reducing their clarity. Other variations may include
the use of a minimal number of setting variables to give the suggestion or
implication of a setting, with the rest of the frame totally blank. Further along,
there may be just irregular shading or patterning, or a regular pattern of light
shapes for example. At the extreme end is the absence of setting; there is totally
unmodulated colour, a black background, and most commonly, a blank or white
background.

Representation relates to the effect of differences or variations in detail between

the foreground and background in a visual, and can be interpreted as a scale
running from maximum abstraction to maximum representation of pictorial detail
(op.cit:166). A visual may show the finest details of the represented participants,
or it may show various levels of abstraction away from this detail. The effect of
reduced representation, or a reduction in detail in the background or setting may
lead to a decrease in the modality of the setting, and concomitant increases in the
modality of the foreground. For example, some of the finest details of the
represented participants may be shown, such as wood grain, skin texture, hair
strands etc., while the background may have lesser degrees of detail, leading to a
variation in the modality between background and foreground.

86
Depth, illumination and brightness are all important modality markers for
naturalistic images, especially in examples from portrayed art and in photography
(though not especially for The Economist magazine visuals). With regard to the
use of depth, the highest modality accrues to the use of central perspective,
(frontal angle only which displays the front of the visual in an attempt to represent
optical reality and engage the viewer in seeing a realistic image of physical space).
Degrees of lesser modality derive from this, ranging from the angular-isometric
perspective, to the frontal isometric perspective, and finally to the depth created
by simple overlapping (see Arnheim 1974:262-285; Saint-Martin:128-144; Kress
and van Leeuwen 1996:152-153).

Variations in illumination project different meanings depending on the techniques

used. In naturalistic images, participants are represented in relation to the sources
of illumination, with highlighting used to draw the viewers’ attention to particular
aspects or participants within the visual frame, while in some less naturalistic
images there may be abstractions from illumination, with shadows or shading
being used to allow just enough to convey the volume or shapes of objects (Kress
and van Leeuwen 1996:167). Other uses of shadowing or shading may be to
indicate areas that recede into the distance, or the use of highlighting to emphasise
prominent areas. Within this shading technique there are also degrees of
illumination, ranging from a full or modulated darkening of the shadowed areas to
the use of dotting or hatching of the shading, to light and shade being completely
abstracted away to the use of only lines rather than shading to indicate the
receding contours of a visual feature.

The scale of brightness values which may be used in a visual ranges from a large
number of different degrees of brightness to only two, that of black and white, two
shades of grey, or two brightness values of the same colour (loc.cit.). There is also
a variation in the ways that brightness can contrast with other scales of brightness,
as in an extreme contrast between the darkest and lightest areas of a visual, or
where there is a slight variation in brightness values, giving a hazy or misted
effect.

87
From the above discussion of modality markers in naturalistic visuals, it follows
that the determination of visual modality is "realised by a complex interplay of
visual cues. The same image may be 'abstract' in one or several dimensions and
'naturalistic' in others, …… yet from this diversity of cues an overall assessment
of modality is derived by the viewer" (op.cit:167-168). Also, naturalistic visuals
are, according to Kress and van Leeuwen, the dominant standard for visual
modality “in our culture”, and are a “yardstick for what is perceived as ‘real’ in
images, even when they are not photographs”, a point which is relevant for the
analysis of Economist magazine sketches (op.cit:168). However there are
instances where this dominant standard for what is ‘real’ does not apply in every
context — there is another continuum of visual modality however which relates to
the use of visuals in scientific/technological contexts. What is 'real' in these
contexts may often be very different to what is considered to be real in terms of
photographic/artistic naturalism. Reality in this context is calculated according to
the tenets of science, which generally deals with concepts of number, weight, and
measurement, as in blueprints or schematic diagrams showing how a machine may
operate, or graphics which display quantitative data and relationships between
phenomena. There are two other important contexts, besides the naturalistic and
scientific/technological which Kress and van Leeuwen only briefly include in their
work. There are firstly instances where a photographic visual attempts to appeal to
sensory qualities in attempt to generate in the viewer some appreciation of the
reality of such sensory qualities as texture, colour and 'feel', and there are the
contexts of abstract art where there is usage of abstract realism in visuals.

These four visual contexts and their relative modalities are defined as coding
orientations, or “reality principles” by Kress and van Leeuwen (op.cit:170). These
concepts and the terminology used are referenced to the work of Jurgen Habermas
(1984) and Bourdieu (1986), as well as to Bernstein (1981), from whom the term
‘coding orientation’ has been borrowed. Coding Orientations are defined as "sets
of abstract principles which inform the way in which texts are coded by specific
social groups, or within specific, institutional contexts" (Kress and van Leeuwen
1996:170).

88
These four Coding Orientations are summarised below.

89
Visual 1: Unemployment - Little shop of horrors plant
(The Economist, April 25th 1992, p57).
(reproduced)

90
1. Scientific/technological coding orientation - the dominant principle here is
the efficiency of the image as an outline or blueprint for explanation,
description etc. The use of colour generally means a lower modality, unless it is
useful in explaining the image’s features more efficiently.

2. Sensory coding orientation - the dominant principle here is the provision of

sensory pleasure in such contexts as visuals for art, fashion, cooking etc. The
use of colour attracts high modality, and is related to the ways colours can be
used to influence the viewers’ emotions (as in ‘hot pink’, ‘soothing pastels’ and
‘peaceful blues’).

3. Abstract coding orientation - the dominant principle here is the ability to

visually reduce the individual to the general, the concrete to its essential basic
qualities. This is the coding used by 'educated sociocultural elites', in areas
such as ‘high art’ and in some scientific and academic contexts etc. The ability
to use this orientation is the mark of social distinction, of being an 'educated
person' or a 'serious artist'.

4. Naturalistic coding orientation - the dominant coding in society, the one

which all members of the society share because they are being addressed as
fellow members, no matter what level of education they have received, or their
social status.

In terms of the application of these coding orientations to The Economist

magazine, the Naturalistic and the Scientific/technological coding orientations are
the most applicable and therefore useful for the interpretation of this magazine’s
multimodal text. Also, as the brief analysis below shows, many of the interactive
meaning realisations in images which Kress and van Leeuwen identify are
applicable to The Economist magazine visuals.

As already shown in the section of representational meanings, visuals are

multidimensional in that they simultaneously utilise a range of different visual
processes in order to address their viewers. Visual 1 (reproduced once more here)
is again a good illustration of this kind of complexity in interactional terms. An
analysis of this visual reveals that it is firstly an offer of information which the
viewers can either accept or refuse. This interpretation is supported by the fact that
the overhanging plant and the human participants are directing their gaze at the
other participants in the visual frame, the little shoots (and not the viewers). This
is therefore a depiction or a scene for the viewers to simply observe, a scene
which can be interpreted as being a medium to long shot in terms of the size of

91
frame, which requires viewers to look at it from a frontal angle and at the same
level position. This therefore requires of the viewers an attitude of little
involvement with, and relative detachment from, the represented participants. This
visual can also be interpreted in modality terms regarding its acceptability as a
believable exemplar of its type as a naturalistic visual. This can be carried out in
accordance with the various modality marker continuums which Kress and van
Leeuwen identify. Thus, the fact that this hand sketch has no colour, but is a
monochrome visual which approximates real people and combines them with
fictional characters, is supported by the relative absence of any background, by the
abstraction from reality afforded by the use of sketching and caricaturisation, and
the relative absence of illumination and brightness to highlight the scene and the
action. This suggests that this visual obviously has less modality than either a
colour or monochrome photograph in the naturalistic coding orientation from
which it has been derived.

The applicability of these interactional realisations in Kress and van Leeuwen’s

work to visuals from the scientific/technological coding orientation in The
Economist magazine is not so obvious however. This is because in their discussion
of interactive meanings in visuals, Kress and van Leeuwen concentrate heavily on
naturalistic visuals, almost to the exclusion of visuals from other coding
orientations. Indeed, all of their illustrative examples in this section of their book
are naturalistic. It would seem, however, that the only applicable interactive
realisations for these scientific/technological visuals would be in terms of the
presence or absence of Colour (the use of colour in these would mean less
modality), in terms of Attitude (as in a top-down angle where the viewer is
looking down at a flat surface or a page which encodes a knowledge orientation),
and in terms of the degrees of Representation of detail (where the delicacy of
analysis or complexity of the data shown would affect the acceptability of a graph
as a believable instance).

4.2.3 Compositional Meanings

92
Up to this point this review has discussed and evaluated Kress and van Leeuwen’s
description of the ways that various kinds of visuals organise and represent their
meanings representationally (ideational), and the ways that visuals attempt to
address their potential viewers in interactional (interpersonal) terms. These have
been examined from the point of view of features which would be most useful for
the analysis and interpretation of The Economist magazine visuals. This section
now deals with a discussion of the relevant compositional meanings (textual) in
their visual grammar. In this area, Kress and van Leeuwen analyse the ways
visuals are composed or structured, and the kinds of meanings conveyed by the
various possible compositions. These include single mode forms, as in the
composition of just a single visual, and multiple mode forms, as in the
composition of a text comprising one or more visuals and accompanying verbal
text. It is their treatment of this latter composition type which is specifically of
relevance to the focus of this study on The Economist magazine text, since the
single mode form is really only used on the magazine’s front page. Kress and van
Leeuwen refer to multiple mode forms as multimodal texts, in that the meanings
embodied within are realised via differing semiotic codes (Kress and van Leeuwen
1996:183).

The compositional features in multimodal texts (in Reading Images the term
composite texts was used, and Kress and van Leeuwen now seem to use both
terms interchangeably) are related to the principles of layout, or the “way in which
the representational and interactive elements are made to relate to each other, the
way they are integrated into a meaningful whole” (op.cit:181). Kress and van
Leeuwen raise the question of how the products of the various codes operating in
one text should be examined; whether the meaning of the whole should be viewed
as the sum of the meanings of its parts, or whether they should be viewed in an
interactive way, such that the textual parts should be viewed as interacting with
and affecting each other (op.cit:183). Kress adopts this latter view, as does this
study. Thus a single image, or a whole page which contains both image(s) and
verbal text (or even sometimes a series of pages in a magazine spread), is
therefore treated as an integrated text, which is the result of the "work of an
overarching code whose rules and meanings provide the composite text with the

93
logic of its integration” (loc.cit.). Kress and van Leeuwen recognise that there are
two such integration codes: that of Layout, or the code of spatial composition, and
that of Rhythm, or the code of temporal composition. Layout is concerned with
texts whose elements are spatially co-extant but in a sense time frozen, as in print-
mode magazines such as The Economist, and in maps and art works, while Rhythm
is concerned with texts which develop over time, as in speech, music, and the
dance etc. (loc.cit.). As this study is concerned with analyzing a multimodal text
from The Economist magazine, the discussion which follows will be concerned
only with the relevant aspects of the Layout integration code in visual
composition. In both a single mode (one image) and a multimodal text (image plus
verbal), this code is essentially concerned with the relative placement of the
represented participants, their placement in terms of top and bottom, left and right,
how much to the top or bottom, or how much to the left or right, and in terms of
perspective, or how much is foregrounded and how much is in the background.
Arnheim (1988) claims that there is a strong physiological element to the
interpretation of this layout code, in that it is based on the viewer's kinaesthetic
sense of balance and weight. Taking this notion one step further in terms of
Halliday’s SFL model, Kress and van Leeuwen suggest that the “fundamental
function of integration codes such as composition is textual. Integration codes
serve to produce text, to place the meaningful elements into the whole, and to
provide coherence and ordering among them” (1996:212).

Again, Kress and van Leeuwen draw heavily on the work of Arnheim (1974,
1988) on pictorial design and visual (gestalt) perception, Dondis (1973) on the
fundamentals of visual literacy, and indirectly on Uspensky (1973) on the poetics
of composition in classical art, for describing the kinds of meanings that are
compositionally organised in visuals. In Reading Images (1990:95-98) they
identify six important structuring principles in layout. These are stated as salience,
balance, vectors, reading paths, framing and perspective. They also discuss the
importance of the vertical and horizontal compositional axes, which are
pinpointed as two important basic organising principles in the visuals produced in
Western cultures (Arnheim 1988; Kress and van Leeuwen 1990:95). In The
Grammar of Visual

94
COMPOSITION SYSTEM GENERAL FEATURES

The placement of the elements (participants and syntagms

that relate them to each other and to the viewer) endows
Information Value them with specific informational values attached to the
various ‘zones’ of the image: left and right, top and
bottom, centre and margin.

The elements (participants and representational and

interactive syntagms) are made to attract the viewer’s
Salience attention to different degrees, as realised by such factors as
placement in the foreground or background, relative size,
contrasts in tonal value (or colour), differences in
sharpness, etc.

The presence or absence of framing devices (realised by

elements which create dividing lines, or by actual frame
Framing lines) disconnects or connects elements of the image,
signifying that they belong or do not belong together in
some sense.

Table 4.4 Interrelated systems of compositional structuring principles (loc.cit.)

95
Design (1996) however, these structuring principles are conflated into three
interrelated systems which relate the “representational and interactive meanings of
[a] picture to each other” (1996:183). These structuring principles are reproduced
in Table 4.4, and will be examined in terms of their applicability not only to single
mode visuals, but also to the multimodal texts which occur in The Economist
magazine (texts which include both images and verbal text, be it within a visual
frame, or across a whole page or series of pages). Each of these compositional
systems will now be discussed in turn.

4.2.3.1 Information Value

The horizontal axis in a visual has been referred to previously with reference to
degrees of involvement, but it is important also because it generates a left-right
distinction in the structural meanings within multimodal compositions that can
have an influence on the information value accorded to the various visual
elements. Kress and van Leeuwen examine the continuous movement between left
and right in horizontal layout structure
in a number of multimodal magazine articles, and based on their examination
assert that generally
when pictures or layouts make significant use of the horizontal
axis, positioning some of their elements left, and other, different
ones right of the centre (which does not, of course, happen in every
composition) the elements placed on the left are presented as
Given, the elements placed on the right as New (op.cit:187).

The meaning of this horizontally-based left-right structure is that the left

approximates the well-established, known, understood, implicitly held view (or
the Given), and the right approximates that which is contestable, to-be-established,
presented as not yet known, to be agreed upon, or to be made explicit (or the New)
(loc.cit.). In stating so, Kress and van Leeuwen draw upon Arnheim's (1988:47)
discussion of “the tendency, largely unrelated to actual eye movements, for
viewers to perceive pictures as organised from left to right, so that the lower-left
corner appears to be the composition’s point of departure [my emphasis]”. They
also establish a direct link to Halliday's (1994) discussion of the structure of the
information unit, which consists basically of two functions, the Given and the
New. The Given is referred to as that which is recoverable, what is already known,

96
or what the speaker/reader is able to access. It may be something which has been
mentioned or shown before, or something in the context of situation. As Halliday
puts it, "the meaning is: this is not news" (1994:298). The New is that which is not
recoverable in the sense that it has not been mentioned before, or it is unexpected.
According to Halliday, "the meaning is: attend to this; this is news" (1994:298).
Thus, Kress and van Leeuwen assert that there is a deal of similarity between the
sequential nature of information in verbal language and the horizontal structuring
of visual layout.

Kress and van Leeuwen also claim that in many multimodal magazine layouts the
left-hand space of the Given is usually taken up by verbal text, while that of the
New (right-hand) is often taken up by one or more images, with the images
providing a way of quickly and directly letting the viewers know what is the new
information coming. Kress and van Leeuwen comment however that there are
exceptions to this verbal-visual, Given-New ordering, such as instances where the
image is presented as the Given and the New is the verbal text. They discuss
specific instances to show that the realisations of this Given-New distinction are
not always predictable, as in an advertisement with an image of a Mercedes car
placed on the left-hand side as the Given (an assumption that viewers see a
Mercedes as a socially accepted symbol of prestige and wealth) and the
description of its features in verbal text on the right-hand side as the New (Kress
and van Leeuwen 1996:189).

These Given-New principles also apply to single naturalistic images, within the
visual frame, where the represented participants are sometimes vectorially
arranged and related in such a way that the left hand participant is recognisable as
a Given and the right hand as a New piece of information (op.cit:190). Conceptual
diagrams can also be organised according to these Given-New principles. This
may be where information is spatially arranged and joined by arrows to show a
process of movement from left to right (as in a model of communication between
company offices, although these can go in either direction), and more definitely to
graphs where information is commonly arranged according to chronology, with
the earlier (Given or known) information placed on the left, and the most recent or

97
current information (the New) placed on the right hand side. This is an aspect of
reading path which will be further developed below (also see Arnheim 1988:47 on
the constraining influence of the mechanism of human vision). This Given-New
mechanism can also be seen in various line graphs which include an element of
prediction of trends based on the behaviour of data in previous time periods. In
those graphs which have that predictive element, the Given is the past figures (for
example, the unemployment rate from 1945 to the article date), while the New is
the predicted movement of the same data beyond the article date. This kind of
predictive line graph is very rare in The Economist magazine — those commonly
presented almost invariably reproduce the relevant data up until the date of the
particular issue in which it appears (see Visual 2), and do not attempt to depict the
ways that the data could behave in the future (although they do attempt this is
various verbal ways).

The vertical axes in visuals are another aspect of information value, which is of
importance in the ways that they generate a top-bottom distinction in the structural
meanings within visual compositions. There is often a difference in modality
between the tops and bottoms of visuals that is conveyed by oppositions between
the visual’s two halves. This verticalised difference can be realised by a variety of
methods, such as framing, differences in focus, colour saturation, salience, blank
spaces etc. Drawing heavily again on Arnheim’s (1974:182-187, 1988:109-148)
work on bipolar composition and the relative visual weightings in art work
produced by the horizontal and vertical axes, Kress and van Leeuwen hypothesise
that in Western cultures (other cultures may utilise a differing structuring), a
common compositional structuring involves the top of a visual occupying a
special place, which is the space of the 'ideal' or most highly valued, with the
bottom of a visual occupying the place accorded to the 'real' or less highly valued
(1996:193). This Ideal-Real structure, (in the same way as the Given-New
structure) also applies to both the composition of single images as well as
composite layouts involving image(s) and verbal text. Kress and van Leeuwen
illustrate this with examples drawn from advertisements in magazines and
illustrations from science high school textbooks. Their rendering of the meanings
of structuring along the vertical axis is summarised in Table 4.5.

98
 TOP
space of the ideal
most highly valued
salient

In advertising: In science:
images the abstract
promise the general/generic

BOTTOM
space of the real, the 'here and now'
less highly valued
less salient

In advertising: In science:
language (verbal) the empirical
product the specific

Table 4.5 Verticalised Ideal-Real meaning in multimodal texts

It is also important to note that often there are often bridging forces between two
opposing halves of visuals, forces which attempt to connect the two opposing
meanings in a kind of visual syntagm or unit of meaning. These can function to
form a bridge between the real and ideal, to lessen or modulate the division — this
is exemplified by Kress and van Leeuwen via the analysis of an advertisement in
which the verticalised difference between an image of a woman in a relaxing bath
(top) and the stark verbal text (bottom) is minimised by the vectors formed by her
gestures, which point to the verbal text extolling the virtues of the herbal bath
product (loc.cit.). Although visually there is a strong division between the image
and the verbal text, this division is lessened so that the whole composition can be
viewed as a single visual syntagm. In many magazines like The Economist this
formation of a visual syntagm is often achieved through the use of a verbal run-
around, where the verbal text is forced to shape itself around the visual or one of
its represented participants, or where the visual in a sense invades the space of the
verbal.

If Kress and van Leeuwen’s hypothesis is accepted, that in Western cultures a

common compositional structuring is that the top of a visual is the space of the
'ideal' or most highly valued, and the bottom is that of the 'real' or less highly
valued, then some relevant comments can be made about observable patterns in

99
The Economist magazine with regard to the placement of visuals vis à vis verbal
text. A glance at any issue will show that in the various departments (Asia,
American Survey, International, Europe, Britain, Business, Finance, and Science
& Technology) there is a strong tendency to place visuals in the top left hand
corners of each leader article. This is especially so of the Finance Department,
which is the source of the text analysed in this study. Accepting Kress and van
Leeuwen’s hypothesis also means that the placement of a visual in the top half of
a leader page would suggest that the graphic designers at The Economist magazine
are presenting their most highly valued or salient meanings in that place. The fact
that this is often done through sketches which include a strong element of
caricaturisation or wry humour is important and will be discussed more fully in
Chapter Six, which presents a visual typology for The Economist magazine and
the results of an interview with the Chief Editor of Graphic Design at The
Economist magazine headquarters in London.

A third aspect of information value of note here is the importance of the central
spaces and outer edges in images and multimodal texts. This is an aspect of visual
composition which is pivotal to Arnheim’s gestalt-based theory of composition, in
which he refers to two opposing forces in composition, that of centricity and
eccentricity. These “differentiate between compositional forces related to an
internal centre and others acting in response to an external centre” (1988:viii).
Kress and van Leeuwen adopt the terms Centre and Margin in order to explain the
ways that some images and composite (multimodal) texts place elements in the
centre of a layout, in order to signal to the viewer that
something to be presented as Centre means that it is represented as
the nucleus of the information on which all the other elements are
in some sense subservient. The Margins are these ancillary,
dependent elements. In many cases the Margins are identical or at
least very similar to each other, so that there is no sense of a
division between Given and New and/or Ideal and Real elements
among them (1990:206).

As mentioned above, the compositional structuring principles are conflated into

three interrelated systems of information value, salience and framing. Thus, the
Given/New and Ideal/Real can combine with the principle of Centre-Margin to

100
provide a composite layout which signals its primary messages through the
relative salience of its elements or through framing devices. Kress and van
Leeuwen also make the point that some of these visual compositional structures
have no real counterpart in language. They suggest that
while language has a specific grammatical form for realising the
Given-New structure, this is not the case with the Ideal-Real and
Centre-Margin structures. This is not to say that the meanings these
structures express cannot, in some form, be expressed in language,
but rather, that they are more readily and frequently expressed
visually, and that language, unlike visual communication, has not
developed ‘grammatical’ forms to express them. (1996:211)

The dimensions of visual space, incorporating all three aspects of the horizontal
and vertical axes, as well as the central position, are reproduced in Figure 4.4.

Margin
Margin
Ideal
Ideal
Given
New

Centre

Margin
Margin
Real
Real
Given
New

Figure 4.4 The dimensions of

physical space (op.cit:208)

Examples of the principle of Centre-Margin operating in layouts in The Economist

magazine are quite common in departmental leader articles, where the centrepiece
of an article is either a photograph or a sketch, and the verbal text is symmetrically
arranged around it. This is suggestive of the kind of semantic connection which, it
is argued in this study, is one of the mechanisms by which intersemiotic
complementarity is realised.

101
SALIENCE INDICATOR FEATURES
Size larger objects are more easily noticed by the eye that smaller
ones.

Sharpness of focus objects are more clearly seen because their features are in
sharp focus and are more easily noticed by the eye than those
which have their features less sharply focused.

Tonal contrasts areas of high contrast, for example black borders placed on
white spaces are higher in salience than a grey-shaded, less
distinct border performing the same dividing function.

Colour contrasts the contrasts between highly saturated colours and softer
muted colours, or the contrast between red, white and blue.

Placement in the visual field the aspect of visual ‘weight’ - objects are ‘heavier’ when
close to the top, and ‘heavier’ when placed on the left.

Perspective objects or entities placed in the foreground are visually more

salient than those in the background, and elements which
overlap others are more salient.

Table 4.6 Visual ‘clues’ for Compositional Salience (op.cit:212)

102
4.2.3.2 Salience
Besides the relative placement of elements on the page and the information value
accorded to them, there is also the relative importance bestowed on them by the
different degrees of Salience. Salience refers to the ability of a viewer to make
judgements about the importance of various elements in a visual in relation to the
other elements, and is related to the viewer's ability to judge the 'visual weight' of
these various elements. One of the important functions of the compositional
integration code is to convey to the viewer the relative significance of visual
elements via an hierarchy which induces the viewer to focus attention on some
elements more readily than on others. Arnheim (1988:15-21) discusses an aspect
of this in regard to the appearance of ‘heaviness’ in objects placed in the top of a
layout as opposed to the bottom (the higher objects appear ‘heavier’), and on the
‘weightiness’ of the left-hand side of a visual frame — the impression that objects
which are placed in the left-hand side of a visual frame as opposed to right
placement are heavier is due to an “asymmetry of the visual field” (op.cit:47).
This has already been discussed in relation to the horizontal axis and Given/New
information structure in composition. Thus, the greater the weight of an element in
relation to others, the greater its salience. According to Kress and van Leeuwen,
salience is not an objectively measurable quality, but is the result of a complex
interplay between various visual elements which act as ‘clues’ to let the viewer
know what is important, and what is more important than other elements
(1996:212). These ‘clues’ are summarised as Salience Indicators in Table 4.6. All
except colour contrasts are relevant to the analysis of Economist magazine
visuals, although conceivably this indicator could be applied to the contrast
between black and white in monochrome images.

Another aspect of the ways that viewers are required to judge the visual ‘weight’
of the elements of a composition covered by Kress and van Leeuwen, is what
Arnheim refers to as the perception of balance or stability created by the
balancing centre (1988:66,109-119). This relates to the ability of the viewer to
make perceptual judgements about the salience (or relative weights) of various
visual elements in combination, so that there is a sense of balance amongst the
elements emanating from a central point or core. This means that within some

103
visuals there is a point which could be deemed to be the centre, and it is often the
area which contains or projects the central message. It may be in the actual centre
or off-centre in the visual, but its salience and 'weight' projects its importance, and
it can also have an effect on the areas in the visual which surround it (Kress and
van Leeuwen 1996:213).

4.2.3.3 Framing
Framing is another aspect of composition which works in combination with
information value and the different degrees of salience organised within visuals.
Framing refers to the degrees of connectedness or boundedness provided by an
actual frame or frame lines around a whole visual, or around the elements in a
visual. Within a single visual or multimodal (composite) text the elements or
groups of elements presented can be connected or joined together by various
graphic techniques, giving a strong sense that they ‘go together’, that they should
be viewed as part of the same message; or they can be marked off or disconnected
from each other, conveying the sense that they should be viewed (read) separately
(op.cit:214). The ways that this separation can actually be interpreted depends on
the context. For example, framed double pictures of two people can convey the
sense that the subjects should be read as being ‘together’ in some way, the nature
of that connection being revealed by a reading of the verbal text accompanying the
framed images. If the two photos are clearly marked off from the surrounding
verbal text, this may suggest that there is some kind of semantic separation — and
the nature of the connection and the separation will largely be provided by the
verbal context. The context thus “colours in the more precise nature” of these
interpretations (op.cit:215).

Framing can also be realised by elements within a visual which convey the sense
of boundedness around, or separation from other elements through their actual
shape or volume (as in a column in a building or in a scene which may divide up
the elements portrayed into sectors). This may also be realised by other methods,
such as the use of discontinuities of colour hue or saturation, of variations in
visual shape, or simply by the use of empty space (op.cit:216). In magazines the
degree of separateness of visuals from the verbal aspect of the text may be either

104
weak or strong, with clear framing being a suggestion of separateness or
distinctness, and weak framing suggesting that both elements should be read
together or as being close in some way (as in subject, or point of view). In The
Economist magazine both clear framing and various degrees of weak framing are
used. Instances of strong framing usually occur with graphs, tables and charts,
where a distinct border has been drawn to clearly demarcate the two modes.
Slightly lesser demarcation through framing occurs with many photographs, which
are marked off from the surrounding text by lines drawn around and on the edges
of the photograph/image frame. The use of framing in The Economist magazine is
realised differently however in the case of sketch visuals where there is often no
clearly drawn line to demarcate the verbal from the visual mode, but the use of
white space to frame the visuals. This kind of choice of layout technique and
visual representation implies strongly, in layout terms, the projection of an
intersemiotic syntagm, which this study hypothesises as being a realisation of
intersemiotic complementarity.

Vectors also can convey this sense of connectedness between elements in a

composition. Besides being lines which can be drawn from or form a part of the
represented elements in a visual, as realised by the direction of gaze, arms, and the
structural features of objects, vectors may also be formed by the arrangement of
abstract graphic elements, where the eye is guided away from the most to the next
salient element (in terms of size or colour saturation etc.). Vectors can also be
formed by differences in size or volume in co-occurring images, and the angles
created by skewed elements, and headlines etc. In some graphic visuals, the
structural similarity between the graphs can convey the sense that they belong
together (the primary source of that sense may be the strong framing around them,
but their structural similarity also acts as reinforcement). The similarities can
include the fact that the graphs are all framed in the same way, they have a
rectangular size and shape, they all have the 3-D ‘shadowing’ effect of being
slightly raised from the paper on which they are placed (a common technique in
The Economist), and they all have the same white-on-black contrasted headings.
4.2.3.4 The Importance of Reading Paths

105
An aspect which operates through all the three compositional structuring
principles of Information value, Salience and Framing discussed by Kress and van
Leeuwen (op.cit:218-223) is that of Reading paths, and the features of what they
term as linear and non-linear compositions, a term which has already been
introduced in Chapter Two in reference to Tadros’ work (1985:25). The former
refers to the strictly coded verbal text of English, which in its single-mode form
must be read from left to right (e.g. a novel), and the latter to multimodal
(composite) texts which can be approached in alternative non-linear (circular,
diagonal, spiralling) ways. Reading paths relate to the hypothetical viewer's eye
movement from the most salient points in the composition to the next or less
salient points. There is of course a cultural element to reading paths in that
viewers from different cultures (and even different age groups or educational
backgrounds) may read a visual in differing ways. There is also a need to
recognise that the reading path may not follow the same path as that followed in
reading verbal text. In other words, left to right and top to bottom — it may in fact
be the reverse or move from the centre outwards. Further, there is also no
implication that the reading path encoded by the producers of the text is the same
followed by the viewer. If what is made salient in a visual is culturally
determined, then members of different cultural groupings are likely to have
different views of what is or is not salient (Kress and van Leeuwen 1996:218-
213).

Kress and van Leeuwen also comment about the reading paths commonly used in
magazines and newspapers, comments that are relevant to the interpretation of The
Economist magazine text in this study. They suggest that the reading paths
followed in magazines often involve the readers flicking through and stopping as
pictures or headlines catch their eye, and then perhaps returning to the articles
which piqued their interest. Alternatively, they may go straight to the article
relating to the front page topic or headline, or to their favourite columnist, or the
sports section. Many readers of newspapers may in fact read the back page first.
Thus the reading path is selective and partial, as opposed to being strictly linear.
They comment further that

106
 “whether the reader only ‘reads’ the figure and the headline
[referring to a magazine text they are discussing], or also part of the
verbal text, a complementarity, a to-and-fro between the text and
image, is guaranteed …… the most plausible reading path is the
one in which readers begin by glancing at the photo, and then make
a new start from left to right, from headline to photo, after which,
optionally, they move to the body of the verbal text. Such pages
can be scanned or read, just as pictures can be taken in at a glance
or scrutinised in their every detail (op.cit:218-219).

As the report in Chapter Six on the interview with the Chief Editor of Graphic
Design at The Economist magazine’s corporate headquarters in London will show,
this is precisely what the graphic designers of each issue of The Economist
magazine assume that their readers will do, and so they attempt to construct their
issues accordingly. Kress and van Leeuwen’s use of the term ‘complementarity’ in
the above quote is interesting, and it is in fact the first time that they use it. Kress
and van Leeuwen’s usage of the term is in a different sense to that of this study
however, in that their usage refers particularly to the fact that a reader of a
multimodal text generally tends to read all elements of a multimodal text in some
culturally-determined way or direction. Complementarity in this study however, is
a theoretically-motivated term which is used to explain not only why a reader does
move “to-and-fro between the text and image”, but also what intersemiotic
semantic resources are used to produce this sense in the viewer that the text before
him or her is a single, coherent multimodal text.

4.3 The Language of Displayed Art

The set of publications by O’Toole (1994, 1995) also applies an SFL perspective
to the language of displayed art, and as such is also a significant contribution to a
functional interpretation of visual communication. O’Toole’s thesis is simply
stated:
Michael Halliday's Systemic-Functional linguistics offers a
powerful and flexible model for the study of other semiotic codes
besides natural language, and its universality may be of particular
value in evolving discourses about art (1995:19).

Like Kress and van Leeuwen, O’Toole approaches the description and analysis of
displayed art from a metafunctional standpoint. Similarly, he reinterprets the

107
metafunctions to be ‘Representational’ for ideational, ‘Modal’ for interpersonal,
and ‘Compositional’ for textual, to provide interesting analyses of the ways that
the visual modes of sculpture, architecture, and especially classical art project
their meanings. Where O'Toole's theoretical focus significantly differs from Kress
and van Leeuwen's however, is in his clearer stress on the key SFL notions of
REALISATION and RANK SCALE in the interpretation of displayed art
(Halliday 1994:15). Realisation in the SFL model refers to a relationship between
levels of meaning. That is, socially-based meanings (such as those deriving from
the ways that people construct the world of objects and happenings, as well as
speaker/listener relationships) are recoded into the grammatical systems of
MOOD, TRANSITIVITY, THEME etc. These grammatical resources are realised
by various syntactic configurations, which are themselves realised by a range of
appropriate phonological or graphemic sequences. Expressed another way, the
phonological/graphemic systems realise the lexico-grammar, which in turn
realises the semantic systems which themselves realise culture (Halliday 1992:24-
25). The application of these fundamental principles to a typical Renaissance
period painting for example, would suggest that the oft-depicted figures of Christ
and his various disciples have the potential to realise certain Judaic-Christian
eschatological concepts for their viewers, in accordance with their specific
cultural backgrounds. These culturally-based meanings are realised however by
the presentation of these figures by means of
a particular manner of representation, with a particular modality of
address to the viewer and involving a complex network of
compositional relationships. These in turn are realised in particular
lines and planes on the painted surface involving chromatically
appropriate colours, rhythms, and degrees of illumination. The
scale of realisation from semiotic systems to graphological form
and substance in painting is analogous to that for language (O'Toole
1995:161).

O'Toole, like Kress and van Leeuwen in their 'grammar of images', asserts that
there is a 'grammar' of painting, and that in any attempt to analyse these types of
semiotic systems, there is a need "to isolate a hierarchy of comparable units of
structure" (loc.cit.). In the SFL model of language the hierarchy of comparable
units at the lexico-grammatical level which Halliday uses consists of the clause
complex, clause, group, word and morpheme (1994). In O'Toole's interpretation,

108
the semiotics of painting or art is viewed and interpreted in terms of the levels of
the Picture (the whole work), Episode (stages in the story or message portrayed),
Figure (animate beings, human or non-human), and Member (parts of the figures)
(1994:14-15, 1995:161). The full framework outlining these levels and the visual
systems which can potentially be drawn upon in any work is outlined in Table 4.7.
Although he also discusses sculpture and architecture as forms of displayed art,
O'Toole focuses mainly on painting to develop his framework, and since paintings
most closely approximate one of the kinds of visuals found in The Economist
magazine (Naturalistic), this review will assess this framework in terms of its
applicability to these kinds of visuals in the present study.

O'Toole claims that the framework he presents for the semiotic analysis of
displayed art is not designed to be a formalised constraint on the interpretation of
an artwork's meaning, but should be viewed as a 'map' which schematicises "the
semiotic space created by the work within which our perceptions and conceptions
are negotiated" (1995:165). He also proposes that there are some advantages to
initially approaching an analysis from the Modal rather than the Representational
functional dimension of meaning, despite the obvious pressure in both linguistic
analysis and visual analysis to start with the subject matter of the text, or the
topical nature of what is being visually represented. The advantages, according to
O'Toole, are threefold: firstly, the Modal systems most probably affect the nature
of the initial engagement with the work; secondly, an effective description would
counteract the common tendency towards a form/content dichotomy among art
critics; and thirdly, a modal-based semiotic analysis may provide students and
lovers of art with a language to describe what is actually seen, rather than what art
historians believe should be known as background knowledge (op.cit:166).

It can also be seen from the framework in Table 4.7 (following page) that many of
the systems which O'Toole identifies as resources for potential usage in art-based
semiotic systems are similar to those described in Kress and van Leeuwen's

109
UNIT \ FUNCTION REPRESENTATIONAL MODAL COMPOSITIONAL
Orientation to Reality and Style:
SCHOOL Typical themes e.g. Baroque - Expressionism - Constructivism - Surrealism - Cubism - Op Art
Pop Art - Installation and Performance.
Actions, events Focus: Frame, Weight Gestalt: Proportion
Agents-patients-goals Perspective Modality: Framing 'Theme'
Narratives, Scenes, Clarity, Light Fantasy Horizontals Line
Settings, Features, Colour, Scale Irony Verticals Rhythm
PICTURE Portrayals Volume Authenticity Diagonals Geometric forms
(WORK) Sitters Gaze: Symbolism Colour Cohesion
'Eyework' Omission
'Paths' Intertextuality
'Rhythms'
Intermediaries
Group and sub-actions, Scale to whole Relative position in Gestalt and to each other
Scenes, Portrayals Centrality to whole Alignment of forms
EPISODE Side sequences Relative prominence Interplay of forms
Interplay of actions Interplay of modalities Coherence of forms
Character Characterisation Relative position in Gestalt, in episode and to
Act / Stance / Gesture Relation to viewer each other
FIGURE Clothing components Gaze, Gesture Parallelism / Opposition
Object Contrast and Conflict: Subframing
Position Scale, Line, Light, Colour
Basic physical forms: Stylisation, Attenuation, Chiaroscuro, Synecdoche, Cohesion:
MEMBER Part of body, Objects Irony Reference, Parallelism, Contrast, Rhythm
Natural form, Components

Table 4.7 Functions and systems in painting (O'Toole 1994, 1995)

110
‘grammar of images' (such as Gaze, Relative prominence, Framing, Rhythm,
Colour, etc.). These kinds of visual systems are of course used by O'Toole in a
distinct way to express how meanings are realised in a specific visual mode, and
there are some additional systems included which obviously apply specifically to
displayed art (such as Stylisation, Attenuation, Chiaroscurio etc.). To explicate the
approach he is taking, O'Toole analyses some famous art works in depth  in the
earlier of his two papers he examines Botticelli's Primavera, and in the latter
Hinder's Flight into Egypt, the winner of the 1952 Blake Prize for Religious Art
in Sydney, Australia.

O'Toole's extensive analysis of the Primavera is an effective illustration of the

ways that the systemic visual options drawn upon and combined by an artist can
be realised in a single painting. One of the strengths of his analytical model,
which in some ways is lacking in Kress and van Leeuwen's attempt to describe a
general 'grammar of images', is that he recognises the need to situate and relate his
analysis to other generalised discourses about art and painting (op.cit:173). His
appreciation of this need is an attempt to utilise the Hallidayan concept of
Register, which is defined by specific values of Field (realised ideationally),
Tenor (realised interpersonally) and Mode (realised textually), and constitutes the
analytical tool which allows the semiotician to relate the social context (context of
situation) to the text (Halliday and Hasan 1985; Halliday 1994). O'Toole also
recognises that any given text (visual or verbal) is a realisation of the social
semiotic out of which it has grown, as well as constituting a contribution to that
social semiotic. A text therefore has the potential to consolidate that social
semiotic by being highly governed by the prevailing conventions and social rules
(be a reflection of it), or it may attempt to question, challenge, and destabilise the
social semiotic from which it is derived (1995:175). O'Toole applies these
concepts specifically to his analysis of the Hinder painting Flight into Egypt,
which he discusses in terms of various art critics' responses to the awarding of the
Blake Prize in Australia.

As already mentioned, many of the Representational, Modal and Compositional

systems which O'Toole refers to in his schematic framework correlate with those

111
used in Kress and van Leeuwen's 'grammar of images' model (e.g. the categories
in the representation of action, of objects and of scenes; the forms of address
realised by gaze, framing, colour, illumination and perspective; the categories in
composition such as framing, positioning in the visual space in terms of the
horizontal, vertical and diagonal axes, and other factors such as colour
coordination and the influence of size and framing on visual salience). What is
distinct however is the application of the Hallidayan concept of Rank Scale for
ascertaining the meanings of its various elements (although this is similar to their
treatment of embedding in images), as well as a stronger application than Kress
and van Leeuwen of the notion that any interpretation of a visual text needs to be
carried out in relation to the social semiotic from which it has arisen, and therefore
tempered with the understanding of why that text is as it is (the concept of
Register). It is this attempt by O’Toole to relate the interpretation of a visual art
work to aspects of its context of situation which is the most relevant aspect of his
framework for this study, and which will therefore be utilised for the
interpretation and analysis of the ways that The Economist magazine visuals
project their meanings.

4.4 Summary and Conclusions

This chapter has attempted to outline and review a set of researchers who have
used linguistically-derived insights from the general theory of language and
communication developed by M.A.K. Halliday (1978, 1985) to examine non-
linguistic modes. This review shows that there is a significant amount of work that
has been carried out by those working from a Hallidayan interpretation of
meaning-making in visual semiotic modes, and that it is concerned mainly with
two areas: various types of images such as photographs, drawings and diagrams,
and displayed art such as painting, sculpture and architecture.

What is new and interesting in Kress and van Leeuwen’s and O'Toole's work is
their application and adaptation of a different, linguistically-focussed paradigm to
examine the ways that visual information is organised and projected. Both sets of
work are characterised by a concentration on and adaptation of differing aspects of
the SFL model. Kress and van Leeuwen's application of Hallidayan SFL

112
principles to the visual mode specifically involves adapting the TRANSITIVITY,
MOOD, MODALITY, and THEME systems from the lexicogrammar to the ways
that the range of images they examine organise their meanings, while O'Toole's is
characterised by a focus on an application of these grammatical systems as well as
Rank Scale and Realisation (which Kress and van Leeuwen only briefly discuss
1996:44). O'Toole's work also more overtly attempts to relate the visual texts he
analyses in terms of their contexts of situation, an appreciation that Register is an
important consideration when any text (of any mode) is analysed. Kress and van
Leeuwen on the other hand examine a wide variety of images and image types to
illustrate how the visual meanings they are referring to are produced, but they do
this with minimal reference to the specific contexts in which the images they
analyse originally occurred. When they do support their interpretations with
reference to specific contexts of situation, it seems that their interpretations are
more powerful (see for example 1996:144-145).

What is also of immediate note from this review is that the metafunctional
terminology posited by Halliday to describe meanings at the semantic level in his
SFL model has been changed or adapted by both sets of authors to suit their
analyses and the modes they focus on. Kress and van Leeuwen begin by using
Halliday's metafunctional terminology, and then proceed to adapt it to what they
perceive as the appropriate terminology for visual sign-making (op.cit:40-41).
Thus, under ideational they refer to Representational (Narrative and Conceptual)
meanings, under interpersonal they refer to Interactive meanings, and under
textual to Composition(al) meanings. O'Toole on the other hand reinterprets the
metafunctions to be Representational for ideational, Modal for interpersonal, and
Compositional for textual. The specific reasons why these changes have been
made are not explicitly outlined in any great depth in either work, beyond
O'Toole's suggestion that "Different labels [than the linguistic ones] are
appropriate for other semiotic codes such as painting, sculpture and music, but the
labels stand for similar functions, or types of meaning relation" (1994:5), and the
statement by Kress and van Leeuwen: “In the form in which we gloss them [the
metafunctions] here they are not specific to any one semiotic mode: for instance
they are not specific to the linguistic” (1996:40).

113
The reasons that they do this may possibly be traced therefore to the fact that they
wish to avoid a confusion of terminology. This is an important point, in that this
study would also aim to avoid terminological confusion. However, since this
study is attempting to analyse the ways that both the visual and verbal modes
work together to realise intersemiotic complementarity, some consistency in
terminology would be appropriate, especially since it is argued that both modes do
in many ways realise the same or similar social meanings, but in ways that are
mode-specific.

This review of Kress and van Leeuwen’s visual grammar and O'Toole's schematic
framework has also assessed them in terms of their applicability to the kinds of
visuals commonly found in The Economist magazine. What was found was that
Kress and van Leeuwen's 'grammar of images' is very useful in elucidating the
kinds of visual meanings which typical visuals in The Economist magazine
project, and how they organise them. O'Toole's model is not as useful in these
terms since the level of delicacy of analysis did not reach that of Kress and van
Leeuwen's. Where his schematic model is most useful is in providing a way of
approaching the analysis of an image, whether it be a naturalistic photograph or a
scientific/technological graph or table, in terms of a visual rank scale, and in terms
of its relationship to its context of situation. The analyses by Kress and van
Leeuwen and O’Toole are therefore useful to varying degrees for the insights they
provide about the ways that various kinds of visuals organise and project their
meanings. However, their work is mode-specific — they concentrate on the visual
mode only and how each type projects its meanings, and do not attempt in any
extensive way to clarify what happens intersemiotically when the verbal and the
visual modes co-occur on the page.

Accordingly, as the focus of this study is to look at the intersemiotic relations

between the visual and verbal aspects of a multimodal text drawn from The
Economist magazine, Halliday’s metafunctional interpretation of communication,
along with selected ideas derived from the reinterpretation of these metafunctions
by Kress and van Leeuwen and O’Toole will be utilised in the development of a

114
theoretical framework for determining the resources used for the realisation of
intersemiotic complementarity. O’Toole’s application of the SFL notion that a text
and its context are dialectically related, will be specifically applied in the analysis
of the Economist multimodal text — a presentation and analysis of the contextual
variables relevant to this text should provide insights in terms of the context-text
relationships operating, and will provide evidentiary support for the interpretations
made in the analysis of intersemiotic complementarity.

For the intersemiotic examination of both visual and verbal modes in ideational
terms, the terminology of the linguistic and visual TRANSITIVITY systems will
be used, as will the terminology of MOOD (speech function) and Modality
(propositional attitudes) for the intersemiotic examination of both visual and
verbal modes in interpersonal terms. The intersemiotic examination of both visual
and verbal modes in textual terms however, will require some adaptation. An
attempt will be made to use Kress and van Leeuwen’s application of Given/New
to layout, but since this part of the intersemiotic analysis will be concerned with
composition on the page, it will be necessary to adopt the terminology of layout
and composition to explain how the two modes complement each other on the
page space. How this will be done will be explained in more depth in the next
chapter, where the framework will be presented. In Chapters Six and Seven
following, this analytical framework will then be applied to the sample
multimodal text drawn from The Economist magazine.

115
Chapter 5 - An Analytical Framework for the Analysis of
Intersemiotic Complementarity in The Economist
Magazine

5.0 Introduction
The reviews in Chapters Two, Three and Four in this study have revealed a
number of salient issues. Firstly, the review of the literature on the analysis of
economics discourse by economists and applied linguists showed that in spite of
an increasing awareness of the nature of economics discourse by both applied
linguists and economists, the published research has not attempted or been able to
in any significant way describe, explain or account for the visual mode in
economics discourse. Secondly, a brief examination of the research by
semioticians and linguists working with other modes of communication, focussing
on the visual mode, showed that there are basically three main schools of semiotic
theory which have attempted to examine other modes of communication utilising
linguistic insights originally derived from Saussure’s treatise. Thirdly, the
overview of the work of those who have attempted to investigate various kinds of
visual modes from within the Hallidayan SFL theoretical framework demonstrated
that while within this particular theoretical framework there is an expanding body
of work examining different manifestations of the visual mode (general images
and graphics, painting, sculpture and architecture), there is virtually no research
into the nature of the intersemiotic semantic relationship between the visual and
verbal modes in page-based multimodal text, and none at all in relation to
economics discourse.

In this chapter, the analytical framework utilised for the analysis of intersemiotic
complementarity in The Economist magazine will be explained and outlined. This
will involve a brief explanation of the theoretical principles and assumptions
underpinning this study, as well as a more focussed discussion of its various
features. A detailed application and discussion of this analytical framework will
be given in the ensuing chapters Six to Seven, where it will be applied in the
analysis of a representative text extracted from The Economist magazine.

113
 Extralinguistic levels

Context of culture

Context of situation

realised in

Linguistic levels

Semantics
(systems of meanings)

realised in

Lexicogrammar
(systems of wordings)

realised in

Phonology
(systems of sounds)

Figure 5.1 - Levels of language and realisation in the

SFL Model (adapted from Butt et al. 1995:11)

114
5.1 The Systemic Functional Linguistic (SFL) Model
As stated in Chapter One, the analysis of intersemiotic complementarity in this
study will take as its theoretical foundation a Hallidayan view of communication,
where language is seen as "social semiotic" (Halliday 1978). It will utilise and
seek to extend the applicability of Halliday's systemic functional linguistics (SFL)
model to page-based forms of multimodal communication. To Halliday, the term
‘social’ suggests two things simultaneously: firstly it refers to the social system,
which is synonymous with culture as a system of social meanings, and secondly it
refers to the dialectical relationship between communication (language) and social
structure. The SFL perspective therefore involves an “attempt to relate language
primarily to one particular aspect of human experience, namely that of social
structure” (Halliday and Hasan 1985:4). This relationship between language and
social situation implies that language is viewed as a system of choices or options
made against a background of other potential options, and against other ways of
communication which human beings have developed over time and in various
cultural contexts. As Hasan points out,
whereas Saussure thought of the various semiotic systems as simply
copresent, systemic functional linguistics attempts to integrate
language as a semiotic system with other semiotic systems in a
community. It is clear that the claim of cogenesis itself rests on that
of the semiotic relation of coding which integrates the context of
culture and situation with language and communication (1993:94).

Halliday (1978:16,21,27-29,109) makes four central claims about language:

1. Language is functional in terms of what it can do or what can be done with it

2. Language is semantic in that it is used to make meanings.

3. Language is semiotic in that it is a process of making meanings by selecting

"from the total set of options that constitute what can be meant" (op.cit:53).

4. The language meanings generated and exchanged are motivated by their social
and cultural contexts.

The central claims about language are represented in Figure 5.1. Here, language is
interpreted as a “complex semiotic system composed of multiple LEVELS or
STRATA” in which “the central stratum, the inner core of language, is that of
grammar” (Halliday 1994:15). This central stratum is referred to as the

115
LEXICOGRAMMAR, because it incorporates both grammar and vocabulary
(loc.cit.). The key concept used to describe the ways that these strata are related in
the overall model is the concept of realisation.

As Figure 5.1 shows, the linguistic levels are related to each other in that the level
of phonology and graphology realises the level of the lexicogrammar, and this
lexicogrammar itself realises the level of semantics or meanings, which also
realises the extralinguistic features of the context. Looking at this from an
opposite perspective, the extralinguistic features of the context are realised in the
choices made in the semantic level, these meanings are realised in choices made in
the lexicogrammar, and the lexicogrammar is realised by choices that are made in
the soundings and graphology (loc.cit.).

A more extensive explanation and application of the main tenets of the Systemic
Functional Linguistic Model to typical examples of economics discourse is
presented in Appendix 3, but according to the SFL view of language, a reader of
any article in The Economist magazine would interpret it as a text realising a
specific context of situation and simultaneously encoding all three of the
metafunctions at the level of semantics. In other words, in ideational terms the
reader would understand the article's processes, their participants, and the
circumstances being represented. He or she would also have to understand the
logical relationships between one process and another, or one participant and
another which share the same position in the text. In interpersonal terms, the
reader would recognise and respond to the encoded speech function, whether the
article is making offers, providing statements, asking questions or issuing
commands, as well as appreciate the attitudes and judgements embodied in the
text. In textual terms, the reader would appreciate the news value and topicality of
the message reported, or its relevance to the context in which it occurs, as well as
the coherence between one part of the article and every other part. All these
elements, in combination with the texture provided by the componential and
organic cohesive devices, operate to convey to the reader that the article he or she
is reading is coherent, relevant, and organised so that it effectively addresses the

116
reader in some recognisable, socially appropriate way (Halliday and Hasan
1985:45).

The following sections will examine the application of these SFL principles to
visual semiosis. The remaining sections will then introduce and outline the
assumptions underlying the analytical framework proposed in this study, and will
present it in the final section. This framework will then be used in the analysis of
the multimodal text extracted from The Economist magazine, which has
previously been referred to as the Mountains text.

5.2 The SFL Model and Visual Semiosis - An Analytical Framework

To be able to say anything meaningful about the interaction between the visual
and verbal modes in a multimodal text, there is a need for consistency in
describing how both verbal and visual meanings are communicated. SFL theory is
an exotropic theoretical paradigm which inherently allows, through its usage, the
ability to explain a range of communicative phenomena, and an ability to
communicate and work with other theoretical paradigms (Hasan in press). A
strength of the SFL model is that the concept of a text in terms of metafunctional
meaning also permits an analysis of semiosis from three different metafunctional
perspectives, the ideational, interpersonal and textual, with the assumption that an
analytical focus on any one necessarily implies that the other two are and should
be considered as operating simultaneously. Further, the view that communication
involves “systems of meanings” and the act of communication involves making
simultaneous selections from those systems in terms of what is going on (the field
of discourse), who is taking part (the tenor of discourse), and the role assigned to
language (the mode of discourse). This suggests that it is a paradigm which can
perhaps be usefully applied to other systems of meaning besides language. This
has been demonstrated clearly by the work of O’Toole (1994, 1995) and Kress and
van Leeuwen (1990, 1996). In their applications to other visual modes, they show
that the Hallidayan SFL model can be utilised not only for the sociolinguistic
analysis of natural language, but can also "offer a powerful and flexible model for
the study of other semiotic codes" (O'Toole 1995:159). However, as stated in
Chapter One, the

117
Extravisual levels

Context of culture

Context of situation

realised in

Visual levels

Semantics
(systems of visual meanings)

realised in

Visual grammar
(systems of visual design)

realised in

Representational Symbology
(systems of display elements)

Figure 5.2 Levels of visual expression and realisation in

the SFL Model

118
question that this study will ask is can SFL theory also accommodate the
synergism of the intersemiotic complementarity between language and those other
semiotic systems. That is, along with providing the means whereby visual
meaning can be described, can it be used to show how meanings interact between
modes within the boundaries of a single page-based text (Halliday and Hasan
1985:26).

Before looking specifically at the framework for intersemiotic complementarity, a

discussion of the visual semiotic framework is presented below, and an adaptation
and application of the SFL model to the visual semiotic is presented in Figure 5.2.
At the lowest level of this model of visual expression and realisation (which in the
SFL or linguistic model is the level of phonology), the term Representational
Symbology is used. The assumption here is that each visual instance consists of
choices that have been made from the systems of fundamental display elements
which are available to the producers of a visual. The term ‘Representational’ is
derived from a common classification used in communication and media studies
research. Generally, this research identifies a communication medium as the
“physical or technical means of converting a MESSAGE into a signal capable of
being transmitted along a given CHANNEL” (Watson and Hill 1997:139). It also
typically identifies a communication medium as being either Presentational,
Representational, or Mechanical. Presentational media refers to linguistic features
(the voice or the spoken word) and gestural features (the face and body) which are
involved in acts of communication, with the person doing the communicating
viewed as the medium. Representational media however is concerned with works
of communication, where the medium is a book, painting, photograph or drawing
which creates some kind of ‘text’ that is independent of its author or designer, and
it is in this sense that it is directly relevant to an examination of page-based
communication. Mechanical media approximates a physical channel and refers
simply to the radio, television, film, telephone, and computer media which act as
transmitters for the presentational and representational media (Fiske 1982:18).

The visual expression level of Representational Symbology in Figure 5.2

specifically refers to the various display elements of which a work of visual

119
communication, at its most basic level, is constructed. These are the visual
elements which in a sense are “the compositional source for all kinds of visual
materials and messages and objects and experiences” (Dondis 1973:15).
Representational Symbology is therefore concerned with the ways that visual
signs and symbols (or works) are produced through the use of various primary
display elements, which Dondis (op.cit:15-16) summarises as:
• the dot; which is the minimal visual unit, pointer or marker of space on the
page.
• the line; this can be a fluid “restless articulator of form” in sketches, or a rigid
line which is used to tightly control visual space (as in a technical drawing).
• shape; this includes the basic geometrical shapes of the circle, square, triangle
and their various combinations and dimensional versions.
• direction; this is the “thrust of movement” (vectors) which arise from the
nature of the various circular, diagonal and perpendicular shapes.
• tone; the presence or absence of light.
• colour; allied with tone, this element is important for its chromaticity (purity
and intensity of hue).
• texture; the surface characteristics, which can be optical or tactile.
• scale or proportion; is concerned with salience, or relative size and
measurement.
• dimension and motion; the use of perspective to give a sense of depth, and the
use of depth of field in still and moving film.

The artist, craftsperson or graphic designer is thus the visualiser who, through the
choices he or she makes, manipulates these basic visual elements to create an
intended effect or to project any number of specific messages. In both the visual
and linguistic systems, there is plenty of opportunity for creating meaning, for
reiterating existing meanings, for generating original meanings; the meaning
potential is thus limitless. Like the linguistic system, the visual system relies on a
set of intersubjective conventions constrained by a specific relevant context. Like
language, any number of existing or new visual messages can be created, and in
the same ways that each spoken or written text is an instance of the language
system, so too is each visual an instance of the visual system. To paraphrase
Halliday (1991:7), the context for this meaning potential - for visual language as a
system - is the context of culture, and the context for the particular instances - for
visual language as processes of text - is a context of use. Just as a sketch or a
diagram is an instance of visual language, so is a situation of visual representation
an instance of culture. Thus, the context for an instance of visual language is an

120
instantiation of choices made constrained by a specific situation, and the context
for the system that lies behind each visual is the system which lies behind each
situation - namely, the culture.

The level of visual grammar, which is characterised as the systems of visual

design in Figure 5.2, relates to the ways that the various systems of display
elements in Representational Symbology are combined to realise visual message
‘syntagms’, or the ways in which visual elements are organised into recognisable
structures (Dondis 1973:20-38). In the same ways that a linguistic grammar
combines sounds into words which then combines these into clauses, sentences
and whole texts, a visual-grammar looks at the ways that people, creatures,
mythological beings, inanimate objects, and spatial representations (places,
scenes, landscapes) have been created by the basic visual elements combining in
meaningful ways to produce coherent visual phenomena of varying degrees of
complexity. Both language and visual phenomena rely on a specific
intersubjective sign system (their respective semiotic systems), and the choices
available in this system are organised in specific ways that make sense to
members of a culture.

The use of the term ‘meaningful’ above is important in that it is the primary focus
of this study’s approach to visual interpretation. The aim here is to view visual
communication, in line with the SFL view of language, as instances of meaning
which are structured according to function, not, as analysts in various structuralist
schools of semiotics have generally done, to examine visuals in terms of their
isolated elements. Their approach was has been to focus on
what linguists would call ‘lexis’ rather than ‘grammar’, on the
‘vocabulary’ — for instance, on the ‘denotative’ and the
‘connotative’, the ‘iconographical’ and the ‘iconological’
significance of the individual people, places and things (including
the abstract ‘things’) depicted in images (Kress and van Leeuwen
1996:1).

As stated in Chapter One, this study recognises that verbal and visual modes
utilise the meaning-making features peculiar to their respective semiotic systems,
in the sense that there are some individual meanings which can be expressed only

121
visually, and some which can be expressed only through language. It also
recognises that there are areas where they both share meanings. Like the
lexicogrammar however, visual-grammar operates as “a means of representing
patterns of experience” which “enables human beings to build a mental picture of
reality, to make sense of their experience of what goes on around them and inside
them” (Halliday 1994:106). Visual-grammar also works as a means of projecting
and exchanging messages, to generate forms of address to potential viewers, and
to ‘colour’ those forms of address in modal and attitudinal terms. Additionally,
visual-grammar works as a means to project a unified, coherent visual message, to
organise the elements of its composition in such a way that the viewers will be
able to see how one part of the visual ‘fits’ with every other part, leading to a
sense of visual coherence. In the lexicogrammar of the SFL model, the clause
plays a central role in embodying experience, organising the nature of the
exchange, and in organising the message. In the visual-grammar, visuals organise
their meanings in similar ways, in what could be viewed as visual ‘syntagms’
which are realised by various arrangements of the core visual display elements
outlined above. Thus, in visual-grammar, as there is in lexicogrammar, there are
various visual ways of relating the participants portrayed (through visual
Transitivity systems), of relating the viewer and the viewed (through visual Mood
and Modality systems), and of relating the elements on a page to each other
(through visual composition systems).

As already indicated in Chapter Four, the analytical framework developed here

will draw on aspects of Kress and van Leeuwen’s (1990, 1996) and O’Toole’s
(1994, 1995) work. Kress and van Leeuwen’s approach is especially important in
that they have attempted, using the terminology of research in graphic design,
media studies, and visual semiotics, to develop a grammar of images which
describes a variety of different kinds of meanings and the ways that visuals can
realise them. Their wide range of specialist terms is very useful for describing the
various ways that visuals can project their meanings; however these terms are
mode-specific, and as such make the task of describing intersemiotic relations
between the verbal and visual modes an exceedingly complex undertaking.
Therefore, the study reported here presents a more consistent use of theoretical

122
concepts and terminology from SFL in order to facilitate the description, in socio-
semantic terms, of how both modes organise and express their meanings. Hence,
this study uses the existing terminology of SFL theory along with selected ideas
derived from the reinterpretation of the metafunctions by Kress and van Leeuwen
and O’Toole to explain intersemiotic semantic relations in page-based multimodal
text.

At the level of the semantics in Figure 5.2, the metafunctions will be interpreted in
similar ways to those used by Kress and van Leeuwen, and O’Toole. The
ideational metafunction is seen as the function of language to represent the
‘goings on’ in the world. In analysing visuals the starting point is to identify the
represented participants, or all the elements or entities that are actually present in
the visual (animate or inanimate), as well as the processes in which they are
engaged and the circumstances in which they are found. The interpersonal
metafunction is the function of language to represent the roles and statuses that
participants hold in any form of interaction, and here the interactive participants
are the foci  this includes the participants who are interacting with each other in
the act of reading a visual, one being the graphic designer or drawer, and the other
the viewer, and the social relations between the viewer and the visual. The textual
metafunction is that function of language through which a text can be recognised
as having coherence and as making sense. The focal point in a consideration of
page-based multimodal text is the coherent structural elements or composition.
This relates to aspects of layout and design which combine and integrate the
elements on the page in a way in which the graphic designers or drawers wish to
present at a particular point in time (Kress and van Leeuwen 1996:183).

Reading (or viewing) a visual therefore involves the simultaneous interplay of

three elements which correlate with Halliday's (1985) three metafunctions: the
ideational, the interpersonal and the textual. For the visual mode these relate to the
represented participants, the interactive participants, and the visual's coherent
structural elements (Kress and van Leeuwen 1990:16-21).

123
The social semiotic view of communication embodied in the SFL model and
adopted for this study of intersemiotic complementarity implies that whether a text
contains only verbal, or both verbal and visual modes, it is viewed as embodying
the pattern of purposeful choices made by its constructors in order to make
meanings for others to receive and respond to in some way. This process of
making selections against a background of other possible choices also applies to
principles of graphic design, as alluded to by Tufte (1983:191):
Design is choice. The theory of the visual display of quantitative
information consists of principles that generate design options and
that guide choices among options. The principles should not be
applied rigidly or in a peevish spirit; they are not logically or
mathematically certain; and it is better to violate any principle than
to place graceless and inelegant marks on paper.

A socio-intersemiotic view of multimodal text makes a number of important

assumptions that are derived directly from the SFL model. Firstly it assumes that a
multimodal text is constructed with a view to projecting or sending meanings
within a social context to some other person, whether that person be a listener, a
reader, or a viewer. The channel used for the transmission of these meanings can
be visual, verbal, or some combination, and each channel will communicate the
socially-based meanings in a form that is appropriate to the medium.

Secondly, it assumes that these social meaning selections, verbal or visual, are
activated by the cultural context in which they are situated. The resultant
multimodal text is an instantiation of these choices, and can as a result be viewed
as a realisation of the contexts of situation and culture-bound choices made by its
constructors. At the same time, in line with the dialectical relationship between
text and culture in the SFL model, a multimodal text also construes the context of
situation and culture in which it occurs. Thus, one can say that visual and verbal
language is in culture, and culture is in visual and verbal language (Hasan 1981,
1995, 1996).

Thirdly, it assumes that the ways that people communicate in various visual and
verbal modes are the result of the choices they have made or the options they have
taken up from each particular semiotic system. If the orthographic mode of writing

124
is chosen, then the person making the choice will have made a range of choices
from a variety of meaningful options available in that orthographic system. In the
same way, someone who designs, draws or develops a visual has made a range of
choices from a visual social semiotic system, choices which, like those from the
written mode, are situated in the social, cultural and ideological contexts in which
they have been made and which they share with others. They are intersubjective
sign systems, by virtue of the fact that in “a community [they] serve to define the
nature of the ‘world’ for its members” and have “a role in the mediation of
meanings” between the members (Hasan 1981:107). It is argued therefore that
when someone makes these choices from more than one semiotic code, and then
chooses to include these choices in the instantiation of one text, then there should
be intersemiotic evidence of these choices, evidence which illustrates how the
different modes complement each other to produce a coherent configuration of
meanings in the form of a multimodal text.

The metafunctional terminology that will be adopted and used in this study, along
with a comparative listing of previous applications to visual phenomena of the
SFL metafunctional concept are outlined in Table 5.1.

SFL PREVIOUS INTERPRETATIONS THIS STUDY

HALLIDAY KRESS/VAN LEEUWEN O’TOOLE ROYCE

(Systemic Functional (Reading Images/Grammar of (Language of Displayed Art) (Intersemiotic
Linguistics) Visual Design) Complementarity)

IDEATIONAL REPRESENTATIONAL REPRESENTATIONAL IDEATIONAL

INTERPERSONAL INTERACTIONAL MODAL INTERPERSONAL

TEXTUAL COMPOSITIONAL COMPOSITIONAL COMPOSITIONAL

Table 5.1 Metafunctional views of visual communication

As stated earlier, the metafunctional terminology of the SFL model will be

maintained as much as possible to avoid terminological redundancy. The term
Compositional however, has been used instead of Halliday’s textual metafunction
because it seems to capture more fully the sense of two modes interacting with
each other to project meaning coherently on the page. This is in general agreement

125
with Kress and van Leeuwen’s usage, but is at slight variance with O’Toole’s
(1994:278) use of composition, where he uses it in the more traditional sense of

126
 EXTRALINGUISTIC AND EXTRAVISUAL LEVELS

Context of culture
Context of situation
realised in

LINGUISTIC SEMIOTIC VISUAL SEMIOTIC

Intersemiotic Complementarity
Semantics ideational Semantics
(systems of meanings) interpersonal (systems of meanings)
compositional

realised in realised in

Lexicogrammar Visual grammar

(systems of wordings) (systems of visual design)

realised in realised in

Phonology Representational Symbology

(systems of sounds) (systems of display elements)

Fig. 5.3 A framework for visual-verbal intersemiotic complementarity in page-based multimodal text (adapted from Butt et al. 1995:11)

127
composition within the art piece, and the artistic interpretation of how the
elements of a work of art cohere to produce a sense of visual unity or wholeness
within the work.

The framework for the analysis and interpretation of co-occurring visual and
verbal modes in multimodal text is presented in Figure 5.3.

In line with Kress and van Leeuwen’s (1990:4) comments, this framework
suggests that both visual and verbal forms of communication can realise the same
systems of meaning in various cultures, and that each mode will do this
independently and via its own specific forms. It suggests further that much of what
can be said about the linguistic semiotic code can also be said about the visual
semiotic code (op.cit:19), and that when they co-occur within the boundaries of
page-based text, they project a coherent multimodal text through semantic devices
which realise intersemiotic complementarity. Figure 5.3 also presents the main
parts of the analytical framework that will be used in this study, and the following
sections will examine the approaches to be taken with each of these. Sections 5.3
and 5.4 will discuss the importance of the contexts of culture and situation and
intertextuality for the interpretation of multimodal text, while Sections 5.5, 5.6,
and 5.7 will expand and examine the procedures that will be followed in the
metafunctional analysis of the Mountains text. Each of these sections corresponds
to a specific intersemiotic metafunction and outlines the main areas where
intersemiotic complementarity may potentially be realised. Section 5.7 will then
summarise the procedures into a full framework of analytical procedures.

5.3 The Context of Situation and Context of Culture

In the SFL model, the semantics forms the interface between any instance of the
linguistic system and the context in which it occurs, the nature of the relationship
between them being one of realisation, where “context is stratified above language
and is thus realised by language” (Matthiessen 1995:33). It was suggested in
Section 5.2 however, that the semantics can also be viewed as forming the
interface between any instance of the visual system and the context in which it
occurs, and that this can apply to a multimodal text in a context where both the

128
linguistic and visual systems co-occur. This section will now set up the ways that
the notions of Context of Situation and Context of Culture can be interpreted and
applied to the analysis of page-based text in which visual and verbal semiotic
systems co-occur.

The three functionally defined domains of Field, Tenor, and Mode are the
contextual variables whose particular configurations and dialectical relationship to
the semantics make up the "semiotics of context of situation" for a multimodal text
(Matthiessen 1995:33; Ventola 1995:6). Halliday defines the semiotic structure of
the situation as that which “relates the situation simultaneously to the text, to the
linguistic system, and to the social system” (1978:142), and it has already been
proposed that this definition is relevant to the relationship between context and
other semiotic systems, such as the visual. The situation is therefore interpreted as
a semiotic structure, where it is an instance of the meanings that comprise the
social system, and is characterised as consisting of the social action, or that
‘which is going on’, [FIELD], the role structure, or the cluster of socially
meaningful participant relationships [TENOR], and the symbolic organisation, or
the particular status that is assigned to the text within the situation, its function in
relation to the social action and the role structure, the channel or medium, and the
rhetorical mode [MODE] (1978:142-143).

In this study, the relationship between context and multimodal text is viewed as a
dialectic or “dialogic exchange” (Matthiessen 1995:33). Here, the intersemiotic
ideational metafunction realises the Field component of the context, the
intersemiotic interpersonal metafunction realises the Tenor of the relationships
between the interactants, and the intersemiotic compositional metafunction
realises the Mode. Thus, a multimodal text is “the product of its environment, and
it functions in that environment" to be both activated by the context in which it
occurs, and at the same time to construe it (Halliday 1978:136).

Halliday (op.cit:110) has also suggested that there are two orders of contextual
categories, referred to as first and second-order context. First-order context is
oriented towards and based on the social system and not towards categories that

129
 FIELD TENOR MODE
Second-order domains of social social roles enacted The role played by
experience brought into through language and/or language and/or visual
existence by language visual expression, as in expression in First-order
and/or visual expression the general speech roles: context, in relation to the
in the course of social questioner - answerer, channel, medium, and
processes e.g. financial, stater - acknowledger, rhetorical mode.
economic, journalistic, offerer - acceptor,
culinary, meteorological, commander - complier.
nurture, common-sense,
etc.

First-order social processes in the social roles and relations

social system, e.g. in the social system,
productive, educational, relating to e.g. family &
informational, kinship, class & caste,
recreational, etc. occupational role, etc.

Table 5.2 First-order and Second-order context in multimodal text (from Matthiessen
1995:35)

130
are specifically linguistic or some other semiotic system (such as the visual). Thus
the first-order contextual variable field is concerned with the social processes in
the social system, while the first-order contextual variable tenor is concerned with
the social roles and relations in the social system. Second-order context however
can only occur through the agency of language or some other semiotic system
(such as the visual). Second-order field consists of the social experience construed
by the semantic system (to what is often referred to as the subject matter), while
second-order tenor relates to the social roles enacted by the semantics (such as the
speech roles). Mode is viewed as being inherently second-order in that it is
oriented towards the role of language or visual expression in context (Matthiessen
1995:34).

A description of the first and second-order contextual categories in relation to

multimodal meaning is given in Table 5.2. For the description of the broader
contextual features of multimodal texts like those which occur in The Economist
magazine (which are constitutive of the social process, as opposed to being
ancillary), Hasan’s (1996) work on context in language is informative. Discussing
what it means to say that the context of situation is illuminated in the text, and
following on from Halliday’s conception of the context of situation (as outlined
above), Hasan makes a distinction between “at least three orders of context” for
those kinds of texts where the role of language is constitutive (op.cit:52). In order
to account for this she refers to these three orders as (i) the context of creation, (ii)
the context of the audience’s contact with the text, and (iii) the reconstituted
context that is specific to that one text (loc.cit.).

The context of creation pertains firstly to the ways that “artistic conventions of the
author’s community are reflected - indirectly or directly - in the created text”
(loc.cit.). For this study, the writing and graphic design policies of The Economist
magazine’s staff in terms of its conceptions of the medium of its message, the
relative primacy of the visual and verbal modes, the subject-matter, the
typographic conventions, the writing/stylistic conventions, and its visual typology
are important. Secondly, the context of creation pertains to the “author’s
conception of the audience” (loc.cit.) which relates to his or her perception and

131
attitudes towards the readership and how they are built into the text. Each text
produced at The Economist magazine is a collaborative effort, so the ‘author’ here
is the staff involved in that creation and editorial process (discussed in Chapter
Six). Thirdly, the context of creation concerns how “the author’s own individual
preoccupations are built into the text” (loc.cit.); again this is concerned with how
The Economist magazine team of writers and editors interpret their roles as a
publisher dealing with economic, financial and eco-political issues. Each of these
three features will be examined in some depth in this study.

More problematic, however, is obtaining information on the context of the

audience’s contact with the text. This may be possible in situations where the
audience is on hand to be observed (as in a lecture), but it is much more difficult
in relation to the audience (readership) of The Economist magazine spread
throughout the world. A major survey of its subscribers would be required, and is
not the focus or within the scope of this study.

The third order of context, the reconstituted context that is specific to that one
text, is explained by Hasan as being
what it [the text] is about, in what relations the characters and
events are placed vis a vis each other, how do the theses hang
together and what are the strategies through which the text
achieves a generally recognisable generic shape (loc.cit.).

To sum up, a multimodal text is therefore a particular contextual configuration or

instance which derives its meaning from the broader UK and world economic,
financial and eco-political culture, as well as The Economist magazine as a
journalistic publishing institution which is comprised of a range of attitudes
towards and perceptions of its readership. This incorporates its various views of its
role as a publisher of reportage dealing with economic, financial and eco-political
issues. This context of culture would also include how The Economist magazine
views the medium of its message, in other words, how it views both its verbal and
visual output relative to priorities, typographic conventions, writing/stylistic
conventions and so on.

132
5.4 The Influence of Other Texts: Intertextuality History
An interpretation of any multimodal text in The Economist magazine as a
particular contextual configuration does however need to be supported by an
appreciation of various relevant aspects of the text’s background history, the
importance of which can be discussed in terms of the concept of intertextuality.
Derived originally from Bakhtin’s attempts to account for the ways that texts are
formed by the previous texts to which they are in a sense responding, and by the
future texts which they are in a sense anticipating, intertextuality is concerned
with the ways that in language
any concrete utterance is a link in the chain of speech
communication of a particular sphere ….. Each utterance is filled
with echoes and reverberations of other utterances to which it is
related by the communality of the sphere of speech communication.
Every utterance must be regarded primarily as a response to
preceding utterances of the given sphere …… Each utterance refutes,
affirms, supplements, and relies on the others, presupposes them to
be known, and somehow takes them into account (Bakhtin 1986:91).

Other researchers who have drawn on and refined this concept in their own work
include Lemke (1995), Halliday (1985) and notably Fairclough (1989, 1992).
Lemke (1995:85) refers to a principle of general intertextuality in his discussions
of intertextuality and text semantics in a social semiotic framework, where the
"meaning of a text depends directly on the kinds of connections made in a
particular community between it and other texts". For Halliday, the concept of
intertextuality means that a text is not just simply an isolate resulting from a
particular contextual configuration, but it is also the result of the influence of other
texts, especially previous texts which are taken for granted by those who share in
its production and reception. Every text therefore has a ‘history’, and having that
history means that the previous interactions between other texts and their contexts
in a sense "leave their mark". This would include not only ideational and
interpersonal features of previous texts being carried over, but also perhaps some
'coded' expressions or formulaic sequences which could signal what is happening,
or act to possibly predict what will occur next (Halliday and Hasan 1985:47). Here
Halliday is touching on what Fairclough (1992:101-136) later refers to when he
distinguishes between two types of intertextual relations, those of manifest
intertextuality and interdiscursivity (or constitutive intertextuality). According to

133
Fairclough, manifest intertextuality refers to “the case where specific other texts
are overtly drawn upon within a text” (op.cit:117). This can be where “other texts
are explicitly present in the text under analysis: they are ‘manifestly’ marked or
cued by features on the surface of the text, such as quotation marks”, or it can
refer to cases where “a text may ‘incorporate’ another text without the latter being
explicitly cued: one can respond to another text in the way one words one’s own
text” (op.cit:104). Interdiscursivity (constitutive intertextuality) however is “a
matter of how a discourse type is constituted through a combination of elements of
orders of discourse” (op.cit:118), wherein the intertextual aspects of a text can be
viewed as “incorporating the potentially complex relationships it has with the
conventions (genres, discourses, styles, activity types) which are structured
together to constitute an order of discourse” (op.cit:103).

In this study the historical or manifest aspects of intertextuality are most relevant
because many articles published in The Economist deal with issues (i.e. Field) that
have developed over time, deal with them in similar ways (i.e. Tenor, Mode), and
therefore may carry explicit or less explicitly apparent features which could
inform an intersemiotic interpretation. Each text in The Economist magazine deals
with a financial issue that may or may not have received lengthy ongoing press
coverage since it first came into prominence. The degree and intensity of coverage
will vary from text to text and from issue to issue, but to provide some coherence
to a discussion of the Mountains text's intertextual history, and of the important
contextual variables which may have an impact on the relationship between its
verbal and visual messages, three specific contextual areas will be examined.
These are:
• The Subject Matter and the Issue (second-order field related)
• The Attitudes (second-order tenor-related)
• Visual Typology (second-order mode-related)

Each of these areas has been chosen for specific reasons. The subject matter is
concerned with general aspects of the topic area, such as its historical
development, or its conceptual and/or informational structure. This relates most
closely to a general background knowledge of the subject area, or those details
which allow or mean that the producers and receivers of the text can be considered

134
as members of a particular discourse community. In the case of The Economist
magazine, this would mean that those who have written, or those who have
produced a particular text would be able to understand and appreciate the
information contained within, and concomitantly, the breadth of that information.
The nature of the subject matter is important because it impinges directly on the
character and source of the issue being addressed. Information on the issue being
addressed is of course the raison d'ê-tre for the text, and deals with the nature of
the issue (or problem) at the time of its production.

The attitudes expressed over time can provide valuable insights into how The
Economist magazine writers attempt to attract and orient potential readers, and
how they structure their analysis of the issue and its future development and
resolution. The attitudes of the writers can be examined from three perspectives:
the first is how they attempt to grab the potential reader’s interest and orient them
to the topic through the use of devices such as article headlines or sketch
caricatures, the second is via the topic focus of the visuals used, and the third is
through an examination of the attitudes expressed in the verbal description and
analysis of the issues, and how that is reinforced by the attitudes expressed in the
visuals used.

A visual typology for The Economist magazine can reveal the ways that this
institution characteristically presents its visual information, and thus says
something about the conventions it uses in graphic design, the ways that it views
its subject matter, and the assumptions that it makes about its readership. Kress
and van Leeuwen have adapted and utilised the term 'coding orientation' to explain
a range of different categories of visuals, categories which they define as "sets of
abstract principles which inform the way in which texts are coded by specific
social groups, or within specific, institutional contexts" (1990:53). Accordingly, it
is proposed that for this study there are basically three categories of visual
communication. They are termed:
• Naturalistic visuals - this describes attempts to portray by some method what
may be considered 'natural' images, or images which the members of a
particular cultural setting would consider to be a recognisable form of a
representation of reality as viewed by the human eye.

135
• Mathematical visuals - this describes the various forms of visuals which are
derived from the scientific/technological coding orientation, and are used to
show structure, demonstrate numerical relationships, illustrate dynamic
relations, clarify processes, and explicate procedures.

• Spatial visuals - this describes the various forms of map visuals. They are
termed spatial because they display their meanings on a flat surface for locative
purposes

The full derivation and explanation of this visual typology for The Economist
magazine will be outlined and explained in Chapter Six.

The view of this study therefore is that a multimodal text is not the isolated result
of a specific contextual configuration, but arises due to the influence of other
texts, especially previous texts. It is proposed that an analysis of the intertextual
history of the Mountains text can usefully inform an interpretation of the
intersemiotic, semantic interaction between its visual and verbal messages. As
suggested earlier by Halliday (1985:47), previous interactions between other texts
and their contexts have left not only the ideational and interpersonal features of
previous texts, but also perhaps some 'coded' expressions or formulaic sequences
which could signal what is happening, or act to possibly predict what will occur
next. Accordingly, the Mountains text will be examined in terms of the three
aspects of intertextual history already discussed above, the subject matter and the
issue, the attitudes, and visual typology. The sections which follow will now set
out the analytical procedures for the intersemiotic metafunctional analysis of the
Mountains text itself.

5.5 The Analysis of Intersemiotic Ideational Features

Before beginning an explanation of the various aspects of intersemiotic
complementarity which will be investigated, a decision needs to be made about
where to actually start with the analysis of the multimodal Mountains text. In this
study, in each instance of textual analysis, the analysis will commence with an
examination of the visual component, and then the verbal component. The reason
for this is mainly for ease of analysis; this is because the visual mode in an
Economist magazine text is less complex than the verbal component in terms of

136
the degree and number of meanings that are being projected. Related to this is the
fact that the verbal component is viewed as the primary vehicle for
communicating by the text-constructors at The Economist's graphic design
department (discussed more fully in the next Chapter), a situation that is
complemented by the traditional primacy of the verbal over the visual in modern
educational systems and modern modes of communication (Barthes 1977, Kress
and van Leeuwen 1996:18-21). Given the greater complexity of the verbal mode
in most page-based multimodal texts it is more than likely that it would carry a far
greater number of meanings, meanings which the visual mode does not or cannot
address. These meanings would in a sense be ‘superfluous’ to the analysis, and
would then need to be ignored because they do not relate semantically to the
visual mode in any clear or specific way. It would thus seem to be more efficient
to start with the mode that is usually packed with less meaning, and then to see
where and how these meanings relate semantically to the verbal aspect of the text.
Related to this are questions of why some meanings are presented visually and not
others, and what is special about the meanings projected visually. It is hoped that
the answers may be found not only in the contextual analysis of the sample text,
but also as a result of the intersemiotic metafunctional analysis.

5.5.1 The Interpretation of Visual Ideational Features

In interpreting visual communication in terms of the ideational metafunction, one
is looking at the representation of experience, the conveyance of meaning in terms
of content, or a presentation of information about something, or someone. As
already discussed, in the lexicogrammatical system, the ideational metafunction is
realised by the clause as representation, and the range of choices that can be made
in the system of Transitivity. This system basically deals with types of process,
and these are semantically represented in the clause in terms of the process itself,
the participants in the process and their attributes, and the circumstances of the
process (Halliday 1978:101). It has also been argued that these semantic
categories are similarly useful for explaining how the constructors of a visual have
represented the ideational meanings they wish to convey. Like linguistic
structures, visual structures and the visual processes embodied within them are
built into the semantics of the various visual communication modes, and they are

137
systematically associated with different kinds of participant roles. Visuals are
representations of reality, or representations of experience and information, and in
that sense they can realise the ideational metafunction, through which patterns of
experience are able to be represented.

To be able to examine the ideational intersemiotic features of an Economist text,

or how the visual and verbal modes interact semantically, the first step is to
examine a visual in relation to its represented participants, the represented
processes, and what those actions represent circumstantially according to the
represented context of situation (Kress and van Leeuwen 1990, 1996, Halliday
1985, 1994). As already alluded to in the previous section, in The Economist
magazine there are basically three categories of visual communication termed
Naturalistic, Mathematical, and Spatial visuals respectively. Each of these visual
presentational modes potentially express their visual semantics singly, or in some
relational combination, and do so in accordance with the inherent nature of the
medium. These visual meanings can be determined by asking various questions of
the visual text which focus on the ways they represent ideational meanings:

For Naturalistic Visuals:

• who are their represented participants, their Identification, or who or what is in
the visual frame (animate or inanimate),

• what is the Activity, or what action is taking place between the actor(s) and the
recipient(s) or object(s) of that action,

• what are the Circumstances, or the elements which are locative (concerned
with the setting), of accompaniment (participants not involved with the action),
and of means (participants used by the actors), and

• what are the Attributes, or the qualities and characteristics of the participants.

For Mathematical Visuals:

• what are their represented participants — the Identification of the represented
participants, which can be dependent or independent variables, physical
objects, forms, formats, symmetrical arrangements.

• what is the Relational Activity between the participants. A mathematical

relation, occurring singly, or in some combination, which can be a derivative
(rate of change), comparative, or proportional relation.

138
• what are the Circumstances in terms of where, what with, and by what means
the relations are realised, which can be in terms of setting, means, and
accompaniment.

• what are the Attributes — what are the qualities and characteristics of the
relationships between the variables represented - a category which is difficult
to apply to mathematical visuals. It is more readily and appropriately applied to
naturalistic representations.

These aspects in a visual, Identification, Activity, Circumstances and Attributes,

correlate with some of the Transitivity features of the clause as identified in the
SFL model. When individually determined, these features will be referred to in
this study as Visual Message Elements (hereafter VME or VMEs), and will form
the reference point for the examination of the verbal aspect of the multimodal text.

VMEs carry semantic properties which are realised by a range of visual

techniques at the disposal of the writers (i.e. drawers, graphic designers etc.), and
are influenced by various contextual features. The previous discussion on the
context of situation, context of culture, and intertextual history in multimodal text
illustrates clearly that a visual’s ideational or representational features do not
occur in isolation — the messages the visuals propose to impart to the viewer are
influenced and constrained by what has been written and/or drawn before by The
Economist magazine writers and the ways it has been typically presented. Each
visual is very much time and context-bound, and the nature and characteristics of
its VMEs largely depends on the myriad other sources of reportage and analysis
which the reader/viewers may have experienced in the days and months prior to
the publication of an article.

5.5.2 The Interpretation of Verbal Ideational Features

As already mentioned, in interpreting verbal communication in terms of the
ideational metafunction, one is also looking at the representation of experience.
The analysis of clauses in a text in terms of their Transitivity features is usually
done clause by clause, and depending on the length of the text, is quite a complex
and revealing process. However, given the decision to start the analysis with the
visual component as suggested above, and since the concern here is to identify
only those features of the verbal component which are related semantically in

139
some way to the identified VMEs, a full Transitivity analysis of the types typically
carried out in an SFL analysis will not be required. Instead, each sentence will be
examined in terms of the particular Transitivity features which in some way are
related semantically to the already-identified VMEs i.e. the participants,
processes, circumstances and attributes. These lexical elements which relate to the
visual meanings also arise according to their Identification (the participants, or
who or what is involved in any activity) and their Attributes (in terms of their
qualities and characteristics), the Activity (the processes, or what action is taking
place, what events, states, or types of behaviour), and the Circumstances (in terms
of where, who with, and by what means are the activities being carried out).

The next stage therefore is to check through the verbal aspect of the text for
semantically-related lexical items using these Transitivity categories. Decisions
about which lexical items to include or exclude in relation to each Visual Message
Element are based on the notion that the lexical items should be the closest
semantically to each visual element, or be reasonably expected to co-occur or
collocate in a text drawn from that particular context of situation. All uses of
language have an immediate environment in which they operate, and any
multimodal text is an instance of communication "structured as a field of
significant social action" (Halliday 1978:43).

5.5.3 The Analysis of Ideational Intersemiotic Complementarity

Once the relevant features of the contexts of situation and culture along with
aspects of a text’s intertextual history have been described, and the VMEs in the
visuals identified, it is then possible to examine these features in relation to how
they complement the verbal aspect of a text in lexico-semantic, intersemiotic
terms. The aim here is to identify how intersemiotic complementarity is realised,
and to try to explain why the writers of this text at The Economist magazine, in an
attempt to produce a coherent article, have chosen to include both verbal and
visual modes.

Starting with the VMEs, and checking through the verbal aspect of the text for
semantically-related lexical items produces a series of lexical inventories. This is

140
an adaptation of the analysis of cohesion chains (also referred to as lexical strings)
which express the semantic relations between lexical items in a text (Halliday and
Hasan 1985:82-94). In this study these lexical inventories express the various
kinds of semantic relations existing between the VMEs and the lexical items
found in the verbal aspect of the text, and as such constitute what may be referred
to as an inventory or list of intersemiotic semantic relations.

Approaches to examining cohesive relations in linguistic text include the

Stratificational Model by Gutwinski (1976), who used the categories of repetition
of item, the occurrence of a synonym or item formed on the same root, and the
occurrence of an item from the same lexical set (co-occurrence group); the
Systemic Model by Halliday and Hasan (1976), which used the categories of same
item, a synonym or near synonym (including hyponymy), superordinate, the
"General" item, and collocation (including antonymy and meronymy); the later
Halliday and Hasan (1985) general lexical cohesion categories (or sense relations)
of repetition, synonymy, antonymy, hyponymy, and meronymy, which represent
an attempt to clarify their previous (1976) classifications; and the approach
advanced by Halliday (1985:310-313, 1994:330-334) which includes repetition,
synonymy (subsuming: Synonymy, Antonymy, Hyponymy, Meronymy), and
collocation.

In this study, this concept of sense relations in lexical cohesion will be extended
and adapted to account for the claim that cohesion occurs across visual and verbal
modes. To elucidate, in a particular multimodal text there may be the co-
occurrence of an image of some person and his or her name in the verbal aspect.
This would appear to be a reiteration of the experiential (nominal) meaning across
modes, and the sense that is created is one of the representation of the ‘same’
experiential meaning in each mode. In this case it would seem that the sense
relation of repetition, as defined by Halliday and Hasan (1985) in their cohesion
model, would be an appropriate way of describing these kinds of explicit instances
of intersemiotic repetition. It would also seem that this kind of adaptation and
usage can be appropriately utilised with the other sense relations of synonymy,
meronymy hyponymy, antonymy and collocation for the analysis of multimodal

141
text, and this will be demonstrated below. The focus here therefore is on the
interpretation of the meaning (sense) relations which arise as a result of the
intersemiotic visual-verbal links in a multimodal text. It is also on experiential
meaning, which is realised through, amongst other devices, explicit lexical
cohesion. Other cohesive devices such as the referential pronominals ‘he’ or ‘she’,
which in a linguistic (verbal) analysis carry out a ‘co-referential’ function between
the instantial verbal elements may also occur. In the example given above, an
analysis of the ways these implicit co-referential features relate intersemiotically
to VMEs would perhaps be an interesting area for further analysis, but it is not the
specific focus in this particular enquiry.

The main categorisation of cohesive relations to be utilised here therefore will be a

combination of the sense relations of Repetition, Synonymy, Antonymy,
Hyponymy, and Meronymy as advanced by Halliday and Hasan (1985), and the
general category of Collocation (Halliday 1985). These are briefly explained in
the following section with some examples drawn from economics text to illustrate
their various meanings, and will hereafter be interpreted as intersemiotic sense
relations.

One aim of this study is to account for an object or person represented visually
(through visual representational techniques) and verbally (through explicit,
identifying lexical items) in the same multimodal text. Intersemiotic Repetition
involves the repetition of a lexical item which express the same meaning
represented in the VME derived from the visual aspect, and therefore the encoded
experiential meaning. It also includes items that are morphologically distinct due
to inflexion or derivation (op.cit:310). Examples of inflexion may be seen in the
sequence ‘supply’, ’supplies’, and ‘supplied’, and of derivation in ‘supply’ and
‘supplier’. This sense relation is represented in Figure 5.4 below.
VISUAL VERBAL
A visual containing the figure A lexical item in the verbal
of Alan Greenspan. The ⇐ REPETITION ⇒ aspect of the text which
Visual Message Element is occurs as:
glossed as:
Alan Greenspan
Alan Greenspan
Figure 5.4 Intersemiotic sense relation of repetition

142
In Intersemiotic Synonymy the experiential meaning of the two lexical items is the
same or almost the same, and may often be used interchangeably. This can be seen
in the synonymous relations between ‘trade cycle’ and ‘business cycle’, between
‘stocks’, ‘equities’ and ‘shares’, and between market structure definitions such as
‘perfect competition’ and ‘pure competition’. This intersemiotic sense relation is
represented in Figure 5.5 below.

VISUAL VERBAL
A visual containing the figure A lexical item in the verbal
of Alan Greenspan. The ⇐ SYNONYMY ⇒ aspect of the text which
Visual Message Element is occurs as:
glossed as:
Chairman of the Fed.
Alan Greenspan
Figure 5.5 Intersemiotic sense relation of synonymy

In Intersemiotic Antonymy the experiential meaning of the two lexical items is

essentially one of opposition, as in ‘long run’ and ‘short run', ‘increasing returns
to scale’ and ‘decreasing returns to scale’, and the opposition of ‘buyer’ with
‘seller’. This intersemiotic sense relation is represented in Figure 5.6 below.

VISUAL VERBAL
A graphic visual showing the A lexical item in the verbal
degrees of bank rates of ⇐ ANTONYMY ⇒ aspect of the text which refers
interest increases. The Visual to interest rates decreases:
Message Element is glossed
as:
decreases.
increases
Figure 5.6 Intersemiotic sense relation of antonymy

Intersemiotic Hyponymy involves a classification of the cohesive relations

between a general class (termed the Super-ordinate) and its sub-classes (termed
the Hyponyms and Co-hyponyms), and vice versa. This cohesive relation may be
seen in Table 5.3:
Super-ordinate (class) taxes market structures
Hyponyms excise duty monopoly
and income tax oligopoly
Co-hyponyms (sub-classes): VAT perfect competition
company tax monopolistic competition
Table 5.3 Cohesive sense relation - hyponymy

143
This intersemiotic sense relation is represented in Figure 5.7 below.

VISUAL VERBAL
A (caricature) visual showing A lexical item in the verbal
the UK banking sector as a aspect of the text which refers
dinosaur. The Visual Message to a specific bank: Lloyds
Element is glossed as the ⇐ HYPONYMY ⇒ Bank. This is the co-
Hyponym: hyponym:

UK banks
Lloyds.
Figure 5.7 Intersemiotic sense relation of hyponymy

Meronymy involves a classification of the cohesive relations between the whole of

something (termed the super-ordinate) and its constituent parts (termed the
meronyms and co-meronyms). This cohesive relation may also be seen in Table
5.4:

Super-ordinate (whole) Balance of Payments Money Supply

Meronyms capital account M1
and current account M2
Co-meronyms (parts): statistical discrepancy M3
official settlements
Table 5.4 Cohesive sense relation - meronymy

This intersemiotic sense relation is represented in Figure 5.8 below.

VISUAL VERBAL
A table listing the sectors of A lexical item which refers to
the US economy which are one of the specific sectors in
experiencing greater profits the US economy: Car
The Visual Message Element ⇐ MERONYMY ⇒ production. This is the co-
is glossed as the Meronym: meronym:

US economy car production.

Figure 5.8 Intersemiotic sense relation of meronymy

The term Collocation, as used by Halliday (1994:333-334), refers to those lexical

items which "do not depend on any general semantic relationship .... but rather on
a particular association between the items in question - a tendency to co-occur". In
his discussion of this term Halliday suggests that lexical items occurring as
collocations are often associated with one or another particular register or
functional variety of language (op.cit:334). This is the case with lexical items that

144
are of a technical nature and commonly appear only in one kind of text, and it is
true of those words which are not specifically technical in nature but collocate
according to a particular variety of text, such as those found in a recipe book, an
academic journal of literature, and in an economic-journalistic publication like
The Economist magazine.

Intersemiotic Collocation as used in this study therefore involves a classification

of the lexical relations between items which do not necessarily enter into the
semantic relations discussed above such as hyponymy or meronymy etc., but do,
in a general sense, have a tendency to co-occur as collocates in texts of different
kinds in various fields. This category may be defined as the relation (or
association) between lexical items which have a tendency to co-occur due to the
particular configuration of the register variable field. This relation will vary in
strength depending on the particular text. For example the word cutting in a text
on rose gardens may suggest collocates such as ‘secateurs’, ‘shoots’ and ‘off-
cuts’, in an economics text words such as ‘budget’, ‘spending’ and ‘debts’, or in a
text on movies words like ‘scenes’, ‘camera’ and ‘lights’. Thus there is an
expectancy relationship occurring between the lexical items, which is an aspect of
the collocational relationship (Halliday 1985:313). This intersemiotic sense
relation is represented in Figure 5.9 below.

VISUAL VERBAL
A visual showing the rate of A lexical item which can be
growth of GDP figures for recognised as appropriate to the
the US economy over a 10 general topic area of growth in
year period. The Visual ⇐ COLLOCATION ⇒ the US economy:
Message Element is glossed
as:

US growth consumer confidence.

Figure 5.9 Intersemiotic sense relation of collocation

145
METAFUNCTION VISUAL MEANINGS INTERSEMIOTIC COMPLEMENTARITY VERBAL MEANINGS

Variations occur according to the coding orientation. Various lexico-semantic ways of relating the Lexical elements which relate to the
In the Naturalistic coding we can look at: experiential and logical content or subject matter visual meanings. These lexical items
Identification: Who or what are the represented represented or projected in both visual and verbal arise according to:
participants (actor, recipient, goal)? Who or what are modes through the intersemiotic sense relations of:
they interacting with? Are the participants Identification (participants): who or
interacting? (vectors). • Repetition: identical experiential meaning. what is involved in any activity?
Activity: What action is taking place, events,
portrayal, scene, states, types of behaviour (gestures, • Synonymy: the same or similar experiential Activity (processes): what action is
facial expressions, stance, physical moves)? meaning. taking place, events, states, types of
Circumstances: where, who with, and by what means behaviour?
are the activities being carried out (setting, means, or • Antonymy: opposite experiential meaning.
accompaniment)? Circumstances: where, who with, and
IDEATIONAL
Attributes: what are the qualities and characteristics • Meronymy: the relation between the part and whole by what means are the activities being
of the participants? of something. carried out?

In the Mathematical coding we can look at: • Hyponymy: the relation between a general class of Attributes: what are the qualities and
Identification: What are the participants (dependent something and its sub-classes. characteristics of the participants?
or independent variables, physical objects, forms,
formats, symmetrical arrangements). What are they • Collocation: an expectancy or high probability to
interacting with? Participant interaction (vectors). co-occur in a field or subject area.
Relational Activity: What is the relation between the
participants, occurring singly, or in some
combination - is it derivative (rate of change),
comparative, or proportional?
Circumstances: where, what with, and by what
means are the relations realised (setting, means, and
accompaniment)?
Attributes — what are the qualities and characteristics
of the relationships represented - a category which is
difficult to apply to mathematical visuals.

INTERPERSONAL

COMPOSITIONAL

Table 5.5 Ideational intersemiotic complementarity

146
The model of intersemiotic complementarity presented in Figure 5.3 can now be
elaborated into an explanatory Table 5.5 illustrating the potential ways that
ideational intersemiotic complementarity can be realised. This table presents these
ideational intersemiotic features in terms of a range of questions which can assist
the analyst in focussing on the VMEs and identifying the range of semantically-
related verbal features. It should be noted however that not all of the questions
will apply to a multimodal text at any one time. The degree of application of these
questions will very much depend on the type of text  for example, an article
dealing with the latest news in a complex financial issue and utilising a sketch
caricature and a photograph will perhaps reveal a far more complex mix of
messages and usages of these realisations than a straightforward advertisement
focussing on a single product or service.

5.6 The Analysis of Intersemiotic Interpersonal Features

The examination of the intersemiotic interpersonal features of a text with both
verbal and visual components involves, in line with the social view taken in this
study, a look at the ways that relations between the visual and the viewer/reader
are represented. The ways in which the producer and viewer/reader of a
multimodal text are placed socially in relation to each other is important because
this can affect the topic, the ways that it is received, and the ways that it is
interpreted. As Kress and van Leeuwen note,
what gets represented, how what gets represented is represented, as
well as how that is read and used, are all effects of the social place
of producers and viewers of images (or of writers and readers of
verbal text). Viewing an image entails first and foremost, before
anything else has happened, being located in a particular social way
by and in relation to the image (1990:23).

In this socially-determined context, an examination of the ways that the visual and
verbal components in an Economist text realise interpersonal intersemiotic
complementarity can be clarified through an examination of the ways that both
modes seek to address the viewer/readers and to inform them about the text
constructors’ attitudes to the subject matter represented.

5.6.1 The Interpretation of Visual Interpersonal Features

147
As already discussed in Section 5.1, in the lexicogrammatical system, the
interpersonal metafunction is realised by the clause as exchange, and the range of
choices that can be made in the systems of Mood and Modality. These systems
basically deal with speech function, and the speaker’s attitudes, comments and
assessments about that which is being exchanged. To examine the interpersonal
intersemiotic features of an Economist text, the first step is to examine any extant
visuals in terms of viewer address, levels of involvement, power relations and
social distance in the visual Mood system, and in terms of the various modality
markers in the visual Modality system.

As with the ideational aspects of visuals, variations can occur according to the
coding orientation. In the Naturalistic Coding Orientation, these variations occur
along a continua of degrees of incidence:
In the Visual Mood System:
• Address: Gaze at the viewer (direct or indirect).
• Involvement & Power: Perspective - horizontal angle, vertical angle.
• Social Distance: Size of frame (close up, medium, long shot etc.).

In the Visual Modality System:

• Modality Markers: Contextualisation: (full to absence of background).
Degree of representing detail or abstracting detail. Texture, Illumination
(light and shade); Colour saturation/differentiation/modulation.

In the Mathematical Coding Orientation however, there is a more restricted

continuum of degrees of incidence due to the nature of the medium:

In the Visual Mood System:

• Involvement & Power: Perspective - horizontal angle; vertical angle.

In the Visual Modality System:

• Modality Markers: Contextualisation: (full to absence of background).
Degree of representing detail or abstracting detail. Texture, Illumination
(light and shade).

Each of these will now be examined in turn.

5.6.1.1 The Analysis of Visual MOOD - Address (“Visual Speech Acts”)

148
In language the roles in a communicative exchange and the nature of the
commodity exchanged can be combined into the four primary speech functions of
Offer, Command, Statement and Question, which can then be matched by a set of
appropriate responses such as accepting the offer, carrying out the command,
acknowledging a statement, and answering a question (Halliday 1985:68-69). In
considering MOOD in visual communication however, it seems that visuals are
not easily analysed with these categories - they often need verbal support to make
the nature of the speech function clear, as in a visual offer of goods and services
supported by a verbal contact address in an advertisement, or the verbal
reinforcement provided by a printed question to complement a questioning facial
expression (Kress and van Leeuwen 1990:30).

Halliday also distinguishes between the exchanging of goods and services, and the
exchanging of information. In the exchanging of goods and services, the semantic
function of the clause is defined as a PROPOSAL, and in the exchanging of
information, the semantic function of the clause is defined as a PROPOSITION
(1985:70-71). This is an important distinction in language, but since The
Economist magazine does not exchange goods and services but information with
its readers, the focus of this study will be on how intersemiotic complementarity
in The Economist magazine is realised through the ways that its propositions are
addressed to its readers via questions and statements in the exchange of
information. This is therefore concerned with forms of Visual Address (“Visual
Speech Acts”), or the ways that The Economist magazine addresses its readers in
order to exchange (or offer) information. This will be discussed in the ensuing
section.

In determining the speech function of a particular visual, the most important

feature to look for is the presence or absence of visual techniques which directly
address the viewer. In the case of a visual which approximates or reproduces a
naturalistic scene (a Naturalistic visual), the absence of any gaze or facial
expressions towards the viewer indicating a question is being asked (realised by
vector drawn from some point of origin to the viewer’s face), or gestures which
command (realised by for example a pointed finger forming a vector directly to

149
the viewer), or offers of goods to the viewer (realised perhaps by a vector formed
from a hand gesture towards some object in the visual frame), would strongly
suggest that it is offering some kind of information to the viewers. In that case
there would be no vectors which can be drawn from the represented participants
directly towards the viewers; all the vectors for example may indicate participants
within the visual frame, and require the viewer to be cognisant of the ways that
they are interacting with each other. It would therefore be a portrayal or a scene
which the viewer can look at with really no requirement to react to other than to
agree with it, or to either acknowledge or contradict its existence/veracity as a
scene (some elements of which will be discussed in the section on visual modality
markers). This offer of information can be supported by some kind of verbal
support, such as labelling to identify the scene or major represented participants
(as is often the case in The Economist).

In the case of a visual which presents information in a mathematical form (as in a

graph or chart), the techniques for addressing a viewer of a naturalistic image can't
really be utilised. This perhaps goes to the heart of the nature of a visual of this
type, in that its primary function is to address the viewer via techniques which
present information in quantitative forms and showing the quantitative relations
between them. Thus there really is no question that these kinds of mathematical
visuals are offering information, since the represented participants (in other words
the data) can form no other relationship to the viewer(s) than to be simply a
display of numbers and graphic lines interacting with each other to indicate some
inter-related information or data. The viewer is not asked anything, is not
commanded, and is not asked to accept/reject something on offer. The viewer is
however offered information which can be agreed or disagreed with, or
acknowledged or contradicted.

5.6.1.2 The Analysis of Visual MOOD - Involvement

The post-Renaissance development of the Western traditions of visual
representation in art and drawing has used perspective to realise visual point of

150
view. A visual with perspective is one which has an inherent or “built-in” point of
view, where the represented situation and the physical space in which the visual is
viewed, or the frame, is cordoned off through various means by a clear dividing
line between the image and its surrounds, thus marking it off and giving it the
function for the viewer as a “window” or “peep-hole” on reality. Indeed, some
early post-Renaissance artists who utilised perspective used a peep-hole as a frame
to draw on a transparent surface what could be seen in order to guarantee an
unchanging point of observation (Arnheim 1974:283-84). The viewer of a visual
utilising perspective can therefore only see the participants and action from a
particular point of view - if the viewer were to walk around the actual object
portrayed (as in a piece of sculpture or an actual person etc.), there would of
course be a different view available from a different angle, but in a visual the
viewer has no choice but to accept the point of view loaded into it by its
constructor. As Kress and van Leeuwen explain, the “perspectival image compels
the viewer to become, together with the institution that has produced the image, an
‘us’ with respect to a ‘them’ or an ‘it’ - that is with respect to the participants
represented in the image” (1990:32).

The level of involvement by a viewer with a visual is realised by horizontal angle,

which is concerned with the interrelationship between two frontal planes: the
frontal plane of the constructor of the visual and the frontal plane of the
represented participants. These planes can be aligned parallel to each other, or
diverge by forming an angle of varying degrees with each other. A visual can
therefore have either a frontal or oblique point of view, the oblique point of view
being a continuum of obliqueness according to the angle of the divergence
(Montague 1993:1-18). The importance of these distinctions is that they are
suggestive of the level of involvement. The horizontal angle encodes whether or
not the person who has produced the visual, and as a consequence the viewer, is
committed or not in some way to the represented participants. The frontal angle is
a statement of some degree of inclusion between the constructor and the visual,
which is realised by the vanishing points all being placed within the visual frame
(Dondis 1973:60-61). On the other hand, an oblique angle encodes varying
degrees of commitment to the subject or represented participants, stating to

151
varying degrees of intensity the level of inclusion. This can be illustrated by
considering the difference in a viewer’s reaction to two photographs, one where
the subject’s face is facing directly at the viewer (a sense of inclusion which
would be strengthened by a direct gaze), and the other with the subject’s face at
forty-five degrees to the viewer, looking at someone/thing else (very little sense of
inclusion since the vanishing points all occur outside the visual frame). The right-
angled or perpendicularly oblique angle would be suggestive of viewing a scene
with no involvement at all, beyond suggesting that ‘this is a scene” to be observed.

There are some fundamental differences between visuals which attempt to portray
participants in some realistic way, and those which attempt to display numerical or
quantitative representations. The former, in effect, have by their nature the system
of perspective fundamentally encoded within them. The drawer of a visual can, by
the very nature of the represented participants, choose to present them from any
angle. With the possible exception of three-dimensional representations of data,
where it is now possible with many computer programs to rotate blueprints and
mathematical representations, the constructors of a mathematically-based visual
are constrained by the inherent nature of this kind of visual form. This visual can
only really be viewed from an oblique point of view, since it is representing
meaning on a two-dimensional plane. All viewers are forced to view the
information portrayed in an abstract graphic form as an offer of information with
which they can agree or disagree, acknowledge or contradict. Unlike the
naturalistic visuals, into which their constructors can encode various degrees of
involvement through degrees of obliqueness, the mathematically-derived visuals
are restricted to a two-dimensional plane which leaves no other choice for the
viewers than to be an uninvolved observer of information or data. There is really
no way that this type of visual can include the viewer.

5.6.1.3 The Analysis of Visual MOOD - Power Relations

The power relations between the viewers and the represented participants in a
visual are encoded in the vertical angle formed between them, which is also an
important aspect of cinematography, where the viewers of film are required to

152
react to the participants in a particular shot according to whether they are looking
down to, up to, or at eye-level with them. There are therefore three possibilities
here for looking at a visual in The Economist magazine: a high angle, a low angle
and an eye-level angle. The high angle means that the viewers are looking down
on the represented participants, suggestive of a superiority to them or some degree
of their insignificance, a low angle means that the viewer is looking up to them,
suggestive of an inferior position to them or some degree of their magnificence,
and an eye-level angle is suggestive of equality between the viewer and the
represented participants.

The differences in perspective between images portraying naturalistic scenes and

mathematically-based visuals are also of relevance here, in that the former is able
to utilise the three angles according to the drawers' or photographers’ views, but
the latter is still constrained by the two-dimensional nature of its presentation of
information; they form an eye-level angle between the viewer and the represented
participants and therefore do not (and cannot) place the represented participants in
a superior or inferior position. This is in line with the previously mentioned
interpretations of these kinds of visuals as offering information, and as offering
this information from an oblique point of view (and therefore requiring no
involvement).

5.6.1.4 The Analysis of Visual MOOD - Degree of Social Distance

A further aspect which is specific to images and their realistic portrayal of
represented participants is the degree of social distance between the represented
participants and the viewer(s), as realised by the size of frame. As already
explained in the review of Kress and van Leeuwen’s work, in television
production the size of frame is related strongly to how much of the human body is
shown in the visual frame (1990:44). There is the close up, the medium shot, and
the long shot etc. These different kinds of television shots can be paralleled with
the varying distances between people when they talk to each other face to face,
where it can be intimate or friendly (as in a spouse or friend or acquaintance), or
unknown (as in a complete stranger). Contextually-based distances may involve
interaction in specific social and public situations, where a well-known, familiar

153
or unknown public figure is delivering a speech to an audience. These distances
are of course culturally determined, but generally speaking one can apply these
ideas to the meanings encoded in visuals when a close up shot of someone's head
and shoulders, as opposed to the top half of the torso, and as opposed to the full
body within three metres or at over thirty metres or more is utilised. As Kress and
van Leeuwen explain further,
the relation between the human participants represented in images
and the viewer is once again an imaginary relation. People are
portrayed as though they are friends, or as though they are
strangers, regardless of their actual relation to the viewer. Images
allow us to imaginarily come as close to public figures as though
they were our friends and neighbours - or to look at people like
ourselves as strangers (op.cit:45).

5.6.1.5 The Analysis of Modality in the Visual Component

The analysis of Modality in the visual component of a text involves an
investigation of viewer acceptance of the truth or credibility of what is
represented, and how this is realised via the use of various visual modality
markers such as contextualisation, the degree of representing or abstracting
detail, and the use of texture, illumination, and colour
saturation/differentiation/modulation. As discussed above, when information is
offered in both naturalistic and mathematically-based visual modes, it can be
affirmed, or denied by the viewer(s). Also, in language the truth or credibility of
what is represented by a speaker or writer is expressed through the use of
modality at the clause level, and the polarity between the affirmation and denial
of this offered information is expressed in terms of whether something is or is
not, or whether it is real or unreal. In between these two extremes there are
other possibilities which express degrees of certainty or uncertainty, where
perhaps something could happen, or of usuality, where something might
sometimes happen, but not always. In the visual semiotic code visuals can also
be interpreted in terms of the truth, credibility, and probability of what they
represent to the viewer(s), and the information they offer can also be affirmed or
denied according to whether something is or is not, or real or unreal, as well as
whether other possibilities exist which can express degrees of certainty or
uncertainty (where perhaps something could happen), or of usuality, (where

154
something might sometimes happen, but not always). In both modes there is the
sense that these attitudes can be expressed through selections made along a
continua of possible choices.

The interpretation of the degrees to which a visual is considered to be real or

unreal, credible or incredible, possible or impossible depends in the first instance
on its coding orientation. So far the discussion of visual representation in The
Economist magazine has generally distinguished between those visuals which are
portraying realistic or naturalistic scenes, and those which derive from a
mathematical source. This distinction, as well as the notion of coding orientation,
is a useful one which will be developed and explained more fully in Chapter Six,
which deals with developing a visual typology of The Economist's visuals. In
terms of the naturalistic type of visuals, it is proposed that there is a continuum of
naturalistic visuals which are considered to be real/unreal or possible/impossible
depending on the degree of accuracy of their representation of reality. This
continuum (shown below in Figure 5.10) will also be outlined in Chapter Six as a
categorisation for describing the various forms of visuals which occur in the
naturalistic coding orientation in The Economist magazine.

colour black/white sketch sketch line

photograph Î photograph Î drawing Î caricature Î sketch

highest median lowest

modality Î modality Î modality

Figure 5.10 Modality in the naturalistic visual continuum in The Economist magazine

This coding orientation illustrates the varying degrees to which these visuals (and
by default their drawers) attempt to portray 'natural' images (visuals which the
members of a particular cultural setting would agree to be a recognisable form of a
representation of reality, as viewed by the human eye).

As will be seen from the analysis given in Chapter Six, the sketch caricature form
of visual is the most common type of naturalistic visual to be found in The
Economist magazine. This form of visual attempts to represent familiar objects,

155
entities, scenes, characters or actions (real or fictional) which are easily
recognisable to the viewers, but are abstracted via caricature. These are not totally
accurate representations, but are a stylistic drawing form in which the main
features of the represented participants have been emphasised to present the
represented participants from a particular attitudinal viewpoint, ones which the
drawer(s) hold and wish to convey to any potential viewer(s). As such, the visual
modality of a caricatured scene in terms of representing what is real or credible is
relatively low. What is presented is not a true representation of reality, as a colour
or to a lesser extent black and white photograph or sketch drawing may, but a
whimsical version of it. In this type of visual there is in a sense more freedom for
the viewer(s) to agree or disagree with the portrayal, to acknowledge it, or to
contradict it. It is therefore a ‘suggested’ or ‘loaded’ interpretation of reality, and
as such carries a lower modality in terms of its representation of ‘truth’ than
photographs or sketch drawings. A viewer of a photograph cannot argue with it as
a representation of reality beyond claiming that it was doctored or taken in
deliberately unflattering circumstances (a high vertical angle for example). The
absence of colour and the abstraction away from many physical details in the
sketch caricature which could have been included, such as additional facial and
physical features, and landscape features such as plants, trees etc., also adds to this
relatively low modality. The emphasis in this type of naturalistic visual is not on
accuracy or realism as in sketch drawings or colour and black/white photographs,
but on presenting a subject in such a way that the main features are caricatured,
and so drawing attention to the attitudes that the drawers wish to express.

Mathematical visuals are on the other hand instances which are derived from the
scientific/technological coding orientation. In this orientation the level of
credibility

156
 1. 2. 3. 4.
highest modality high to median median to low lowest modality
modality modality

Figures: Figures: Figures: Figures:

• no colour. • no colour. • some colour in the • fully differentiated
• high degree of • complex headline, data, use of colour
complexity in the presentation of figures, or line within and around
presentation of variables and graphs. the visual.
variables and relationships • less complex • data presented in
relationships shown presentation of non-mathematical
shown. • one or more variables and form e.g. stick-
• high degree of mathematical relationships figures, symbols.
complexity in the relationships shown. • not severely
mathematical Î shown (as in three Î • one or two Î decontextualised
relationships line graphs on one mathematical in that colours or
shown. set of x,y axes). relationships some aspect of the
• decontextualised, • decontextualised shown. data may ‘bleed’
set apart from the by a framing • decontextualised into the verbal
verbal severely device like a thick by a framing aspect of the text.
e.g. Placed on a black border. device like a black • a single
separate page or in • a verbal headline border, or an even relationship
an appendix. labelling the clear space around between variables
• no interpretative quantitative topic the visual. shown.
verbal support of the visual, but • verbal headline • visually salient
beyond a small not interpreting it. interpreting the verbal headline
sub-heading data shown - interpreting the
indicating the indicating its data shown -
topic of the data significance. giving a very
(the figures speak strong ‘slant’ to its
for themselves) messages.

Figure 5.11 Mathematical visual modality in The Economist magazine

157
does not rest on the degree to which something is accurately represented, but on
how efficiently it represents data in a quantitative form. What is ‘real’ in this
coding orientation is dependent on how much the visual conforms to the accepted
scientific and mathematical norms for presenting data in a visual format in the
various contexts in which it may occur. This orientation therefore focuses on how
phenomena are counted, weighed and measured, and how these measured features
interact with each other in an abstracted, two dimensional way. What is ‘real’ can
be looked at in terms of whether it utilises these accepted norms of mathematical
interpretation of interacting data. There is in a sense a continuum of
scientific/technological acceptability which, depending on the degree of use of
certain kinds of scientific/technological visual techniques, places the viewer(s) in
various attitudinal positions. These attitudes will vary across different contexts (or
disciplines), and the means of graphic representation in some of the ‘hard’
sciences like physics and chemistry will differ to those acceptable to a social
science-based discipline like economics or commercial studies.

As will also be seen in Chapter Six, the kinds of mathematical visuals used in The
Economist magazine are largely line graphs, bar graphs, pie charts, and tables.
These kinds of visuals vary in the amount of data they show and the ways that are
deemed to be appropriate by the members of the discipline for whom they are
produced. Since The Economist magazine is produced for a wide audience of
professional and non- professional readers in the eco-financial sphere who are
both initiates and non-initiates with mathematical presentation techniques,
mathematical modality in The Economist could be interpreted by making reference
to a continuum of mathematical modality markers as suggested in Figure 5.11.

According to the mathematical visual modality continuum in Figure 5.11, one

would expect that mathematical visuals of highest modality would be found in
professional economics journals, given the absence of colour and the high degree
and complexity of information required. The types of visuals with the lowest
modality would on the other hand be expected to be found in a very introductory
economics textbook, or a general newspaper or magazine that dealt with
economics issues in such a way that would be palatable and understandable for the

158
uninitiated. Thus, one would expect that the kinds of mathematical visuals utilised
in The Economist magazine could most likely be identified as being placed
approximately between 2 and 3 (median modality) because they typically combine
elements of both these descriptions (see Chapter Six on this for a fuller
explanation).

5.6.2 The Interpretation of Verbal Interpersonal Features

In interpreting verbal communication in terms of the interpersonal metafunction,
one is also looking at the clause in terms of an interactive event in which the
speaker, or writer and audience are involved (Halliday 1985:68). The
interpersonal metafunction in the lexicogrammar is realised by the clause as
exchange and the range of choices that can be made in the systems of Mood and
Modality, which are basically concerned with speech function, speaker attitudes,
comments and assessments.

5.6.2.1 The Analysis of MOOD in Verbal Propositions

In the SFL model, the interpersonal aspects of communication are covered in
terms of how the roles in the exchange and the nature of the commodity
exchanged can be combined into the four primary speech functions of OFFER,
COMMAND, STATEMENT and QUESTION, which can be matched by a set of
appropriate responses: accepting the offer, carrying out the command,
acknowledging a statement, and answering a question (Halliday 1994:68-69).
When information is exchanged in an interaction between a speaker/listener and a
writer/reader, as it commonly is in The Economist magazine, it is the MOOD
element in the clause which carries the components of information that are dealt
with in the exchange, and which realise the speech function. In lexicogrammatical
terms this is realised through the choices made in the Mood structure of the clause,
in terms of the ordering of the Subject, Finite, Predicators, and various aspects of
the Residue.

In relation to Visual Address, in The Economist magazine the main focus is the
exchange of information between the magazine and its viewer/readers. Given this,
the relevant grammatical category used in the exchange of information is the

159
indicative. Within this category, the usual expression of a statement is the
‘declarative’, and the usual expression of a question is the ‘interrogative’
(op.cit:74). As discussed, for both the order of the elements is significant, in that:
• The order Subject before Finite realises the ‘declarative’.
• The order Finite before Subject realises ‘interrogative’, of which there are two
further distinctions: there is the yes/no interrogative for polar questions (the
response is either a yes, no or a maybe), and there is the WH-interrogative for
content questions (where the answer is concerned with who, what, which,
why, how, and when).

Having established that The Economist magazine is basically concerned with the
exchanging of information and that this is realised by the ordering of the Mood
elements in the indicative grammatical category, the next stage therefore is to
check through the verbal aspect of an Economist magazine text for the ways that it
addresses its readers in the act of exchanging information. As already mentioned,
this will basically be in terms of whether the readers are addressed via being asked
questions, or by being given statements; this means that the verbal aspect of the
text will be examined in terms of its Subject-Finite ordering.

5.6.2.2 The Analysis of MODALITY in Verbal Propositions

In the lexicogrammar, it is the function of Modal Adjuncts to add interpersonal
meaning to the clause (ideational meaning is added by Circumstantial Adjuncts
and textual by Conjunctive Adjuncts). Modal Adjuncts do this by influencing the
Mood element through the addition of some kind of modification to the
Subject/Finite relation, or more indirectly by simply adding some attitudinal
expression or attempting to control the nature of the interaction. Halliday
(1994:81-83) identifies two principle types of Modal Adjunct which express
interpersonal meaning: Mood Adjuncts (which are most closely associated with
the meanings in the Mood system and therefore tend to occur close to the Finite
operator), and Comment Adjuncts (which do not have a direct impact on the Mood
constituent of the clause but affect the speaker’s attitude to the proposition as a
whole). These are summarised with

160
 TYPE ADJUNCT MEANING PRINCIPLE EXAMPLES
polarity not, yes, no, so
polarity probability probably, possibly, certainly, perhaps, maybe
& usuality usually, sometimes, always, never, ever, seldom, rarely
modality readiness willingly, readily, gladly, certainly, easily
obligation definitely, absolutely, possibly, at all costs, by all means
time yet, still, already, once
Mood typicality occasionally, generally, regularly, mainly,
temporality for the most part

obviousness of course, surely, obviously, clearly

intensity just, simply, merely, only, even, actually, really, in fact
mood degree quite, almost, nearly, scarcely, hardly, absolutely, totally,
utterly, entirely, completely

opinion I think in my opinion, personally, to my mind

admission I admit frankly, to be honest, to tell you the truth
persuasion I assure you honestly, really, believe me
entreaty I request you please, kindly
presumption I presume evidently, apparently, no doubt, presumably
desirability how desirable? (un)fortunately, to my delight/distress, regrettably,
hopefully
Comment reservation how reliable? at first, tentatively, provisionally, looking back on it
validation how valid? broadly speaking, in general, on the whole, strictly
speaking, in principle
evaluation how sensible? (un)wisely, understandably, mistakenly, foolishly
prediction how expected? to my surprise, surprisingly, as expected, by chance

Table 5.6 The principle types of Modal Adjunct (from Halliday 1994:49, 81-83).

161
examples in Table 5.6. Each will be discussed in turn.

Within the category of Mood Adjunct, Modality is used by speakers to express

their attitudes towards the arguable point contained in a clause (i.e. the Mood
constituent), whether they be Propositions or Proposals. When information is
exchanged, the Proposition is something that can be argued about in terms of
whether it is or is not, real or unreal, and whether it can be agreed or disagreed
with (polarity). However, many communicative interactions are not expressed in
such extremes, for participants in interactions often temper their communication
by projecting something about their attitude to the message, or making some kind
of comment about its relevance, reliability, interest etc.

The system of Modality allows for the information being exchanged to be adjusted
in various ways and to express these attitudes and judgements. When Modality is
used to argue about the probability or frequency of a proposition, it is referred to
as Modalisation, and basically relates to the expression of two kinds of meanings
(Halliday 1994:88-92):
• probability: judgements are expressed as to the likelihood of something
happening or being through the usage of certainly, surely, probably, maybe,
perhaps, etc.
• usuality: judgements are expressed as to the frequency with which something
happens or is through the usage of always, often, usually, typically,
occasionally, rarely, seldom, once; ever, never, etc.

The meanings produced through this form of modalisation, which basically means
that the speaker is being tentative about his or her proposition, can be expressed in
the clause through the usage of:
• the choice of a Finite modal operator such as in ‘The Chancellor might have
raised interest rates’.
• the use of Mood Adjuncts which express probability, certainty such as ‘The
Chancellor was possibly thinking of raising interest rates’.
• the use of Finite modal operators and Mood Adjuncts together as in ‘The
Chancellor might possibly have been thinking of raising interest rates’.

Halliday distinguishes and classifies these modalisation resources into three

degrees or values of modal judgement: high, median and low. Some of the

162
common lexical resources used to realise these can be summarised in Table 5.7
below.

Value: Low Value: Median Value: High

Verbal Operator may will must

Adjective possible probable certain

Adverb possibly probably certainly

Noun possibility probability certainty

Table 5.7 Degrees [Values] in Modality (1994: 337-339)

Speakers not only express their judgements about the likelihood, frequency, or
generality of propositions, they also make Comments about them. Comment
Adjuncts are used by speakers to add an expression of attitude and evaluation to
the proposition as a whole, but they do not say anything about the probability,
generality or frequency. They typically occur in the initial position in a clause or
directly after the Subject; however Halliday claims that because these kinds of
adjuncts are commenting about the entire clause rather than its individual
Mood/Residue elements, they should be considered as operating outside that
structure altogether. In fact they tend to occur at points in the clause which are
significant for the textual organisation (op.cit:83). Some typical examples of
Comment Adjuncts include:
1. ‘Unfortunately, the Reserve Bank raised interest rates today.’ [Desirability]
2. ‘The Reserve Bank, understandably, raised interest rates today.’ [Evaluation]
3. ‘The Reserve Bank raised interest rates today, as expected..’ [Prediction]
4. ‘Tentatively, the Reserve Bank could raise interest rates today.’ [Reservation]

Both types of Modal Adjuncts, the Mood Adjunct and the Comment Adjunct,
relate to varying degrees to the system of Mood, and therefore to the statements or
questions realised in the Mood structures in Propositions. The Mood Adjunct is
obviously closer to the Mood elements than the Comment Adjunct, but they do
form a continuum, and as in Halliday’s interpretation, both will be treated as being
part of the Mood constituent of the clause and as important vehicles for adding
interpersonal meaning to the clause (loc.cit.).
There remains one further way that interpersonal meaning can be realised in
Propositions. This is through the use of what Halliday refers to as the attitudinal

163
Epithet 1994:184, Matthiessen 1995:692). These primarily lexical items operate at
the next rank down in the lexicogrammar: that of the grammar of groups and
phrases. Epithets are important in the nominal group in both experiential and
interpersonal terms. When they are used in experiential terms, they are seen as
being ‘objective’, where they simply describe an entity as in ‘I bought a small
share issue’, or define it as in ‘I bought the small share issue’. In these examples
the distinction between describing and defining is realised by the ‘a/the’ contrast.
When Epithets are used to add interpersonal meaning however, they are usually
referred to as ‘subjective’ adjectives which express the speaker’s subjective
attitude towards the Head (noun) by providing information about it in either
appreciative terms as in ‘it was a wonderful day of trading’, or in pejorative terms
as in ‘it was an appalling day of trading’ (Downing and Locke 1992:451-453). It
is this latter attitudinal Epithet which will be examined in The Economist
magazine text.

To summarise, the next stage of an analysis of a multimodal text in terms of its

verbal Propositions therefore is to check through the verbal aspect of the text for
the ways that the readers are addressed in terms of the speech function realised in
the exchange of information. This will involve an examination of the text’s Mood
features in terms of the ordering in the main Subject/Finite constituent of the
clause complexes, to ascertain whether the Propositions the readers are being
addressed with take the form of statements, or questions, or some degree of both.
Once this is done, the next step is to see how or in what ways the Propositions are
treated in terms of their Modalisation features: where there is argument about the
probability or frequency of a Proposition, and any comments that may also be
made. In addition to this, the next step would be look at the ways that the
speaker’s attitudes are realised through the use of such features as attitudinal
Epithets. Once these analyses are completed, it will then be possible to examine
the multimodal text in terms of how both the verbal and visual elements address
their reader/viewers, as well as how the mode-based attitudes are expressed to
realise interpersonal intersemiotic complementarity.
5.6.3 The Analysis of Interpersonal Intersemiotic Complementarity

164
In the same ways that the relevant features of the contexts of situation and culture
along with aspects of a text’s intertextual history inform the interpretation of a
multimodal text’s intersemiotic ideational features, these features can also play a
part in the interpretation of a text’s intersemiotic interpersonal characteristics.
Once the main ways that the viewer/readers have been addressed by both modes
have been described, along with any attendant Modality features, it is then
possible to examine these features in relation to how they complement each other
to produce an interpersonally coherent multimodal text.

The aim here therefore is to identify how intersemiotic complementarity is

realised in the various ways that the reader/viewers and the text are
intersemiotically related through the systems of MOOD (Address via Offers,
Commands, Statements, Questions) and MODALITY (Attitude re something as
real or unreal, true or false, possible or impossible, necessary or unnecessary,
and other attitudinal positions). It is proposed that intersemiotic
complementarity can be realised semantically through the intersemiotic
interpersonal relations of:
• Reinforcement of address: an identical form of address, where the visual and
the verbal modes, utilising the methods inherent in their modes, both ask
questions or make statements in the exchange of information. When this
occurs, the fact that both modes address the reader/viewers in the same way
leads to them to interact in such a way that a single, coherent multimodal text
is the result. Reinforcement of address therefore realises intersemiotic
complementarity. One could well ask what is the opposite or alternative to
Reinforcement of Address — but since there really is no opposite form of
address to a statement or a question, except the logical opposite of no-
question or no-statement, there seems on the surface that there is no category
of intersemiotic complementarity which can account for situations where one
mode asks a question and the other makes statements, except perhaps
Differential Address; but this is unrevealing, since there is still the sense that
the two modes are complementary in some way. To account for this perhaps
intersemiotic complementarity could be viewed in terms of two potential
options: one in terms of reinforcement of address (or roles) as explained
above, the other as a realisation of some kind of “correspondence”
relationship or reciprocal association between the modes in the text, as in an
advertisement where a photograph may directly address viewers with a
question and the accompanying writing mostly makes a series of statements
which answer the question(s) raised (see Royce 1998b on this in an analysis
of an advertisement from The Economist magazine). The terminology of
Barthes (1977) may be helpful here, where he refers to the ‘relay’
relationship in image-text relations (although his usage of the term restricts

165
 this kind of relationship to the speech balloons used in comics, and the
various modes involving moving images such as TV or film, where the
dialogue works in concert with the images).

• Attitudinal congruence: the same or a similar kind of attitude, where the

propositions presented by both modes are treated in the same or similar
attitudinal fashion. Here intersemiotic complementarity can be realised via
the Modality Markers in visual(s) and the Modality features of the clauses
projecting the same or similar attitudinal meanings in concert.

• Attitudinal dissonance: an opposite or ironic attitude, where the propositions

presented by both modes are treated in different ways attitudinally. Here the
attitude presented in the visual for example can be one which is opposite or
ironic in relation to the attitudes expressed in the verbal aspect of the text. In
these situations the intersemiotic complementarity produced by the ideational
aspects of the text in a sense ‘anchors’ the reader/viewers’ interpretation —
they see that ideationally the multimodal text is a coherent one, but that
interpersonally there is opposition produced by the differing attitudes. This is
a recognisable rhetorical device in literature, where irony is used for dramatic
or even whimsical effect. In these senses therefore intersemiotic
complementarity is realised via the obvious dissonance between the attitudes.

The model of intersemiotic complementarity presented in Figure 5.3 can now be

elaborated into an explanatory table illustrating the potential ways that
interpersonal intersemiotic complementarity can be realised. This table presents
these interpersonal intersemiotic features in terms of a continuum of options
which may be simultaneously selected in both modes. It should be noted however
that not all of the continuums will apply to a multimodal text at any one time, and
in fact one or two may dominate the ways that each mode attempts to address the
viewer/readers and to project various attitudes. The range of potential visual and
verbal realisations of these interpersonal, intersemiotic complementarity relations
are outlined in Table 5.8 (following page).

5.7 The Analysis of Intersemiotic Compositional Features

The examination of the intersemiotic compositional features of a multimodal text
involves an examination of those features of the page’s layout which facilitate
intersemiotic complementarity between the visual and verbal modes. As
mentioned previously, the term ‘composition’ has been bracketed with Halliday’s
textual

166
METAFUNCTION VISUAL MEANINGS INTERSEMIOTIC COMPLEMENTARITY VERBAL MEANINGS

IDEATIONAL

Variations occur according to the Coding Various ways of intersemiotically relating the reader/viewer Elements of the clause as exchange
Orientation. In the Naturalistic Coding - it is a and the text through MOOD (Address via offers, commands, which relate to visual meanings.
continua of the use of: statements, questions) and MODALITY (Attitude re These arise according to:
Address: something as real or unreal, true or false, possible or
Gaze at the viewer (direct or indirect). impossible, necessary or unnecessary, and other attitudinal The MOOD element in the clause
Involvement & Power: positions) through the intersemiotic relations of: realising speech function (offer,
Perspectival aspects of the role of the horizontal command, statement, question) as
angle, and the vertical angle. • Reinforcement of address: an identical form of address. determined by the Subject/Finite
Social Distance: relationship.
Size of frame (close up, medium, long shot etc.) • Attitudinal congruence: a similar kind of attitude.
affecting the affinity between viewer and image. The MODALITY features of the
Modality Markers: • Attitudinal dissonance: an opposite or ironic attitude. clause which express attitudes about
Contextualisation (full background to a complete the Proposition presented in the
INTERPERSONAL absence of background). clause. Modalisation which indicates
Degree of representing detail or abstracting detail. views on the possibility, probability,
Texture and certainty of the Proposition, as
Illumination (light and shade). well as the use Comment Adjuncts.
Colour saturation showing degree of reality. Also the use of attitudinal Epithets in
the form of subjective adjectives.
In the Mathematical Coding -it is a continua of the
use of:
Involvement & Power:
Perspective - horizontal angle; vertical angle.
Modality Markers:
Contextualisation: (full to absence of
background).
Degree of representing detail or abstracting detail.
Texture, Illumination (light and shade).

COMPOSITIONAL

Table 5.8 Interpersonal intersemiotic complementarity

167
because it captures more fully the sense of two modes interacting with each other
to project meaning coherently on the page. Also, the concept of ‘composition’
refers not only to systems of layout, but also to a text’s positioning within a whole
magazine or book, as well as a particular section or department.

This section will, in line with previous sections, examine the ways that both the
visual and verbal modes project their meanings to produce compositional
intersemiotic complementarity. The visual mode will firstly be examined in terms
of general principles of layout and composition, principles which would be
considered as the conventional wisdom when applied to codes of graphic design.
The verbal mode however will not be examined at the clause level in terms of
Halliday’s analysis of Theme-Rheme, but will be treated as the ‘body copy’; in
other words as an orthographic whole realised by many of the same compositional
principles used in the visual mode. These principles organise the verbal element as
a coherent piece of orthography, which simultaneously interacts with the visual
element(s) placed on the same page. This analysis of composition will adapt and
extend the work of Arnheim (1974, 1988) on pictorial design and visual (gestalt)
perception, Dondis (1973) on the fundamentals of visual literacy, Uspensky
(1973) on the poetics of composition in classical art, Kress and van Leeuwen
(1996) on compositional structuring principles in images, O’Toole (1994) on
composition in displayed art, and especially White (1982a, 1982b, 1983, 1984,
1988, 1990,) on graphic design principles and techniques for page layouts in
magazines.

5.7.1 The Interpretation of Visual Compositional Features

As discussed in the review on composition in Chapter Four, Kress and van
Leeuwen utilise the work of Arnheim (1974, 1988), Dondis (1973) and Uspensky
(1973) for describing the kinds of meanings that are compositionally organised in
visuals. In The Grammar of Visual Design (1996), the structuring principles they
previously discussed in the earlier Reading Images (1990) version are conflated
into three interrelated systems which relate the “representational and interactive
meanings of [a] picture to each other” (1996:183). These principles were:

168
a) INFORMATION VALUE: The placement of the elements (participants and
syntagms that relate them to each other and to the viewer) endows them with
specific informational values attached to the various ‘zones’ of the image:
∗ left and right: the left approximates the well-established, known,
understood, implicitly held view, or the Given, and the right approximates
that which is contestable, to-be-established, presented as not yet known, to
be agreed upon, or to be made explicit, or the New.
∗ top and bottom: where the top of a visual is the space of the 'ideal' or most
highly valued, and the bottom is that of the 'real' or less highly valued.
∗ centre and margin: the centre means that it is represented as the nucleus of
the information on which all the other elements are in some sense
subservient, and the margins are ancillary, dependent elements.

b) SALIENCE: The elements (participants and representational and interactive

syntagms) are combined to attract the viewer’s attention to different degrees by
assigning 'visual weight', as realised by such factors as:
∗ placement in the foreground or background: objects or entities placed in the
foreground are visually more salient than those in the background, and
elements which overlap others are more salient.
∗ placement in the visual field: the aspect of visual ‘weight’ - objects are
‘heavier’ when close to the top, and ‘heavier’ when placed on the left.
∗ relative size: larger objects are more easily noticed by the eye that smaller
ones.
∗ contrasts in tonal value: areas of high contrast such as black borders on
white spaces are higher in salience than a grey-shaded, less distinct border.
∗ colour contrasts: the contrasts between highly saturated colours and softer
muted colours, or the contrast between primary red, white and blue.
∗ differences in sharpness: objects are more clearly seen because their features
are in sharp focus and are more easily noticed by the eye than those which
have their features less sharply focused.

c) FRAMING: The presence or absence of framing devices disconnects or connects

elements of the image, signifying that they belong or do not belong together in
some sense. This is realised by a range of elements:
∗ by drawn in frame lines and borders.
∗ by variations in visual shape or volume, as in a column in a building or in a
scene which may divide up the elements portrayed into sectors.
∗ by discontinuities of colour hue or saturation.
∗ by the use of empty space.
∗ by vectors formed by the arrangement of abstract graphic elements: besides
being lines which can be drawn from or form a part of the represented
elements in a visual, as realised by the direction of gaze, arms, and the
structural features of objects, vectors may also be used where the eye is

169
 guided away from the most to the next salient element (in terms of size or
colour saturation etc.).
∗ by vectors formed by the differences in size or volume in co-occurring
images, and the angles created by skewed elements.

Kress and van Leeuwen also state that these three structuring principles of
information value, salience and framing apply not only to individual visuals but
also to layouts, or what they refer to as composite visuals, or “visuals which
combine text and image, and perhaps, other graphic elements, be it on the page or
on a television or computer screen” (loc.cit.). As shall be outlined below, these
compositional principles are similarly useful in explicating how intersemiotic
complementarity is realised in compositional terms in magazine multimodal texts;
but first the ways that the verbal aspect of a multimodal text will be interpreted in
this study will be outlined.

5.7.2 The Interpretation of Verbal Compositional Features

As mentioned above, the verbal element will be referred to and treated as the body
copy, or as an orthographic ‘visual’ whole which is organised, in a similar way to
the ways that visuals are compositionally organised, through the structuring
principles of information value, salience and framing. These structuring principles
in the verbal mode are realised through the use of various typographical
conventions which are in general use in many magazines of the same type as The
Economist magazine. These orthographic compositional or layout techniques are
generally referred to as principles of typography. Typography is defined by
Watson and Hill as
the art or style of printing, integrally allied to visual design …… [wherein]
the choice of typefaces and the marrying up of text with graphics, both to
arrest attention and to please the eye, lend force to the old adage that it is
not only what you say but how you say it that counts (1997:236).

The importance of typography is also highlighted by White, who suggests that

a coordinated system of typography is a complex aesthetic and functional
calculation requiring the balancing of a number of factors. If it is well
worked out, it becomes a basic — and important — visual tool as well as a
constant definer of the magazine’s personality (1982b:84).

These structuring principles of information value, salience and framing, and how
they apply to the body copy will be discussed below.

170
In terms of Information Value, the relative placement of typographic elements
endows them with specific informational values attached to the various ‘zones’ of
the body copy. Here the left and right orientation in English is important. The
space on the page is conventionally seen as consisting of two parts or zones —
there is the zone of the outer margins which must be kept clear of type, and the
zone of the “live matter”, or the area within the page margins where the type is to
be placed (White 1988:81). In the lexicogrammar, the culturally-determined left to
right reading path in English works textually to realise the Theme-Rheme
ordering in the clause. In this the well-established, known, understood, implicitly
held view typically occurs in the initial, or left, position of the clause, and that
which is contestable, to-be-established, presented as not yet known, to be agreed
upon, or to be made explicit occupies the right. This left-right orientation can
work in the whole page space as well, and realises what Kress and van Leeuwen
(1996:186-187) have referred to as Given-New ordering in composition (see
Chapter Four on Information Value). On the page, the copyfitting of columns (in
magazines it has typically been the two-column and three-column breakup), the
selection of line length (justified or ragged, noting that the longer the line the
harder it becomes to read), and the use of runarounds (fitting or shaping the text
type around a visual, the degree of deformation and line length working to disrupt
or improve the reading path), all work in an integrative way to facilitate a reading
path of left-right orientation, and to potentially realise a Given-New ordering.
Associated techniques which influence information valuations include the use of
line spacing (or leading), where the increased difficulties in reading longer lines
can be compensated by increasing the space between the lines, and devices like
numbering or lettering, which allow for the listing of information and facilitates
faster scanning.
There is also the importance of placement in terms of the top and bottom of the
page, where the top of the page is the space of the 'ideal' or most highly valued,
and the bottom is that of the 'real' or less highly valued. This valuation is
specifically relevant to the placement of headings and sub-headings, blurbs,
boldface lead-ins, by-lines, and captions, which are typically located in the top
(and left-hand) part of the page (White 1988:93).

171
• Headings in magazines are usually referred to as the ‘Headline’, incorporating
its function to catch or attract the reader’s attention, as well as to announce the
topic

• Blurbs usually summarise or expand the meaning of the headline, explain its
significance to the reader, and generally operate to attract the reader to the
article; they are thus placed in such a way that they can be easily seen and
understood.

• Boldface lead-ins are pieces of text which are designed to draw the reader into
the article — they start the article off and are presented, in their most effective
form, as complete sentences which stand alone in terms of making sense and
setting up the reader’s interest.

• Captions are provided to ensure that the reader will interpret a visual in the
way that the author of the article intends, and commonly they are placed at the
bottom of the visual in ragged right format.

• By-lines let the readers know who or which reporting bureau has prepared the
article, and can be placed anywhere on the page, typically above or below the
headline, or just after the blurb. In The Economist magazine they are almost
invariably found below the headline and sometimes after the blurb in the form
of a statement which gives the city bureau from which the article has been sent
(never the names of the reporters).

The ways that these features work compositionally is also related to the left-right
orientation of readers. As White points out, these features should be placed on the
page where they are most likely to be noticed. The left-right orientation and the
tendency to read from top to bottom therefore means that each feature should
generally be placed in “the first place we look on seeing a new page [which] is the
top-left corner (northwest)”, and that later “our eyes [can] travel downward in a
diagonal direction (toward the southeast), searching for whatever else might be of
interest” (op.cit:97). Thus each of these features is accorded a specific status in
terms of its top-bottom (and left-right) placement on the page.
The aforementioned division of space on the page into the zones of the outer
margins, and the “live matter” is also relevant in terms of the ways that the centre
and margin operates to balance the composition of a page. Besides being sized
according to a magazines’ binding conventions and practices, the margins are
generally used to give a publication a ‘special look”. According to White, the
“conspicuous consumption of empty space” (i.e. the use of deep margins) can be
used to “add a touch of class as well as individuality”, with the empty space

172
around the type making it “appear more valuable”, and working to “give a little
space for the eye to rest [which] is appreciated by the reader”. The space of the
margin is not to be wasted by being crammed with the type, but is viewed as “the
velvet cushion setting off the jewel” (op.cit:81-82). The zone of “live matter” on
the page implies of course that this is the centre or focal point of the page as a
whole (not as a single point, but in terms of the central area of influence). This
centre-area can be divided up in many ways, but as pointed out above, in
magazines it is usually into the two and three column make-up. The centre area
where the columns are arranged, as the area of presentational influence, is the part
of the page to which all the other elements are in some sense subservient. In
typographical terms the margins and even the visuals, while important, are
ancillary, dependent elements. As White asserts rather pointedly,
the typography is the fabric of the publication. It is seen on all the
pages and, as such, it is the fundamental raw material you can
control. Pictures, charts, headlines, logos, and all the rest of the
embellishments are additional to the typography. You can have a
product without the aforementioned elements, but you cannot
publish without words, and that means type (1983:42).

In terms of Salience, various typographic elements are utilised to attract the

viewer’s attention to different degrees by their ability to assign 'visual weight'.
Here “tone of voice typography” is of prime importance. Type can be likened to
human speech in that speakers can speak more loudly for emphasis, or vary the
voice pitch or tone to modulate up and down a high or low scale. In White’s
(loc.cit.) view, the volume or loudness of a voice is translated in typographical

terms to the size of the type, where ‘shouting’ can be realised by the use of large

letters (this has its counterpart in Email discourse and IRC chat lines where it is
considered rude to ‘shout’ by using capitalisation). Typographical salience here is
thus realised through relative size, where larger letters are more easily perceived
by the eye than smaller ones — the larger the type, the more easily it is noticed by
the reader, as in headlines.

Changes of pitch or voice tone can be (tentatively, as White does point out)
translated into varieties of boldness, where lightface can be used for a ‘high

173
voice’, and extra-bold for a ‘low’ voice (loc.cit.). The editorial judgement
involved here is significant; that which is deemed important should be
emphasised, and what is merely background or unimportant should be treated in a
less emphatic way. Thus contrasts in tonal value are often utilised, where areas of
high contrast such as black letters on white spaces are higher in generating
salience than grey-shaded, less distinct letters.

Salience is also generated by contrasts in type, where different kinds of typeface

can be used to express differing senses, ranging from the ‘working’ style of the
more traditional typefaces (such as Courier, Arial, Times Roman etc.), to

the use of more ‘exotic’ or decorative typefaces (such as Gaelic, Handwriting

etc.). Other ways of creating emphasis in type, all of which White (1988:38)
recommends to use sparingly, are the use of underlining (or underscoring) an

important word, phrase or whole sentence, switching from Roman (upright) to

Italic typeface (to be balanced by the fact that italics is lighter in shade than the
Roman), and varying line length of the important type or using ragged right if the
rest is justified etc. There are also special variations of typefaces which can serve
to intensify salience, such as the use of various forms of outlining, shadowing, and
ornamental or swash characters, and by contrasts in colours, as in a contrast
between type in highly saturated colours and softer muted colours, or the contrast
between type in primary red, white and blue. These special variations are not
really relevant to The Economist magazine since it rarely outlines or shadows the
type and exclusively uses monochrome type; it is important in advertising pages
however. As White elaborates, typography in a magazine such as The Economist
should be subjected to consistency in styling and not be used as an exploratory,
esoteric art form (an area that should be left to typographic experimentalists),
because a magazine
depends on its typography to hold it together (sic). Typographic
fireworks, albeit fun and games in themselves, cannot help but tear
it apart (sic). Words (i.e. type) are the one element occurring on all
pages, and they must be used to help unify those pages into a
common package. To achieve this unity — which is actually
nothing but easy recognition — it is essential to retain basic
simplicity throughout (1982b:80).

174
In terms of Framing there are various compositional elements which work to
demarcate or connect the parts of the body copy, signifying that they belong or do
not belong together in some sense. White (1982a:89-102) demonstrates that this
can be realised by a range of elements. The first and most obvious is the use of
actual frame lines under, above and between the typography, and borders around
the typography to set it apart from the margins and any visual elements. These can
be seen in some magazines where the entire space of the live matter is enclosed or
‘boxed in’, or in magazines such as The Economist where each page is separated
from the top margin by a single line, and each article is separated by a column-
wide single line and square symbol. Single lines are especially used on pages
where a number of articles of the same length are presented.

Another very important framing device is the use of clear, empty space around the
typographical elements to mark them off from both the space of the margins and
the gutters. Both the frame lines and empty spaces can be used to form lines of
division between columns and to act as a divider along the central spine of the
page, essentially dividing the page into two typographical halves (this is common
in magazines which include short item leader-reports, such as The Economist
magazine’s ‘Leaders’ section). A variation of these linear dividers is the use of the
runaround, which can be either actual lines or blank space that is used to conform
to the shape of another page feature — in this the type is forced to ‘run around’
another page element, and lines on the page or clear space act as the element
which realises the division (not highly recommended however since it can lead to
a disjointed typographical presentation). A further element which is related to
salience is the ways that large or variable type in sub-headings, bold-face lead-ins,
bylines and blurbs can act as framing devices or dividing lines between
typographical elements.

Framing through borders and frame lines can be weak or strong depending on the
typographical tonal values — the framing values of black in white spaces are of
course higher than in that of grey-shaded, less distinct borders and frame lines.
Framing can also be readily viewed as a continuum of ‘framedness’ if the use of
colour to frame is considered. The divisions between typographical elements on

175
the page can be gradually realised by gradations of background page colour, and
in some more experimental pop magazines such as Raygun, even the colour of the
type itself can be used to frame the page elements. In both these cases there is not
a clear ‘in or out’ sense created, but a graduated balancing act between inclusion
and division. An important device to balance strong framing is the use of bleeding
across page boundaries. This mainly involves visuals being presented across two
pages for effect, but it can also be used in typography, where headings can spread
across the top of two pages, or even start at the top and finish at the bottom of the
same page — this can work to mediate any strong typographical framing, and
combined with visual bleeding, can serve as a strong unifying force on the
page(s).

To summarise then, it can be seen that the structuring principles of information

value, salience and framing operate simultaneously in both the visual and body
copy (verbal) aspects of a multimodal text, and that these principles are realised in
accordance with the inherent characteristics of the mode — in the visual it is
through various pictorial, photographic and illustrative methods, while in the
verbal it is through typology. The next section will address how both the modes
compositionally realise intersemiotic complementarity through these common
structuring principles.

5.7.3 The Analysis of Compositional Intersemiotic Complementarity

Once the ways that each of these modes organise themselves in compositional
terms have been examined, the next stage is to examine the multimodal text in
terms of how the structuring principles of information value, salience and framing
are utilised by both the verbal and visual elements on the page. The previous
sections discussed how these principles operate within the modes, but now the
discussion will focus on how they work together on the whole page space. The
argument presented here is that an intersemiotically coherent multimodal text is
potentially realised via the following compositional intersemiotic
complementarity relations of:
• Information valuation on the page
• Salience on the page
• Degrees of framing of elements on the page
176
• Inter-visual synonymy
• The importance of potential reading paths.

Each of these will be discussed in turn, but before this a number of compositional
aspects regarding the magazine as a whole should be considered, because they
have an important impact on the ways that these intersemiotic relations operate on
the page. White (1982b:17-43) discusses at some length the importance of “how
we look at, and then read, the product [the magazine]”. The first important
element is the physical properties of the whole published document, be it a
newspaper, magazine or textbook. This involves aspects of its size, shape, weight,
stiffness, floppiness, thickness and the techniques used for binding (op.cit:15).
These all affect the ways that it is held in the hands. If it is a magazine, then the
ways that a reader can leaf through it is important in influencing the magazine’s
construction. Readers of magazines like The Economist tend to hold them by the
spine and them flip the pages looking for elements which catch the eye. As White
points out,
Readers see the magazine twice: the first when riffling through the
pages to determine whether it is interesting; then the second,
slowly and deliberately, when they have committed attention to it.
The process results in two distinct sets of impressions: (1) a
cumulative muchness, and (2) a set of memorable highlights. These
highlights stand out in contrast to the overall background texture
and their function is to beguile the superficial page flipper into
becoming an involved participant - a reader (op.cit:17).

The impression of a “cumulative muchness” is generated by the repetition of a

range of compositional devices or signals which work to guide the reader through
the pages and allow him or her to apportion time for the reading. The sense these
repetitive, recognisable devices produces is a coherent pattern that allows the
reader to efficiently interact with the whole magazine. This is commonly realised
via the consistent use of department headings, similar pictorial and graphic
(visual) elements, numbering conventions, common framing styles, and
consistency in general typology (e.g. display type). Additional important elements
are the ways that the colour and texture of the body copy typography, and the
ways that the margins and spaces are handled — all these contribute to an
impression that the elements contained within are all part of a piece (op.cit:18).

177
These regular, controlled ‘signposts’ can be contrasted with the elements which
are designed and used with the reader in mind as the recipient of ‘memorable
highlights’. These are the devices which work to fulfil the ‘attract’ function, such
as brightly coloured graphics, sketch caricatures, incongruities in layout,
unexpected words or catchy word-plays and puns in headlines, or anything which
may break the sense of conformity, the repetitive background patterning of the
magazine. Once the reader has been persuaded by these ‘attracts’ to go further into
the work, to begin to read, he or she is influenced by the background patterning in
terms of such features as
good clear writing. Good clear illustration that reinforces the
words with apt images. Good clear organisation that encourages
the flow of ideas in logical sequence. Good clear presentation that
acts as a lubricant, helping to slip those ideas off the page and into
the reader’s mind ….. (loc.cit.)

These aspects will be discussed not only in terms of the three structuring
principles of information value, salience and framing on the page already referred
to above, but also in terms of two additional compositional intersemiotic
complementarity relations. These are inter-visual synonymy, and the importance
of potential reading paths. All five of these intersemiotic complementarity
relations will now be explicated.

A consideration of information valuation on the page requires an examination of

the relative placement of the visual and verbal modes in terms of their vertical or
top/bottom placement on the page, their horizontal or left/right placement on the
page, and their relative placement on the page in relation to the centre and margin.
An important element in this discussion of information valuation on the page is
the already-mentioned physical properties of the whole published document, be it
a newspaper, magazine or textbook. As was pointed out, readers of magazines
tend

178
Prime viewing Prime viewing
area area

Area least visible

on first
examination
of the magazine

Secondary
viewing area Secondary
viewing area

Figure 5.12 Information valuation on the page (from White 1982b:16)

179
to hold them by the spine and then flip the pages looking for elements which
catch the eye. Whether they flip from left-to-right or vice versa, the inside pages
of the magazine (near the gutter) are virtually “useless in that first, crucial
viewing” (op.cit:15). Readers of magazines tend to pay attention, in a seemingly
random way, to what grabs their eyes — given this information, the prime viewing
areas of a magazine page, in the initial stages, are therefore those identified and
represented in Figure 5.12.

However, once a reader has been attracted to a particular page, these whole-
magazine prime viewing areas shift to those relevant for a single page; then the
influence of the left to right, Given-New structuring orientation becomes more
marked, and the inside pages of the magazine take on a single page-based
compositional importance.

A consideration of salience on the page requires an examination in terms of the

relative sizes of the visual vis à vis the verbal aspect of the text, and perhaps vice
versa. The proportion of space taken up by each on the page can illustrate how
important the elements are to the compositional makeup of the text. According to
White, size is an indicator of importance, so a graphic designer should
signal the Big Idea of the story in the Big Picture - and make that
big picture as big and as dominant as possible. Then, reinforce the
big idea within the big picture with smaller ideas carried in the
smaller pictures, to flesh out the big idea, add details, corroborate
the thesis …… In visual terms, such attention-focusing is achieved
most easily by the simple expedient of giving the important picture
the size that shows its importance (1982b:127).

On the page salience is realised also by variations or contrasts in tonal values, type
and colours in much the same ways as when they are used within the visual and
verbal modes. There may be situations where a visual uses intense colours or
heavy black tones to attract the eye of the reader, or where larger than normal type
is used to direct the eyes to the headline. Gradations of tone can also be used to
guide the eye from the intense (or hyper-real) colour or deep black from one
element on the page to another — this can work in a step-by-step fashion where,
as the tone decreases in intensity, the eye is guided away (or towards) the next

180
salient feature on the page by the gradual ‘fading’ or ‘deepening’ effect. This is an
aspect of horizontal sequencing which White sees as an important feature of
magazine layout and design (op.cit:26). This can also occur across whole pages,
where the visual or verbal feature can bleed across the pages (this is more
common in images however).

An examination of the degrees of framing of elements on the page involves the

ways that the visual and verbal elements are divided between themselves and the
remaining empty space on the page. Again, the methods by which elements on the
page are bounded or divided by actual frame lines, borders, and empty space work
in much the same ways as when they are used within the visual and verbal modes.
The degree of this boundedness is modulated also by the various ways of using a
runaround, variations in tonal values, and the use of visual-verbal bleeding across
pages.

The two additional compositional intersemiotic complementarity relations of

Inter-visual synonymy and potential reading paths can also be used to convey a
sense of compositional unity in a multimodal text. Inter-visual synonymy is
concerned specifically with the degrees of semblance in form across visual modes
which can work to present a kind of inter-visual harmony, realising an inter-visual
intersemiotic complementarity. In many multimodal texts there is often more than
one visual utilised, and as we shall see in The Economist magazine, this can
potentially be naturalistic photographs, sketch caricatures and drawings, or
mathematical tables, charts, graphs and diagrams. When visuals have been derived
from the same coding orientation, there is potential for them to co-operate on the
page; this can occur even if they are from differing coding orientations. This
complementarity is realised through inter-visual synonymy, where there is some
semblance in form, shape or colour across the page, which in a sense mirrors the
meaning expressed by both visuals.

An aspect which is interdependent with the compositional principles of

information value, salience, framing and inter-visual synonymy on the page is that
of potential reading paths. Kress and van Leeuwen (1996:218-223) discuss this

181
when they refer to ‘linear’ and ‘non-linear’ compositions. The former refers to the
strictly coded verbal text of English, which in its single-mode form must be read
from left to right (e.g. a novel), and the latter to multimodal texts which can be
approached in alternative non-linear (circular, diagonal, spiralling) ways (as in
some newspapers and magazines). Reading paths relate to the hypothetical
viewer's eye movement from the most salient points in the composition to the next
or less salient points. Arnheim’s (1988) work on left-right directionality and the
reading path that a typical reader will follow is of note here, as well as the already-
discussed assumptions that graphic design experts like White make about how
magazines are typically read (White 1982a, 1982b; Swales 1990:15). There is of
course a cultural element to this in that viewers from different cultures (and even
different age groups or educational backgrounds and contexts) may read a
multimodal text in differing ways.

This is supported by Kress and van Leeuwen’s comments about the reading paths
commonly used in magazines and newspapers. They suggest that the reading paths
followed in magazines often involve the readers flicking through and stopping as
pictures or headlines catch the eye, and then perhaps returning to the articles
which piqued their interest. Alternatively, they may go straight to the article
relating to the front page topic or headline, or to their favourite columnist, or the
sports section. Thus the reading path can be selective and partial, as opposed to
being strictly linear.

The model of intersemiotic complementarity presented in Figure 5.3 can now be

elaborated into an explanatory table illustrating the potential ways that
compositional intersemiotic complementarity on the page can be realised. This
table shows that an examination of the compositional features of a multimodal text
involves an examination of those features of the layout or composition which
allow the elements on the page to be viewed as coherent parts of the one
composite text. Again, the assumption is that these features have not been placed
on the page in a

182
METAFUNCTION VISUAL MEANINGS INTERSEMIOTIC COMPLEMENTARITY VERBAL MEANINGS

IDEATIONAL

INTERPERSONAL

Variations in visual meanings occur Various ways of mapping the modes to The body copy as an orthographic whole
according to choices made in terms realise a coherent layout or composition by realised by various structuring principles:
of: • Information Valuation on the page: Information Value:
Information Value: Left/right placement (horizontal axis); Top/bottom placement: visual ‘weight’
Top/bottom placement (vertical Top/bottom placement (vertical axis); assigned to headings, sub-headings, blurbs,
axis). Balance (centre and margins). boldface lead-ins, by-lines, & captions.
Left/right placement (horizontal • Salience on the page: Left/right placement: columns, margins, line
axis). Foregrounding and-backgrounding; spacing, line length (justified or ragged).
Balance (or balancing centre & Relative sizing; Centre & margins: balance between margins &
margins). Tonal contrasts and variations in focus. ‘live matter’.
COMPOSITIONAL Salience: • Degree of framing of elements on the Salience:
Foregrounding & backgrounding. page: Clear spaces or actual frame lines; Relative sizing: size of the type face.
Relative sizing. Contrasts in tones and/or colours. Contrasts in tonal values: lightface vs. extra -
Contrasts in tones and differences Visual-verbal bleeding; run-arounds; bold.
in sharpness of focus. margins; gutters. Contrasts in type: traditional vs. decorative
Framing (weak and strong). • Inter-Visual synonymy: typefaces, roman vs. italics and various forms
Clear spaces or actual frame lines. Degrees of semblance in form across visual of outlining, shadowing, and ornamental or
Contrasts in tones and/or colours. modes. swash characters.
Structural elements which divide • Potential Reading paths: The impact of Contrasts in colours: highly saturated vs. softer
visual space. potential reading paths. muted.
Vectors producing an hierarchy of Framing (weak and strong).
elements via arrangements of Frame lines, Empty space, Borders.
abstract graphic elements. Contrasts in tonal values and/or colours.
Similarity and differences in size or Elements which divide/control visual space:
volume in co-occurring images. runarounds, bleeding.

Table 5.9 Compositional intersemiotic complementarity

183
random way, but have been placed there for various purposes within a particular
context of situation and culture, the most important purpose being to convey to the
readers a sense of unity, of co-operation, and of coherence in terms of the central
and supporting messages. The range of potential visual and verbal realisations of
these compositional intersemiotic complementarity relations are outlined in Table
5.9.

5.8 The Theoretical Framework and Summary

In this chapter, the analytical framework which will be utilised for the analysis of
intersemiotic complementarity in The Economist magazine has been introduced,
derived and explained. Each section dealt with the main areas where intersemiotic
complementarity may be realised, and a brief explanation of the theoretical
principles and assumptions of the SFL theory underpinning this study was given,
as well as a more focussed discussion of its various features and their relevance to
economics discourse. This chapter also presented an interpretation of the ways
that visual meanings in the visual semiotic are realised, and this also was related to
the kinds of visuals extant in The Economist magazine.

In each of the last three metafunctionally-based sections in this chapter, an

analytical framework which showed how visual and verbal meanings can
potentially be realised was presented, and this was used as the basis for a tabular
presentation of the ways that intersemiotic complementarity can be potentially
realised in a multimodal text. Each of these tables can now be combined and
conflated to present an overall descriptive framework showing a range of potential
ways in which the intersemiotic metafunctions reflect relations between different
semiotic codes in a multimodal text. This is shown in Table 5.10 (following page),
which will now form the analytical basis for the analysis of a representative text
extracted from The Economist magazine, referred to previously as the Mountains
text.

The analysis of this text will be applied and outlined in the next two chapters.

184
METAFUNCTION VISUAL MEANINGS INTERSEMIOTIC COMPLEMENTARITY VERBAL MEANINGS

Variations occur according to the coding Various lexico-semantic ways of relating the experiential Lexical elements which relate to the visual
orientation. In the Naturalistic coding we can and logical content or subject matter represented or meanings. These lexical items arise according
look at: projected in both visual and verbal modes through the to:
intersemiotic sense relations of:
Identification : who or what Identification (participants): who or what is
Activity : what action • Repetition: identical experiential meaning. involved in any activity?
Circumstances where, who with, by what means • Synonymy: the same or similar experiential meaning. Activity (processes): what action is taking
IDEATIONAL Attributes the qualities and characteristics • Antonymy: opposite experiential meaning. place, events, states, types of behaviour?
• Meronymy: the relation between the part and whole Circumstances: where, who with, and by what
In the Mathematical coding we can look at: of something. means are the activities being carried out?
• Hyponymy: the relation between a general class of Attributes: what are the qualities and
Identification : what something and its sub-classes. characteristics of the participants?
Relational Activity : what is the relation • Collocation: an expectancy or high probability to co-
Circumstances : where, what with, by what occur in a field or subject area.
means
Attributes: qualities and characteristics

Variations occur according to the Coding Various ways of intersemiotically relating the Elements of the clause as exchange which
Orientation. In the Naturalistic Coding - it is a reader/viewer and the text through MOOD (Address via relate to visual meanings. These arise according
continua of the use of: offers, commands, statements, questions) and to:
Address MODALITY (Attitude re something as real or unreal, true The MOOD element in the clause realising
Involvement & Power or false, possible or impossible, necessary or unnecessary, speech function
INTERPERSONAL Social Distance and other attitudinal positions) through the intersemiotic
Modality Markers relations of: The MODALITY features of the clause which
express attitudes. Modalisation views on the
In the Mathematical Coding - it is a continua of • Reinforcement of address: an identical form of possibility, probability, and certainty of the
the use of: address. Proposition, as well as the use Comment
Involvement & Power • Attitudinal congruence: a similar kind of attitude. Adjuncts. Also the use of attitudinal Epithets in
Modality Markers • Attitudinal dissonance: an opposite or ironic attitude. the form of subjective adjectives.

Variations in visual meanings occur according to Various ways of mapping the modes to realise a coherent The body copy as an orthographic whole
choices made in terms of: layout or composition by realised by various structuring principles:
COMPOSITIONAL Information Value • Information Valuation on the page Information Value
Salience • Salience on the page Salience
Framing (weak and strong). • Degree of framing of elements on the page Framing (weak and strong).
• Inter-Visual synonymy
• Reading Path

185
Table 5.10 Analytical framework for intersemiotic complementarity in The Economist magazine.

186
Chapter Six will focus on the contextual variables which are relevant to the
interpretation of the Mountains text. This includes an examination of The
Economist magazine as a cultural institution, and the stylistic and graphic design
conventions it utilises in the context of creation of its texts. It also includes an
examination and discussion of the contextual variables which have influenced the
Mountains text. This will involve an examination of the text in relation to its
Context of Situation, and in terms of its intertextual history. Chapter Seven will
involve a detailed application and discussion of this analytical framework to the
Mountains text — this will be carried out according to each metafunction, and will
constitute an attempt to answer the questions raised about how the visual and
verbal modes work together on the page, and to test the argument that
intersemiotic complementarity occurs when certain conditions are met. It is argued
that intersemiotic complementarity will obtain when one or more of the following
occurs:
• when the ideational meanings in both modes are related lexico-semantically
through intersemiotic sense relations of repetition, synonymy, antonymy,
hyponymy, meronymy, and collocation.

• when the interpersonal meanings in both modes are related through

intersemiotic reinforcement of MOOD, and through intersemiotic attitudinal
congruence and attitudinal dissonance (Modality) relations.

• when the compositional meanings are integrated by the compositional relations

of information value, salience, visual framing, visual synonymy, and potential
reading paths.

187
Chapter 6: The Context of the Mountains Text
6.0 Introduction
The next two chapters in this research will analyse the representative multimodal
text previously referred to as the Mountains text. As already outlined, this research
will constitute an attempt to test the SFL model’s assertion of the interrelatedness
of semiotic systems, and to demonstrate the ways that this interrelatedness can
potentially be realised in page-based text.

The multimodal text which is the subject of this research realises, like all such
texts of this type in The Economist magazine, a particular contextual configuration
of the variables Field, Tenor and Mode. However, apart from being situated in the
broader UK and world economic, financial and eco-political environment, this
specific configuration is situated in a context of creation (Hasan 1996:52). As
pointed out in Chapter Five, this context of creation includes the ways that the
artistic conventions of the authors’ community, their conception of their audience,
and their individual preoccupations, are built into the multimodal text. In the
context of this research this pertains to the writing and graphic design policies of
The Economist magazine’s staff in terms of such aspects as its conceptions of the
medium of its message, the relative primacy of the visual and verbal modes, the
general subject-matter, the typographic conventions, the writing/stylistic
conventions, and its preferences in terms of visual and verbal typology. It also
pertains to The Economist magazine staff’s perception and attitudes towards its
readership, and to how The Economist magazine and its writers interpret their
roles as writers/publishers dealing with economic, financial and eco-political
issues. These extra-visual and extra-linguistic variables are an integral part of
what could be considered The Economist magazine’s context of creation, and as
such are aspects which can have an important bearing on the production and
subsequent interpretation of the context of situation of a text like the Mountains
text.

This chapter will firstly examine The Economist magazine in terms of the context
of creation of its multimodal texts, and will draw information about them from
various sources. This will be followed by an attempt to relate these broader issues

186
to the Mountains text by examining the more specific features of the text’s context
of situation (Field, Tenor and Mode), as well as aspects of its intertextual history.

6.1 The Economist Magazine and its Context of Creation

This section will examine the context of creation of the Mountains text according
to Hasan’s (1996) categories: the artistic conventions of the authors’ community,
the authors’ conception of the audience, and the authors’ preoccupations
(regarding their roles as publishers of economic journalism).

6.1.1 The Artistic Conventions of the Authors’ Community

The artistic conventions of the authors’ community will be examined according to
the following categories: The Economist magazine’s institutional background and
published reviews, The Economist Style Guide Book and Numbers Guide, an
interview with the magazine’s Chief Editor of Graphic Design, and a description
of the magazine’s visual and verbal graphic design typology.

6.1.1.1 The Institutional Background of The Economist Magazine

In relation to the institutional background of The Economist Magazine, there are a
number of important details of note. The Economist is a British magazine which is
often considered to be one of the more respected magazines available in business
journalism in the world. It has a broad, international scope, combining articles on
politics, finance and industry, with some specialisation in discussing localised and
international finance. It is published weekly, and its regular departments include
the following:
· Arts, Books and Sport
· American Survey
· Asia
· International
· Europe
· Britain
· Business
· Finance
· Science and Technology
· Economic and Financial Indicators

187
In a recent review on the World Wide Web by Luhman (1998), The Economist is
described as dealing with a wide variety of issues on finance, economics, politics
and industry with “insightful depth”. It is viewed by many of its readers as not
simply reporting the economic and business news, but trying to examine what the
news may actually mean by attempting to provide some degree of analysis.
Further, it is claimed that
The Economist also has some of the best articles on finance. Many of
these articles are about the international banking and securities
businesses, and they often translate important new laws or academic
studies into laymen's terms. Articles are loaded with useful and
cogent information and graphs. They're also written with a fair
amount of wry humour.

Another advantage for the reader is that

all of the articles published in The Economist are anonymous [which]
helps to keep the focus on the article, and not on the author. It also
allows leading industry experts to air their views without fear of
retribution.

All these points accord with the aims outlined in The Economist magazine’s own
published policies or explications on its magazine, The Economist Style Book
(1990) and Numbers Guide (1991).

6.1.1.2 The Economist Style Book and Numbers Guide

The Economist Style Book (1990) and Numbers Guide (1991) are the only
published works which give any indication of the ways that The Economist
magazine as a publishing institution views the medium of its message, and of
some of its attitudes to its subject matter and the personalities involved. Both are
in-house publications.

The Economist Style Book (1990) is concerned specifically with providing general
advice to its editors and writers on common errors in writing and to give them
some usage rules. The philosophy behind this Style Book is fundamentally
commercial, in that it has been produced so that The Economist can successfully
compete with its publishing rivals — both in terms of its analysis of issues, and in
terms of its writing. The first rule for writers in The Economist magazine is that
their work should be easily understood, given that “Clarity of writing usually

188
follows clarity of thought” (1). The writers are charged to think about what they
want to say, to say it as simply as possible, and to follow six rules of writing
(derived from George Orwell’s 1946 “Politics and the English Language”) which
are paraphrased as:
1. Never use commonly-read metaphors, similes or other figures of speech.
2. Choose shorter words rather than longer to express the same thing.
3. If a word can be avoided/deleted, then do it.
4. Don’t use the passive if you can express the same thing with the active.
5. Use everyday language over foreign borrowings, and scientific jargon.
6. Break any of the above rules before you write anything that is “outright
barbarous” (1).

Specific guidelines for editors are also given in the latter part of the book. They
are asked to be careful of over-editing, noting that one of the sources of The
Economist magazine’s excellence in its field is the fact that there is a lot of variety
in their articles. This is supported by a stated editorial policy — “The Economist
has a single editorial outlook, and it is anonymous” (10). Thus it is not the
intention of the Style Book to impose a single style on all the journalists around the
world who submit articles to The Economist magazine, but the editors do ask for
and give respect to good writing; they are also required to look for certain
infelicities of style, and they require the journalists to follow the various rules
which they set out in some detail.

The Numbers Guide (1991) is a very different kind of publication to the Style
Book, being a guide to reading and understanding how mathematical (numerical,
statistical and visual) data in The Economist magazine can be interpreted and used
for commercial decision-making. It is designed as a companion volume to the
Style Book, which is a little surprising since this book (the Numbers Guide) seems
to be aimed at the reading public, while the Style Book is aimed primarily as a
guide for the writers and editors who work for the magazine (although it is
available to the general public). Thus, the Numbers Guide is not very revealing
about the artistic conventions used at The Economist beyond giving some
indication of the kinds of mathematical visuals which it typically uses: graphs,
tables, charts and the occasional pictogram.

189
THE FUNCTION OF VISUALS How important do you consider visuals to Economist articles? Are they essential, or
IN THE ECONOMIST performing some filler function? Why?

What are visuals supposed to be doing in an article?

Are they there to be relevant in a general way to the overall topic of the article?

Are they there to provide extra information in support of the text?

Are they there to repeat and refer to specific information in the article, or are they there
simply to attract a reader's attention to read the article?

THE PROCESSES INVOLVED Is there a general policy on the graphic design of each article which fulfills the same
IN ARTICLE CONSTRUCTION purpose as the Economist's Style Guide? If so, what is it?

Could you describe for me the process involved in the creation of an article, from the
beginning, for each issue? (At which point does the graphic design department become
involved? Why?)

How is the choice of visual (caricature, graphic, photo) decided? How is it

decided/who decides to use a photo or a graphic or a caricature, or some combination
of each?

Who draws or designs each of the visuals? Do they have some background in the area
(caricatures = artists; graphics = mathematics background in training; photos =
photographic training)

Is a bank or collection of visual drawn upon, or do you sometimes create visuals

specially to suit the article?

In deciding the placement of text relative to the visuals, what guidelines do the
designers use?

Are you given a text and then do you try to fit the visuals around it? Or is the text
arranged around the visuals?

Are the visuals placed according to some guidelines? If so, what are they?

Who decides on the headings for the visuals and is there a policy about the kind of
heading and its tenor (humour, punning etc.)?

GENERAL SALIENT POINTS What are the Economist's expectations about its audience or what is its reader profile?
REGARDING THE GRAPHIC
DESIGN POLICY
Why does the Economist use humour in many of its visuals and headings? Is this
deliberate, and why?

Do you think that the Economist is trying to make assertions about its coverage to try
to influence its readers in some way? Which ways do you think are acceptable to the
magazine?

Table 6.1 Graphic design at The Economist - interview questions

190
6.1.1.3 The Economist Magazine’s In-house Graphic Design Policy
The nature of any ‘official’ graphic design policies which may exist within the
institution of The Economist magazine itself can also reveal much about the
artistic
conventions of the authors’ community. To ascertain if there were any in-house
graphic design policy, and to get some idea of the daily workings of the graphic
design team at The Economist magazine, an interview/discussion was conducted at
the London headquarters of The Economist Magazine on 17th August, 1994 with
Ms. Penny Garrett, the Chief Editor of Graphic Design. The interview was
conducted while Ms. Garrett was working on the design of the cover page of that
week’s issue of the Atlantic edition of the magazine. In order to focus the
interview, and to ensure that as many as possible of the relevant points concerning
graphic design policies at The Economist would be covered, a series of prepared
questions were used to guide the discussion. Not all of the questions were asked,
they were not necessarily asked in any particular order, and they were not simply
read out, but used as guides or prompts. These focus questions are listed in Table
6.1, and the results are summarised and collated in the following points.

In terms of the overall function of visuals in The Economist, Ms. Garrett indicated:
• the verbal is primary, and is the starting point for all aspects of the magazine's
texts. (she couldn't imagine why or especially how you could construct
magazine texts starting with the visuals, since The Economist is a magazine
largely to be read).
• visuals are used because economics is generally considered a pretty dry, boring
subject area which needs eye-catching aspects to attract the reader (the primary
reason why humour in the form of puns or wry statements are used in the
headings in almost every article, headings which often relate to the visuals
used).
• while the primary focus of visuals is to attract the attention of the reader, there
are differences in perceptions as to the relative importance of each mode in that
role. Photos and caricatures are believed to be primary ‘attractors’, while the
graphics (tables, graphs and charts) are there to add/provide information,
inform or support the main text.
• given this ‘attract’ function of visuals, information that is presented in too
complex a form will not fulfil that aim. Relatively simple graphics are thus
used for information purposes only, to support the information already
provided or to inform the reader on aspects relevant to the topic.
• visuals are never used as fillers to fill up spaces in the page. If there were extra
spaces that needed to be filled, then the writers would perhaps be asked to write
more; however it is more common to ask the writers to reduce the volume of

191
 their writing than to deal with the problem of trying to find page space for the
visuals.
With regard to the place of visuals in an assumed reading path, Ms. Garrett
suggested:
• the attention-grabbing role of visuals starts with the front page; they are used in
that place to attract the reader to the magazine as a whole.
• it is assumed that once a reader is attracted by the front page, then he/she would
start to flip through and look at the various sections. Thus, eye-catching aspects
of the magazine are important. As the reader flips through it is assumed that the
visuals then work mostly to catch the eye of the reader.
• the most likely reading path is suggested as a photo or caricature first, then the
graphics, and then the heading and blurb. If ‘hooked’, the reader then would
start to read the article more carefully.
• it is preferred that caricatures are not labelled overtly, the preference being for
the reader to figure out or quickly identify the participant or subject that is
being visually caricatured (in fact, approximately 65% of the corpus collected
for this research contained labelling in some form on its caricatures — in these
a represented participant is identified by name or according to some specific
attribute, so this is presumably a question of subtlety, where perhaps it is
assumed that in some cases the reading audience cannot readily identify the
caricatured participants or the specific attribute emphasised, and need some
verbal prompting, or ‘anchoring’ in Barthes’ terms).

In relation to the process of article construction, Ms. Garrett stated:

• there is a deadline element to the construction process, with time being a major
constraint on the design decisions made.
• there is no set policy of rigid guidelines laid down in the construction of each
weekly issue of The Economist, but a collaborative team effort which starts
with an initial meeting of the editors to decide topics, areas of interest etc.
• the next stage involves the editors, writers and graphic designers in various
meetings to work out the actual texts in each issue. In the majority of cases, the
verbal is written first (either finished or at least started). The writer then meets
with the editor and graphic designers to discuss the article and to give the
designers some idea of its content, thrust of argument etc. Generally there is a
requirement for at least one visual, or one, two or three of varying types,
depending upon the article. These joint meetings decide on the visual number
and type.
• in rare instances, there are situations where the writer is influenced by the
graphic designers' visuals. This is commonly in the situation where only 2 - 3
paragraphs of an article have been written and the writers collaborate on the
types of visuals to be included with the finished product. Sometimes the writer
may be impressed by the graphics' elements and may include words or phrases
in the text which directly follow the humour or proposition of the visual. In this
sense therefore the visual shapes the elements in the verbal.
• the construction process depends on the article, but the aim is for visuals to be
cohesive, or be relevant to the article's content, so they are chosen based on a
collaboration between the writers, the editor and the graphic designers.

192
• this collaborative decision-making process works within the parameters of a
common typological format to The Economist texts (this is explicated in
Section 6.1.1.4, which deals with visual-verbal typology). Articles which do
not fit this general format are shorter articles of straight reportage, reader
comments, reportage/comments on headline issues, and even economic gossip
columns.
• in deciding the relative placement of the visuals to the verbal in the layout of
each issue, it is very rare for the verbal text to be designed around the shape of
a visual. In fact the visuals are able to be sized and re-sized using computer
graphic design software (which Ms. Garrett was doing during the interview).
Decisions are made based upon the number and type of visuals per page and
the need to maintain reader interest. Again there are no set guidelines, but
important considerations are spatial criteria in terms of page size, the number
and type of visual, and the need to alternate types of visual.
• an important point is that the verbalised headings for the graphics are done
after the graphs have been drawn. This also applies to the headings for the
whole article, as well as for captions for photos and caricatures. There is no one
person who decides on these captions or headings; sometimes it can be the
editor, the writer, and at other times the graphic artist. Again, it is a
collaborative effort.
• the remainder of the magazine is developed according to the same background
principles, and thus basically follows the same format in varying lengths.

Ms. Garrett also provided some background on the visual designers or drawers in
this construction process:
• the sketch caricatures are drawn by people who have qualifications in graphic
arts — with major drawings for a leader article or a cover page, the graphics
department also commissions artists who are informed of the nature of the
cover or article.
• the graphics are produced by people who have some kind of statistical
background.
• the photos are often drawn from a bank of photos from an agency — in this
case they tell the agency the nature of the article or cover, and the agency sends
them what they consider appropriate choices, from which the final visual is
chosen.

6.1.1.4 Visual and Verbal Typology at The Economist Magazine

There is a common verbal and visual compositional format in each issue of The
Economist magazine which in this section will be described firstly according to
the most common ways that the verbal aspect of the text is organised in terms of
its usage of headings, typefaces, columns and framing between the articles, and
secondly according to a categorisation of the types of visuals most frequently
used.

193
As outlined above, the bulk of The Economist magazine consists of a number of
departments or sections, which in every issue are labelled as American Survey,
Asia, International, Europe, Britain, Business, Finance, and Science and
Technology. There have been very few changes to these department names in
recent years, the most notable being in the Finance department, which in current
issues is referred to as the Finance and Economics department. Each of the
departments in the magazine follows basically the same page format, with the
following verbal characteristics:
• each commences with a major, leader article, which is followed by a series of
smaller, supplementary articles.
• a large bolded and capitalised heading identifies the department, such as
‘BUSINESS’, or ‘EUROPE’.
• the same identifying word ‘Business’ or ‘Europe’ occurs in smaller, bold
capitalised type to signal that each of the pages in which it is placed is part of
that particular department. This is placed on the top corner of each subsequent
page (on the left of the left-hand page, on the right of the right-hand page).
• under each of these department headings there is a single black line at the top
of the page which separates the department heading from the textual page space
below.
• the ending of the department is signalled by the large bolded heading of another
department.
• the articles within each department are different in size and length, whether
they include visuals or not, but each has a beginning which is realised by a
visually salient bolded heading, and an ending signalled by a column-wide dark
line with a small black square on the right-hand side. The placement of this
small black square in the right-hand position on the dividing (ending) line is in
a sense a visual ‘period’ or ‘full-stop’ to the text, signalling to the reader that
the article ‘ends here’.
• in leader articles, the blurb, which basically summarises the main argument or
point of view of the article, occurs directly below the main headline.
• in the shorter articles which follow the leader article there is usually a small
heading identifying the topic area or place in which the story occurs, followed
then by the more visually salient article headline.
• directly under the blurb in leader articles is the source of the article; the article
can be from any number of overseas bureaux or ‘desks’, such as Tokyo, Paris,
New York etc. The names of the writers are never given.
• all articles are started with drop capitals, in agreement with the Style Book
requirements (1990:31).
• the ‘live matter’ space on the page is typically organised into three columns,
which allows some articles to be either composed of a single column, or spread
across most or all of the page. Two columns occur only occasionally in ‘gossip
columns’ like ‘Bagehot’ or ‘Lexington’.

194
• visuals are placed and arranged in accordance with the structure of these
columns; visuals can be blocked in by justified type, or the columns are made
to runaround them.

In terms of the visual elements, The Economist magazine typically uses instances
which are recognisable as being members of the three categories of visual
communication already referred to in Chapter Five (Section 5.4) as the
Naturalistic, Mathematical, and Spatial visuals. The derivation of some of these
categories of visuals in The Economist may be understood by making reference to
the term 'coding orientation'. Visual coding orientations have been defined by
Kress and van Leeuwen as "sets of abstract principles which inform the way in
which texts are coded by specific social groups, or within specific, institutional
contexts" (1990:53).

The NATURALISTIC visual is a categorisation developed for this research to

describe the various forms of visuals which are derived within the naturalistic
coding orientation (op.cit:54). This orientation involves an attempt to portray what
may be considered 'natural' images, or images which the members of a particular
cultural setting would consider to be a recognisable representation of reality as
viewed by the human eye. In this coding orientation, it is proposed that there is a
continuum of representation from the 'hyper-real' or 'more than real' photograph
(op.cit:52) to a line drawing of a simplified or abstracted reality. This continuum
of possibilities, which is summarised in Figure 6.1, can also be divided into the
sub-categories of Photographs and Sketches.

hyper- colour B/W sketch sketch line

real Î photo Î photo Î drawing Î caricature Î sketch
photo

Figure 6.1 The naturalistic visual continuum

With the first category, that of photographs, there is the hyper-real photo, which is
basically a photograph which has been processed and developed so that it can
"show colours so rich, and be modulated in such detail, that it seems 'more than
real', or 'hyper-real'" (loc.cit.). The next type, the colour photo, is in fact the

195
naturalistic 'real' standard, where the colour represented in the photograph would
be considered to be an accurate or credible representation of a 'real' scene or
subject by most people in a particular cultural setting. Neither of these types of
photographs are very frequently used by The Economist, except on the front page.
By far the most frequently used type is the B/W (Black and White) photo, a
photograph which has had colour abstracted away to the point where there is an
absence of any natural colours, apart from the two colours of black and white.
There is a continuum of variations of these two colours, which runs from their full
intensity of colour (total black or total white), to various values in terms of their
brightness, where various degrees of darkness or lightness are used.

The next type of naturalistic visual is the sketch. The first of these, sketch
drawings, are basically hand-drawn sketches which may or may not include
colour, and which attempt to portray a real scene or human subject(s) as
realistically or as accurately as possible. An example of this could be a sketch
drawing of the same naturalistic quality as a court scene drawing in a newspaper,
where the aim of the sketch artist is to approximate a photographic image, because
in that context the law prohibits photographs. These types of visuals are used
occasionally in The Economist, and are always in black and white.

One step further away from this type of naturalistic sketching is the use of
caricature in the visual, here referred to as sketch caricatures. These are sketches
of scenes or characters (real or fictional) which are easily recognisable to the
viewers, because they represent familiar objects or entities. However, these are not
accurate representations, but images in which the main features of the subject have
been emphasised in caricature. Here the emphasis is not on accuracy or realism as
in sketch drawings, but on presenting a subject in such a way that the main
features are over-emphasised, but not to the point that the viewers can’t quickly
identify the subject. This is often done when caricatures of famous people are
sketched, where a head, or a prominent nose or ears are made overly large, as in a
caricature of Prince Charles, for example. Further, the subject is sometimes
portrayed in an unnatural, or unusual way, in circumstances or situations which
cut across most viewers' realistic expectations. This combination of the use of

196
caricature and the insertion of the unusual often produces the comic but evaluative
effect for which The Economist is well-known. The result is that the visual could
even be interpreted as a cartoon. These types of sketch caricatures are very
frequently used in The Economist, where the real or fictional subject of a visual is
familiar to the viewer, but is shown in an unusual situation which is semantically
related in some way to the story.

The next type of sketch in The Economist is the kind of sketch which constitutes a
strong abstraction away from what may be considered a realistic portrayal by the
viewer. This is the line sketch, which may or may not portray its subject
comically, and may or may not portray real or fictional subjects. This type of
visual is not often seen in The Economist, but when it does appear, it involves the
use of simple, caricatured line drawings which include a minimum of detail,
almost to the point of being a 'stick-figure' style of visual. Many of the finer
details of the other sketch types are omitted such as strands of hair, patterns in
clothes etc., the main aim seemingly being to simply portray an activity or a scene
which is relatively easily recognisable to the viewer. Usually it is describing an
activity rather than a person or famous scene.

The MATHEMATICAL visual is a categorisation used to describe the various forms

of visuals which are derived from the scientific/technological coding orientation
(Kress and van Leeuwen 1990:54). This orientation has as its guiding principle the
measuring of the usefulness or effectiveness of a visual as a 'blueprint', or in the
accurate presentation of information. This orientation covers a wide variety of the
visuals used in diverse disciplines across the academic spectrum. Visuals are used
inter alia to show structure, illustrate dynamic relations, clarify processes, and
explicate procedures. The types of visuals used in The Economist invariably
congregate around the kinds of scientific/technological coding realisations that are
readily identifiable as being mathematical or statistical in nature and origin. They
are never blueprints for building (as in engineering, architecture etc.), or
instructions for use (as in technical manuals), and rarely are they used to explain
processes (as in biology, physics chemistry etc.). They are however used to
portray sets of data or variables in terms of other sets of data, which usually fall

197
within the province of the mathematical or statistical academic disciplines. For
these reasons, the term mathematical visuals will be used to describe the second
major type of visuals used in The Economist.

Unlike naturalistic visuals, mathematical visuals do not seem to be easily

describable in terms of a continuum, although there may be a continuum in terms
of levels of complexity according to the information displayed. Rather, they may
be most easily described in terms of presentational modes, and according to the
potential number of visual relations expressed. In The Economist there are
basically four types of mathematical visuals, which are defined by the Random
House Webster's Electronic Dictionary and Thesaurus (1992) as :
(1) the line graph - a series of discrete or continuous points forming a linear
curve, each point along which represents a given (x,y) value.
(2) the bar graph - a graph using parallel bars of varying lengths, to illustrate
comparative costs, exports, or some other variable.
(3) the pie chart - a graphic data display in which sectors of a circle correspond in
area to the relative size of the quantities represented.
(4) the table - an arrangement of words, numbers, or symbols, usually in vertically
or horizontally parallel columns, which display a set of facts or relations in a
compact and comprehensive form.

Each of these visual presentational modes potentially express visual semantic

relations which may occur singly, or in some combination, and may be glossed as
the derivative (rate of change), the comparative, and the proportional types of
visual relations. For example, some mathematical visuals may express only one
type of relation, as in a line graph which expresses a simple derivative relation, or
may include two, as in a table which attempts to portray comparatively a number
of economic features which change over time, or a couple of line graphs in one
visual which show comparative rates of change, or even a couple of bar graphs
which portray proportion, rate of change and comparison relations.

The derivative (rate of change) visual relation is derived from the fact that it is
concerned with the mathematical relation of the derivative, which is defined in
mathematics (and in econometrics) as "the instantaneous rate of change of one
quantity in a function with respect to another" (Random House Webster's
Electronic Dictionary and Thesaurus 1992). Expressed another way, these kinds of

198
visual relations demonstrate graphically how the rate of change in one variable
(called the dependent variable) behaves in relation to the rate of change in another
(called the independent variable).

In The Economist the line graph is the visual method used to portray rate of
change in a dependent variable. The independent variable is most often time,
while the dependent can be any statistically measurable feature of an economy,
such as unemployment, stock prices, inflation rates etc. This rate of change
graphical relation in economics can also be expressed in functional algebraic
terms, as in the functional equation Y = f(X), which may be rewritten as Y is a
function of X. The process of calculating the derivative is called differentiation
with respect to the independent variable. Calculation of this function produces the
first derivative, a pivotal concept in economic explanations of the ways that
variables in an economy interact. It is important because the first derivative of any
total function can be interpreted as the marginal function, which gives an
indication of the rate of change in the dependent variable. This is graphically
interpreted in the slope of the graph of the function (Pearce 1992:104).

Line graphs which express only a single rate of change relation are very common
forms of mathematical visuals in The Economist. Variations of the single line
graph usually involve the portrayal of multiple line graphs, which show the
various rates of change of a number of dependent variables for the purposes of
comparison, or may include the use of bar graphs to express the same relations,
the top of each bar representing the points (x,y) along which a line graph could
possibly be drawn. There is also a comparative relational element to this kind of
visual (discussed below).

The comparative visual relation is derived from the fact that it is concerned with
presenting visually two or more visual modes of representation (linear, parallel
blocks, figures/words data, circle segments, symbols) in order to emphasise their
similarities and differences. In The Economist this seems to occur only in
conjunction with other visual relations.

199
The proportional visual relation is derived from the fact that it is concerned with
presenting visually the comparative relation between variables as to magnitude,
quantity, and height/depth on a visual plane. This visual relation is promoted
specifically through the use of parallel blocks, circle segments, numbers, and
sometimes symbols in bar graphs, pie charts and tables. This relation is very close
in purpose to the comparative relation, in that it seeks to emphasise the similarities
and differences between proportions, and that it seems to occur only in
conjunction with other visual relations.

The SPATIAL visual is a categorisation used to describe the various forms of map
visuals. They are termed spatial visuals because they represent their meanings on
a flat surface some selected features of a part of the earth, or a portion of the
heavens in relation to each other's cartographic positioning. These are shown in
their respective relationships according to some convention of representation, or
codes (Random House Webster's Electronic Dictionary and Thesaurus 1992). This
kind of visual is a scaled attempt to inform the viewer about the relative spatial
arrangement of symbolised (or codified) features, which in The Economist is most
often carried out via the occasional use of various types of maps showing the
location of countries or resources in particular areas.

6.1.2 The Authors’ Conception of their Audience

Discerning the nature of the perceptions that The Economist magazine, as an
institutional ‘author’, has of its audience is a difficult task, given that this is
viewed as commercial information that is usually kept in-house. Nevertheless, the
sources utilised to glean this information do indirectly reveal something of the
ways that The Economist views its reading public. The interview with Ms. Penny
Garrett (1994) is perhaps the most revealing — when asked about the Economist's
expectations about its audience or the actual existence of a reader profile, Ms.
Garrett stated that they assume their magazine is aimed at an audience which
ranges from professional economists and business people, to general readers who
have attained a certain level of education and sophistication which allows them to
have an interest in a magazine of its type. There is no explicit reader profile (that
she was aware of). Luhman’s (1998) review (in Section 6.1.1.1 on institutional

200
background) supports these comments in that The Economist magazine’s audience
is generally assumed to be professional white-collar people, but suggests also that
one of the attractive features of the magazine is that the writers are able to
translate important new laws or academic studies into laymen's terms, thus
appealing to both peripherally and deeply involved readers.

The fact that the magazine is also published in two regionally-focussed editions
per week, one for the Atlantic region and one for the Asian means that the
audience is perceived in cultural and/or geographical terms — both issues
concentrate on issues relevant to their respective regions, and they often have
differing front pages containing images which are designed to attract the differing
clientele (Penny Garrett 1994).

That many of the articles are produced with a fair amount of wry humour is also
indicative of the authors’ conception of their audience. When asked specifically
about this, Ms. Garrett mentioned that it is used visually and verbally to attract the
reader to the article. Humour in the form of puns or wry statements is used in the
headings in almost every article — the reader is therefore viewed as a potential
reader, someone who is discerning and needs to be persuaded or attracted in some
way in order to read. To this can be added the fact they are also viewed as
potential (or continuing) subscribers or customers.

This ‘attract’ function of humour is confirmed by the Style Book (1990), which
also refers to its importance in drawing readers in and keeping them (op.cit:37).
The Style Book deals with issues of writing style by extracting examples from its
own issues. The aim expressed in the discussion of these examples is again the
need to keep the reader of The Economist magazine interested, and to encourage
that interest to continue; writers are warned about overuse of metaphors and
jargon with “some of these [metaphors] are tired, and will tire the reader”
(op.cit:2), and “Avoid it [jargon]. All sections of The Economist should be
intelligible to our readers, most of whom are foreigners” (op.cit:4). With regard to
the use of active versus the passive, the only comment is “it is not incumbent upon
you to be pompous” (loc.cit.), and for tone, the writer is asked not ‘be stuffy’,

201
because the “reader is primarily interested in what you have to say. By the way in
which you say it you may encourage him either to read on or to stop reading”
(op.cit:5). This is supported further by a series of ‘do nots’, as in “do not be
hectoring or arrogant”, “do not be too pleased with yourself”, “do not be too
chatty”, and “do not be too didactic”, all of which are designed to “persuade them
[the readers] to renew their subscriptions” (op.cit:6).

The Numbers Guide (1991) is less revealing of audience perception, but it is

addressed to the same reading audience, and gives advice about basic numeracy,
common financial errors, and the various techniques useful for financial problem-
solving and analysing information for commercial decision-making. It is useful for
a general reader of The Economist magazine because it uses mathematical
information and visuals drawn mostly from its own publications, and draws on
mathematics, statistics, accounting and economics to illustrate its ideas (op.cit:1).
It is thus an attempt to give its readers more interpretative skills and to in a sense
maintain their interest by doing so  if The Economist uses professional
terminology and jargon, as well as technical visuals, then the magazine wants to
ensure that its readership is able to understand what is going on. It is in a sense an
attempt to maintain and to increase the readership by educating them into the
field.

6.1.3 The Authors’ Individual Preoccupations

Some important attitudes are revealed in the Style Book (1990) in terms of The
Economist magazine’s views of itself and the subjects of its articles in the
publishing culture in which it operates. It certainly views its magazine product
highly, referring to its own reputation for excellence in writing and publication,
the source of this being in its quality of analysis and quality of writing (op.cit:10).

The Style Book also reveals an element of ‘anti-establishment’ to the magazine.

Thus, while the subjects or people covered in its articles are first and foremost to
be treated with respect (which means using the title a person has been given or
wishes to use), judgements however should be made about some titles, in order to
avoid indulging “people’s self-importance” and those with “an inflated ego”

202
(op.cit:63). With reference to ethnic groups, the aim should be not only to avoid
giving offence, but also to “avoid mealy-mouthed euphemisms and terms that
have not caught on despite promotion by pressure groups” (op.cit:59). Above all,
being in the end a commercial product, the Style Book advises that The Economist
magazine should treat its readers as customers; in that sense the magazine wants to
have and keep a readership, so there are requirements to try to identify with them,
to avoid “being stuffy” (op.cit:5), and to avoid confusing them with long
sentences (op.cit:9). The writers are asked to use the “language of everyday
speech” (op.cit:5), and to avoid boring or offending readers by being too abrupt,
arrogant or pushy (op.cit:6).

6.2 The Mountains Text in The Economist Magazine

The sample text analysed here is an extract from the issue of The Economist
magazine published in March 1993, bearing the title heading “Mountains still to
th
climb” (The Economist, March 27 , 1993:77-78). It is presented in full in
Appendix 1. This text is a typical instance of a multimodal text in The Economist
Magazine, and can be seen as a representative example of the kinds of leader
articles that are regularly placed at the beginning of each department in the
magazine. It is a text which demonstrates The Economist Magazine’s own stated
policies (in The Economist Style Book and Numbers Guide) regarding writing and
presentation methods, and as well as the policies and customs of its graphic design
team. It is also representative in terms of the kinds of visuals characteristically
utilised in the magazine, and in terms of the layout it uses.

As stated above, The Economist magazine consists of a number of departments or

sections, each of which contains a series of articles and commences with a leader
article. This is the case with the issue in which the Mountains text occurs. Each
leader article in this issue occurs under the banner of a large bolded and
capitalised heading, such as ‘BUSINESS’, or ‘EUROPE’. The department in
which the Mountains text is placed is ‘FINANCE’, and it is the leader article for
this department. This particular article is structured in the same fashion as the
leader articles in the other departments in this issue, as well as those that have
been published in previous issues. The same word, ‘Finance’ in smaller, bold

203
capitalised type, is placed on the top corner of each subsequent page (on the left of
the left-hand page, on the right of the right-hand page) to signal that each of the
pages in which it is placed is part of the Finance department. The ending of this
Finance department is signalled by the large bolded heading at the beginning of
the next department, which in this issue is the ‘SCIENCE & TECHNOLOGY’
department.

Each of the articles within the Finance department in this issue are different in size
and length, whether they include visuals or not, but each has a beginning which is
realised by a visually salient bolded heading, and an ending signalled by a
column-wide dark line with a small black square on the right-hand side. In the
Mountains text, the text begins with the large bolded heading “Mountains still to
climb”, and ends with the same black line and square indicated above. Further, it
contains instances of both naturalistic and mathematical visual modes which co-
occur with the verbal mode. In this particular multimodal text, the naturalistic
visual is a sketch caricature, and the mathematical visual is a “boxed” set of two
line graphs. Both are very commonly used in The Economist, and both are typical
in terms of their representational features, graphic conventions, and style.

As explained in Chapter Five, the relationship between context and a multimodal

text is viewed as a dialectic or “dialogic exchange” (Matthiessen 1995:33), that a
multimodal text is the product of its environment, and that it therefore functions in
that environment. It is both activated by the context in which it occurs, and at the
same time works to construe it (Halliday 1978:136). However, as also pointed out,
a multimodal text is influenced by other texts, especially those which have
appeared before it — a text has an Intertextual History. This intertextual history is
important, because the many articles published beforehand in The Economist deal
with issues that have developed over time, and therefore carry common features
which will both inform and confirm an intersemiotic semantic interpretation. This
section will initially provide a specification of the context of situation of the
Mountains text in

204
FIELD of Discourse
experiential domain Maintenance of economic institutions; maintenance of economic
(subject matter) journalism; a description, analysis and prediction about an
economic and financial problem;
short term goal To describe current situation; to inform; to analyse and suggest
potential outcomes;
long term goal To maintain economic institutions and discussions about them;
To maintain subscriber readership;

TENOR of Discourse
agentive or social roles Journalist and drawers to readers/viewers and subscribers;
status Non-equal or hierarchic; magazine writer/drawers are
considered as authority; writer/drawers control the flow of
information. Audience unseen. Reader/viewer can accept or
disavow;
social distance Maximum social distance; institutionalised (journalist/drawer to
readers - journalist/drawer to subscribers);

MODE of Discourse
role of language/visual Constitutive (language constituting social process or whole of
activity); written to be read and drawn to be viewed;
mode of interaction Public act (magazine to be bought); monologic (reader does not
take part in the creation of the text);
medium Written and drawn page-based magazine;
channel Page-based visual and orthographic;
rhetorical thrust Semi-technical journalistic analysis; descriptive and persuasive;

Table 6.2 A description of the Mountains text’s context of situation.

205
terms of the particular values of the contextual variables Field, Tenor, and Mode.
Then it will present an analysis of the text's intertextual history. This will involve
a description of those aspects of its context of situation and culture which
specifically concern its place in relation to pre-existing news events.

6.2.1 The Context of Situation of the Mountains text.

The context of situation is characterised as consisting of the social action (that
which is going on, or the FIELD), the role structure (the cluster of socially
meaningful participant relationships, or the TENOR), and the symbolic
organisation (the role the text is playing, or MODE). Using these variables and
applying them to a multimodal text can allow the analyst to produce a situational
description. Table 6.2 provides a general description of the Mountains text’s
context of situation (adapted from Butt et.al. 1995:126-127).

The description of the Mountains text’s context of situation is a generalised

characterisation of the text’s multimodal presentation of the field of human
experience in terms of who or what is participating, the actions performed, and the
circumstances of this action. It is also a description of the relationships between
the participants and represented participants in terms of their roles and statuses,
whether they are temporary or permanent. Finally, it is a description of how the
two modes, the visual and the verbal, act in the text in terms of the channel,
medium, and rhetorical focus etc. These three descriptive concepts “serve to
interpret the social context of [the] text, the environment in which the meanings
are being exchanged” (Halliday and Hasan 1985:12). The next section will now
delve more deeply into relevant aspects of this text’s context, and it will do this by
describing elements of the text’s history and connection to other texts.

6.2.2 The Intertextual History of the Mountains text.

The analysis of intertextual history will involve an analysis and recognition of
those aspects of the Mountains text’s context of situation and culture which
specifically concern its place in relation to the news events which occurred prior
and subsequent to its publication. The Economist magazine produces a journalistic
form of economics discourse which reports on (or provides background for) and

206
then analyses a financial news story from a particular point of view. No financial
or economic news story occurs in a vacuum, but is both time and context-bound.
Further, the nature of The Economist magazine is such that it is not reporting
“breaking” news, but is dealing with a story aimed at an audience that is assumed
to have at least a minimal knowledge of the story’s existence, some background
knowledge of the practical and theoretical aspects of the story, and perhaps some
familiarity with the main issues as they have developed over time. Therefore, any
analysis and interpretation of the visual and verbal meanings extant in any text
would need to be predicated on some understanding of what has gone on before in
the social, eco-financial, and political spheres which produced the story, as well as
the attitudes expressed in other media contexts. It is not in the scope of this study
to look at these wider issues in any great depth other than through the information
provided by The Economist magazine. The Mountains text is in effect a
microcosm or sub-culture of various intertextual relations which embody the
social, eco-financial, and political meanings which have influenced its
development. In order to clarify these meanings in the Mountains text, an
examination of the story’s topic development in The Economist magazine prior to
and concurrent with this particular text’s publication will be provided, using a
framework which can potentially be utilised for all texts of this type.

The Mountains text deals with a financial issue that has received ongoing press
coverage since it first came into prominence just over two years earlier. The story
relates to the financial problems of the Lloyd’s Insurance Group. An examination
of articles dealing with this topic in The Economist magazine back to January
1991 shows that it is a complex issue which assumes a modicum of familiarity
with the Lloyd’s Insurance Group’s history, structure, operational characteristics,
administrative personnel, and current circumstances. The degree and intensity of
coverage varies from text to text and from issue to issue, but to be able to interpret
the Mountains text effectively in order to provide some coherence to a discussion
of the Mountains text's intertextual history, and of the important contextual
variables which may have an impact on the relationship between its verbal and
visual messages, four specific contextual areas will be examined. These were
defined in depth in Chapter Five, and include the subject matter and the issue

207
(Field-related), the attitudes (Tenor-related), and the visual and verbal typology
(Mode-related) used.

The information discussed in the following sections is derived from a series of

nineteen preceding articles on this issue published by The Economist magazine
since January 26th 1991. All these articles were published in the Finance
Department of each issue, the last one occurring on February 6th 1993, some six
weeks prior to the publication of the Mountains text. Included also is a selection of
subsequent articles, to give an indication of the most recent news on the Lloyd’s
situation.

6.2.2.1 The Subject Matter and the Issue (Field-related)

Before discussing the specific issue with which the Mountains text is concerned
and its intertextual relationship with previously published texts, it may be helpful
to provide some background details about Lloyd’s institutional history. Lloyd’s of
London is a British institution which has a history dating back over 300 years. It is
an insurance market which is unique in that the underwriters are individuals who
are subject to unlimited liability. This means that the people who pledge their
wealth to back or underwrite the insurance market, the “names”, are potentially
liable to lose all their wealth if enough claims are made against the insurance
policies. A well known aspect of Lloyd’s insurance philosophy is the notion of
risking individual wealth for individual gain - the names risk all to gain all. As
The Economist (June 27th, 1991:77) states, a “popular phrase has it that names at
Lloyd’s are liable to their last collar stud if things go wrong” (in practice names
are not truly liable to be totally bankrupted, since there are internal regulations and
“stop-loss” insurance policies to avoid the potential hardship and allow the names
a modest “survival” income). Existing company law supports this limited liability.
The fact that the Lloyd’s names are subject to unlimited liability (contrary to
current law) is a result of an historical oversight in 1720 which gave two corporate
insurers a monopoly, and banned other companies and partnerships but not
individuals. It is also a result of the fact that Lloyd’s had become too established
in its methods by the time of the drafting of the limited liability law by Parliament
in 1837 to be able to change, or be forced to change by the political establishment.

208
The opportunity to become a name is open to everybody, but is subject to an
applicant being able to pass the solvency test, which requires proof that the
potential name has accessible wealth in excess of £250,000. The opportunity to
become a name is therefore restricted to a rather small percentage of the British
and foreign investing public, and is also seen as prestigious or a sign of being part
of the moneyed classes. The application to become a name is usually processed
through large underwriting agencies, insurance brokers and member’s agents, who
collect these names into “syndicates” which collectively underwrite insurance
policies in designated marine, non-marine and motor insurance markets.
According to the notion of unlimited liability, if a syndicate incurs claims in
excess of its reserve holdings, then the names’ other assets may be called upon to
meet the debt. The syndicates formed back insurance policies for specific time
periods at various levels of risk for various premiums; if no claims are made, or
claims are made then settled, then any profits made are distributed to the
syndicate’s individual names. These syndicates are then closed and their accounts
announced (finalised) three years later (thus giving time for outstanding claims to
be finalised).

The syndicates which can’t be closed are termed “open years”. They are open in
the sense that the syndicate is unable to close outstanding accounts or claims for
insurance payments i.e. there are existing claims against the syndicate’s various
policies that are still being processed, claimed or paid out (as in a continual claim
for damages for asbestos-related damages etc.). A name who is in an open year
syndicate is thus continually liable for any current and possible future claims, a
situation which can be rather unsettling, and resulting in the underwriting cycle, or
the situation where either a series of losses tends to drive names away (leaving or
not taking up the chance to become a name), or a series of profitable years attracts
names.

The administration of the Corporation of Lloyd’s is carried out by the Council of

Lloyd’s. This Council has seen some changes in personnel in recent years - in
January 1991, a new chairman took over, Mr. David Coleridge. He commissioned

209
a taskforce to prepare a report on the market’s capital base (discussed below)
which was overseen by Mr. David Rowland, the chairman of Sedgwick’s, a major
Lloyd’s broker. The problems at Lloyd’s, discussed below, continued to multiply
to the point that in August 1992 David Coleridge announced his resignation
effective December 1992, and that David Rowland would be taking over as the
new chairman of the Lloyd’s Council soon after. At the same time the Chief
Executive of Lloyd’s Alan Lord announced his resignation, to be replaced by
Peter Middleton, transferring from Thomas Cook, a travel firm. Both Rowland and
Middleton figure prominently in the Mountains text, and are discussed in depth in
various articles published by The Economist magazine - Rowland continuously
since the release of the Taskforce report in January 1992, and Middleton
specifically in an article published in November 1992 which introduces him to the
reading audience. Both Rowland and Middleton are discussed in all but one of the
subsequent articles published in The Economist magazine leading up to the
publication of the Mountains text.

The issue, or the current problems which the Mountains text specifically deals
with have their roots in a number of problem areas and issues which have
confronted the Corporation of Lloyd’s since the late 1970’s and early 1980’s.
During those years Lloyd’s experienced a number of problems with fraud on the
part of the underwriters, brokers and agents who, it is often claimed, were
incompetent, ill-trained, or all too ready to pursue their own interests at the
expense of the members of the syndicates they were supposed to represent (The
Economist June 22nd, 1991:90). There is more self-regulation of the market now,
but these problems were at that time so important and intrusive that they affected
the ability of the market to make profits for its names. The last reported loss for
Lloyd’s was in fact in 1967, and the highest recorded profits ever were in 1986.
With regard to the reporting of these losses, an important consideration already
mentioned is the fact that profits or losses are reported three years in arrears. So
the 1980’s were mostly good years, where any losses incurred on underwriting
were compensated by investment profits. It is against this backdrop that the source
of the current problems referred to in the Mountains text can be summarised,

210
many of them referred to in the “inauspicious background for Lloyd’s first-ever
business plan” given between lines 6 and 22 (see Appendix 2):
• The 1987 published account figures were only slightly down on the previous
year’s record profits, however in the years 1987 to 1990 there was a series of
natural and human-made disasters which lead to a rapid increase in the number
of insurance claims. These were, for example, the Piper Alpha oil rig fire, the
oil spillage from the Exxon Valdez, the San Francisco earthquake, Hurricane
Hugo and a spate of European gales. Further, Lloyd’s was rather heavily
exposed to the highly litigious US market, where “long-tail” claims for
damages from mismanagement in the American savings and loans institutions,
and claims for damages from pollution and asbestosis cases were still
occurring. The result of all these claims lead to predictions that 1988 would
produce a significantly reduced profit, and that the accounts for 1989 would
show the first loss since 1967 (The Economist January 26th, 1991:71). The
figures for these years were released in 1991 and 1992 respectively, and did
show major losses.

• This aspect of the insurance cycle meant that names wanted to withdraw their
investment in the market’s syndicates. In January 1991 Lloyd’s had 26,000
names registered, but just over half could not be allowed to leave because they
were locked into 58 of the 401 syndicates which were considered to be “open
years” (71). Since Lloyd’s names have unlimited liability, many of the names
faced the very real prospect of being bankrupted.

• The prospect of being bankrupted caused consternation amongst many names

who wanted a “market” solution whereby the risk could be shared out amongst
the other unaffected names. This mutualisation has often been suggested but
refused by the Lloyd’s Council because it did not accord with the traditional
notion of individual risk for individual gain. Other names groups wanted to be
assisted by a government rescue package, a proposal denied despite the fact
that the most vocal proponents included sixty Tory members of parliament (The
Economist June 22nd, 1991:90).

• The problems with the troubled names and syndicates (such as the Outhwaite
and Feltrim syndicates) caused the new Chairman David Coleridge to
commission a taskforce report in January 1991 to examine the market’s capital
base and to make recommendations. This report was overseen by David
Rowland. While this report was being prepared there was a rapid increase in
names trying to withdraw their investment in Lloyd’s, political pressure
brought to bear, and much consternation expressed by those who were locked
into open years.

• The Taskforce Report, released in January 1992, made a series of

recommendations for Lloyd’s to follow to “salvage itself” (The Economist
January 18th, 1992:69). The issue of The Economist magazine at that time
supported many of the recommendations such as reducing costs in combination
with other common-sense business solutions, introducing a compulsory stop-
loss scheme which would limit member’s liability, and a pooling scheme which

211
 would spread capital across syndicates. However, The Economist also criticised
the report’s recommendation to place Lloyd’s under the protection of the
Financial Services Act and to reduce the three year lag in reporting annual
accounts. The Economist suggested that the Council had already shown signs of
rejecting the report’s recommendations based on past decisions when changes
were suggested, and that a big part of the Corporation’s problems were caused
by incompetent management. Thus The Economist concluded, “Change that,
and the market’s prospects would improve. Do nothing, and its decline might
well continue - even if the other reforms are brought in” (The Economist
January 18th, 1992:69).

• The losses, political pressure and syndicate problems continued throughout

1992, and as mentioned above, David Coleridge resigned in December and was
replaced by David Rowland as the new chairman of the Lloyd’s Council. The
Chief Executive of Lloyd’s Alan Lord also announced his resignation, and was
replaced by Peter Middleton.

• The situation facing these two men, and their responses to it forms the
immediate background to the Mountains text. Peter Middleton, the new Chief
Executive, was introduced to The Economist readers in the magazine’s
November 1992 issue. In it he is referred to as “a breath of fresh air”, is
characterised as being more sympathetic to the position and plight of the
names, and is seen as one who “exudes optimism” about Lloyd’s future. In line
with the Taskforce Report, he aims, with the support of the Report’s author
David Rowland, to implement many of its recommendations, such as reducing
costs by one-third, reducing the number of writs served against non-paying
names, finding a solution to the problems of closing open years, and settling
names’ negligence suits brought against syndicate brokers (The Economist
November 21st, 1992:84).

• The two articles published in The Economist prior to the publication of the
Mountains text both deal to varying degrees with the problems outlined above.
The earlier article reiterates the problems of costs, open years and declining
names, and reports on the progress of attempts by Rowland and Middleton in
dealing with them. It also specifically targets Lloyd’s problem with a shortage
of capital brought on by the declining number of new names and the departure
of existing names, and discusses the previously touted and rejected solution:
that of allowing corporate investors to join alongside individual names (The
Economist January 30th, 1992:72-3). The later (penultimate) article deals with a
negative forecast report on Lloyd’s future prospects, the more specific issues of
helping the names who have been hit with unlimited losses, and the decision on
whether to impose another special levy to help meet the payments that various
names have been unable or have refused to pay (The Economist February 6th,
1992:78).

Article Headline Verbal Topic(s) Visual Topic(s)

212
1. Leaking at the seams Falling profits and future Photo: inside of the Lloyd’s building;
(January 26, 1991:71) losses Graphs: underwriting capacity, names
current and resigned.
Chart: pre-tax profits.

2. Losses at Lloyd’s, continued Losses by a major syndicate Sketch caricature: Latin scrawled on a
(May 11, 1991:77) wall

3. Lloyd’s of London Unrelieved misery Refusal by the government to Sketch caricature: a rich aristocrat
(June 22, 1991:90) rescue the Lloyds’ names in begging with top-hat in hand.
trouble

4. Lloyd’s of London Losses unlimited The first overall loss since Sketch drawing: Lloyd’s offices/names
(June 29, 1991:77-78) 1967 and its effects over 100years ago.

5. Lloyd’s of London Bleeding away The increasing resignation of Table: syndicates losing money least and
(August 31, 1991:64) names as a result of the most money.
losses
6. The trials of Lloyd’s The losses and resultant law Sketch caricature: Lloyd’s ship being
(October 19, 1991:95) suits by affected names boarded by pirates.

7. Lloyd’s tries to salvage itself The Taskforce Report and its Sketch caricature: Lloyd’s building
(January 18, 1992:69) recommendations being salvaged at sea by ships.

8. Lloyd’s of London Feeling mutual? Names suing a syndicate Sketch drawing: Lloyd’s names over
(February 15, 1992:82) agent and mutualising the 100years ago.
losses
9. Lloyd’s of London Toughing it out MP names complaining and Photograph: doorman in top-hat
(February 22, 1992:72) Lloyd’s bad publicity below/in front of Lloyd’s .

10. The liquidity gap at Lloyd’s Names refusal to pay and the Sketch caricature: financial tap being
(May 16, 1992:103) loss of capital (liquidity) as a blocked by lawyers.
result Graphs: Lloyd’s compared to other
insurers’ performances.
Chart: Lloyd’s US trust funds
performance.

11. Undermining the foundations The effect (lawsuits) due to a Sketch caricature: Lloyd’s building
(June 20, 1992:75) refusal to help names falling apart.
Chart: Lloyd’s profit and loss
performance.

12. Standing, sinking The results and analysis of a Photograph: a rich aristocrat begging
(June 27, 1992:84) Lloyd’s annual general with top-hat in hand.
meeting

13. Lloyd’s of London Name-calling The effect on names of a Nil

(July 4, 1992:68) report by Lloyd’s on Lloyd’s
Bank

14. Lloyd’s of London Distressed gents A meeting by angry names Photograph: of David Rowland and
(August 1, 1992:68,70) and the resignation by David Coleridge
Rowland

15. Lloyd’s of London Name dropping Discussion of Lloyd’s Photograph: boats wrecked as a result of
(August 29, 1992:72-73) prospects and problems a hurricane

16. Lloyd’s of London Time to sue The lawsuits by angry names Photograph: doorman in top-hat talking
(October 17, 1992:90) to name at Lloyd’s.

17. Lloyd’s of London New broom, old dust Profile of Peter Middleton, Photograph: Peter Middleton in front of
(November 21, 1992:84) Lloyd’s new Chief Executive a Lloyd’s sign.

18. Lloyd’s of London Corporate cure? The new bosses Chart: Lloyd’s underwriting capacity
(January 30, 1993:72.74) performances, and the
shortage of capital
19. Unlimited losses Forecasts and possible Sketch caricature: Lloyd’s building with
(February 6, 1993:78) answers hands trying to stem leakage of money.

Table 6.3 Headlines setting the scene

213
The Mountains text, like its immediate predecessors, deals in visual and verbal
terms with the interrelated problems of open syndicate years and declining names
(see specifically sentences 17-19, 20 in Appendix Two; and both line graphs), and
discusses the shortage of capital brought on by the declining number of new
names and the departure of existing names (see sentences 16 to 18). It reports in
some detail about the prospects for the Rowland and Middleton business plan to
address these difficulties, and then analyses the plan’s prospects in terms of
solving three big problems: the number of open years, continuing litigation and
the financing of the 1990 losses (see sentences 23 to 25, 36-37, 45, 56 and 63).
The final paragraph of the text (see sentences 63 to 67), like the penultimate
article, gives a negative forecast on Lloyd’s future prospects with the business
plan. The publication of the Mountains text therefore constitutes a direct
continuation of the Lloyd’s story, and as such intertextually manifests various
aspects of the general subject matter and the specific issues raised in these
previous discussions.

6.2.2.2 The Attitudes (Tenor-related)

As mentioned previously, information on the attitudes expressed over time can
provide valuable insights into how The Economist magazine writers attempt to
attract and orient potential readers, how they structure their analysis of the issue
and its future development and resolution, and how this then becomes salient for
the situational setting of the Mountains text. These salient features can then inform
any interpretation of the interaction between the visual and verbal messages of the
text. The attitudes of the writers can be examined from three perspectives: the first
is how they attempt to grab the potential reader’s interest and orient them to the
topic through the use of article headlines, the second is via the topic focus of the
visuals used, and the third is through an examination of the attitudes expressed in
the verbal description and analysis of the issues.

The article headlines of the nineteen texts summarised in Table 6.3 point
consistently to the background issues which the Mountains text addresses directly.
The issues highlighted by these prior texts and thus made salient in terms of the
situational setting for the Mountains text are:

214
• the financing of the reported major losses,
• the loss in numbers of names,
• the resultant shortage of capital,
• the number of syndicates with open years,
• the prospect of further litigation by disgruntled names,
• the related problem of unlimited liability,
• the rescue plans put into operation to try to deal with these mounting
problems.

Table 6.3 also shows that the structure of Lloyd’s as an institution, its place in
British financial culture, the difficulties with which it is faced, and its future
prospects, are described and analysed to varying degrees in all the articles prior to
the publication of the Mountains text. The writers of these articles also express
their attitudes verbally through various comments, asides, and the way questions
are framed to the readers, and visually through the use of various kinds of visuals.
A number of direct quotes, listed below, should serve to indicate these attitudes,
which, supported by the visuals, may be briefly summarised as satirising the
“establishment”, and expressing a less-than-sympathetic, slightly suppressed glee
at the problems they were facing at that time. These texts also provide support for
the views expressed in The Style Guide that seem to be ‘anti-establishment’ in
nature.
• Those who work at Lloyd’s are referred to as “those smooth scions of the
British upper class” (Text 1: January 26th, 1991:71).
• In relation to the names’ financial problems, the reader is exhorted: “Do not
weep too hard. Lloyd’s names do not, as many suppose, deserve unlimited
sympathy when the unlimited liability that they gaily signed up to suddenly
seems unlimited indeed” (Text 2: May 11th, 1991:77).
• A comment on the response given to a speaker’s rousing speech at the Feltrim
Syndicate name’s meeting to discuss their financial woes: “The response he
got from these normally placid scions of Britain’s moneyed class would have
been more in place….” (Text 2: May 11th, 1991:78).
• A comment on the government’s refusal to help the names, accompanied by a
sketch caricature of a begging aristocrat: “A main reason for the brush-off was
some much-merited ribaldry from politicians and newspapers about the
country-house-and-yacht-set running hat in hand to the Treasury when the
going gets tough.” (Text 3: June 22nd, 1991:90).
• In the same article as above: “the name’s hope of a free lunch at the Treasury
has been disappointed. Canny speculators expect a small flood of paintings,
yachts, and country houses onto the market. All, no doubt, insured at Lloyd’s”
(Text 3: June 22nd, 1991:90).
• In a number of the articles there is a pointed (my italics for emphasis)
reference to names as being the “the rich people who provide the market’s

215
 capital” (Text 6: October 19th 1991:95), or the “rich individuals whose capital
supports underwriting” (Text 7: January 18th 1992:69), and the “rich
individuals who provide Lloyd’s capital” (Text 8: February 15th 1992:82). As
time passed there were further pointed references to the names, but in many
instances they were also variously referred to as “those mostly rich
individuals” (Text 11: June 20th 1992:75), then “the once rich individuals” ”
(Text 12: June 27th 1992:84), “the formerly rich providers of capital” (Text 13:
July 4th 1992:68), then the no-longer-so-rich individuals who provide the
market’s capital” (Text 16: October 17th 1992:90), and then finally in the
Mountains text, “the individuals who provide Lloyd’s capital”.

Of further relevance here, and further evidence which can inform a multi-modal
interpretation of the Mountains text, is the fact that the subsequent articles in The
Economist magazine have continued to express some of the attitudes outlined
above in their analyses of the Lloyds’ problems. Good examples of this
‘intertextual future’ for the Mountains text are two Finance department leader
articles which occurred later in the same year as the Mountains text. The first is
titled ‘Sweeping up Lloyd’s messes’ (May 1st, 1993:97-98), and includes a sketch
caricature of Messrs Rowland and Middleton attempting to clean the surface of a
large carpet labelled as ‘Lloyd’s of London’ which, unfortunately, has a great deal
more dirt (labelled for example as ‘litigation’) swept under it. This is of course
suggestive of past actions and problems, and the text verbally discusses and
analyses these in terms of the Lloyd’s business plan, which is also the subject of
the Mountains text.

th
The second text is titled ‘The end of the name’ (September 18 , 1993:77-78), and
includes a sketch caricature and two line graphs within one visual frame. This text
deals with a discussion and analysis of one aspect of the Lloyd’s business plan,
that of including corporate membership and thus ending unlimited liability for
individual names. The sketch caricature depicts the very large Lloyd’s bell being
pushed (rung) by two business-suited arms (attached to a body out of frame), and
this pushed bell as a result striking similarly be-suited individuals (names) rather
violently out of the way. The line graphs combined within one visual frame show
the falling underwriting rates for Lloyd’s in dollars and pounds sterling since 1989
(underwriting rates being indicative of the level of investment by names).
Accordingly, the verbal aspect of the text reports, discusses and analyses the

216
names’ financial problems as a result of unlimited liability, the reduction of
underwriting capacity which occurred as a result, the losses which Lloyd’s has
been experiencing in recent years, and the details and ramifications of allowing
corporate membership in this troubled context. The attitude of the writers may be
summed up in the following reduced extract in the concluding and commenting
part of the discourse: ‘So Lloyd’s future, if it has one, will be as a market
(describes characteristics) …… In short, it will pull Lloyd’s out of the 20th century.
th
’
Provided that is, that it survives the 20 (The Economist September 18th,
1993:78).

It should be pointed out at this juncture that the problems Lloyd’s has been
struggling with, concerning the issue of the names, the open syndicates, and the
rescue plan(s) that Messrs Rowland and Middleton have been trying to carry out,
finally reached its conclusion with the announcement that a recovery plan had
been finally approved. One of the many media outlets announcing this was the
Business News section of The Daily Telegraph newspaper, which carried the
headline asserting that ‘Future of Lloyd’s is Assured’ (August 31st, 1996:B1) in
reference to the successful approval of a rescue plan. This refers not to the 1993
rescue plan mooted in the above articles and discussed in the Mountains text
(which was eventually rejected by the names in December 1993), but with the
subsequent ‘Reconstruction & Renewal’ plan in which all pre-1993 liabilities were
to be re-insured in a new company with the name of ‘Equitas’. This was approved
by over 90% of the names, and this particular article is referring to the Council of
Lloyds’ confirmation of the plan, thus essentially making it irrevocable. The
article is interesting in that it includes a table giving a time-line of events, and uses
a colour photograph of David Rowland with his hands together in a ‘prayer’
gesture and the accompanying caption: ‘Deliverance: Lloyd’s chairman David
Rowland yesterday’.

6.2.2.3 Visual and Verbal Typology (Mode-related)

As mentioned above, the information about the Lloyd’s issue is derived from a
series of nineteen articles published since January 26th 1991, and all were
published in the Finance department of each issue. The articles discussed as

217
subsequent to the Mountains text were also from the Finance department, except
for the newspaper article announcing the resolution of Lloyds’ problems. Of these
twenty-one Economist articles, eight were leader articles (the same as the
Mountains text), and the remaining thirteen were smaller, supplementary articles.

Table 6.3 also shows that like the Mountains text, all the articles except number
thirteen are multimodal, incorporating both naturalistic (eighteen) and
mathematical (six) visual forms. The most common naturalistic visual is the
sketch caricature (nine), followed by the black and white photograph (seven), and
sketch drawings (two), while the most common mathematical visuals are charts
(three), graphs (two) and tables (one). This accords with the identification of the
most commonly used visuals in The Economist magazine given in Chapter One:
these are sketches (mostly caricatures) and photographs (mostly portraits and
scenes), which are supported by mathematical visuals such as graphs, charts. This
also provides support for the views expressed by Ms. Penny Garrett in the graphic
design unit regarding visuals acting as attracts to the readers through the use of
humour and caricaturisation, and as a way to make a seemingly uninteresting topic
area a little more enticing.

The verbal elements of these texts of course vary in their placement on the page,
but they do share some common elements in that they:
• are all in the three-column format.
• are identified as belonging to the Finance department via the large bolded and
capitalised heading (leaders) and the smaller, bold capitalised heading at the
top of the page (supplementary).
• have a single black line at the top of the page.
• have an article ending signalled by a column-wide dark line with a small black
square on the right-hand side.
• have a blurb (leaders).
• a small heading (supplementary) identifying the topic area or place in which
the story occurs, followed by the article headline.
• all started with drop capitals.
• have visuals that are placed and arranged in accordance with the placement of
the columns.

The list of the common visual and verbal elements above establishes The
Economist magazine’s conventions of Mode — all these elements are common to

218
the particular issue in which the Mountains text occurs, and show that it is a
typical text in terms of its visual and verbal typology. There is thus consistency for
the readers in terms of the subject matter and issue addressed, the attitudes
presented, and the mode-specific ways that these are represented on the page.

6.3 Summary
This chapter has been concerned with a description of the context of culture of
The Economist magazine in terms of its general characteristics as a business
magazine, its published policies on its verbal and visual output, its internal
policies on graphic design, and the magazine’s general visual and verbal graphic
design typology. It has also attempted to relate these broad contextual factors to
the Mountains text by examining aspects of its intertextual history, covering the
text’s subject matter and issue addressed, the attitudes expressed towards this
issue, and the ways that the magazine has produced them in visual and verbal
terms. By doing this, an inter-semiotic interpretation and analysis of the semantic
interaction between the Mountains text’s visual and verbal modes can be more
effectively informed, and should as a result produce some interesting and reliable
insights into how this page-based multimodal text, like others in The Economist
magazine, is coherent as a result of the intersemiotic complementarity between the
visual and verbal modes. The analysis of these features will be carried out in the
next chapter.

219
Chapter 7: The Analysis of Intersemiotic Complementarity in the
Mountains Text
7.0 Introduction
This chapter will attempt to analyse and interpret the Mountains text in terms of
evidence for intersemiotic complementarity between its verbal and visual modes.
The insights gained from an analysis of the Mountains text’s contextual features
will be utilised to inform this analysis. The actual analysis of the Mountains text
will involve a detailed application and discussion of the analytical framework as
derived and outlined in Chapter Five.

This chapter will begin with an outline and overview of the procedures which will
be followed in the application of the analytical framework. This will be followed
by three main sections, each pertaining to an intersemiotic metafunction. This
chapter will constitute an attempt to test the applicability of the concept of
intersemiotic complementarity in answering the questions raised earlier about how
the visual and verbal modes work together on the page in The Economist
magazine. It was proposed that intersemiotic complementarity obtains when one
or more of the following features occur. The first is when the ideational meanings
in both modes are related lexico-semantically through intersemiotic sense relations
of repetition, synonymy, antonymy, meronymy, meronymy, and collocation. The
second is when the interpersonal meanings in both modes are related through
intersemiotic reinforcement of address (MOOD), and through intersemiotic
attitudinal congruence and attitudinal dissonance (Modality) relations, and the
third is when the compositional meanings are integrated by the compositional
relations of information value, salience, visual framing, visual synonymy, and
potential reading paths.

7.1 Analytical Procedures

The analysis of the Mountains text that is given in this chapter will involve a very
detailed examination of the ways that the visual and verbal modes project their
respective meanings, as well as to how those meanings relate to each other
intersemiotically on the page. The parameters of the analytical framework have
already been outlined in detail in Chapter Five, and the actual sequence of steps

220
Stage I - Analysis of the Visual(s)
1) Examine the visual(s) in relation to its represented participants (Identification), the represented
processes (Activity/Relational Activity), the qualities or characteristics of the participants (Attributes),
and the represented context (Circumstances).

2) Interpret the meaning of these features according to what they represent in the represented context of
situation, the context of creation, the text’s intertextual history, and its specific context of situation.

3) Based on this interpretation, assign these Identification, Activity, Circumstance and Attribute features a
meaning-based VME label.

§
Stage II - Analysis of the Verbal
1) Using the identified VMEs as the reference point, examine each clause in terms of the particular
Transitivity features which are in some way related semantically to each identified VME i.e. the
participants, processes, circumstances and attributes.

2) Make a series of lexical inventories of the verbal items that are semantically-related to the experiential
meaning of the VMEs.

§
Stage III - Analysis of intersemiotic complementarity
1) Taking each VME as a starting point, examine each lexical item in the inventory.

2) Interpret the semantic relationship between each VME and lexical item in terms of whether the it is
intersemiotic Repetition, Synonymy, Antonymy, Hyponymy, Meronymy, or Collocation sense relations.

3) Label each identified intersemiotic sense relation as ‘R’, ‘S’, ‘A’, ‘H’, ‘M’, or ‘C’.

4) Collate, quantify and interpret the frequency and relative prominence of the identified intersemiotic
sense relations.

Figure 7.1 Procedural steps in the INTERSEMIOTIC IDEATIONAL analysis

221
taken in this involved analysis will be outlined in Figures 7.1, 7.4, and 7.5 in this
chapter (each gives an overview of the detailed analytical steps actually taken). An
important point to note also is that to ensure that the interpretation of the visuals
given in this study are reasonable, they have been checked against the
interpretations of over one hundred people in informal polls taken at various
conference and workshop presentations.

7.2 An Intersemiotic Ideational Analysis of The Mountains Text

This section will be concerned with testing the proposition that ideational
meanings in both verbal and visual modes in the Mountains text are related lexico-
semantically through what has been referred to as the intersemiotic sense relations
of repetition, synonymy, antonymy, meronymy, meronymy, and collocation. The
actual sequence of steps taken in this analysis are outlined in the Figure 7.1.

As outlined in the analytical framework presented in Chapter Five, in order to

examine the intersemiotic ideational features of the sketch caricature and the
mathematical line graphs in the Mountains text and how they interact semantically
with the verbal aspect of the text to realise intersemiotic complementarity, one has
to firstly analyse them in terms of:
• their represented participants, in terms of their identification, or who or what is
in the visual frame (animate or inanimate), including both those which perform
as actors and those which act as recipients or objects (goals) of the action of the
other represented actors (participants).
• the activity, or what action is taking place between the actor(s) and the
recipient(s) or object(s) of that action.
• the circumstances, or the elements which are locative (concerned with the
setting), of accompaniment (participants not involved with the action), and of
means (participants used by the actors).
• the attributes, or any qualities and characteristics of the participants
represented.

These visual features carry semantic properties which have already been referred
to as the Visual Message Elements (VME or VMEs), elements which are realised
by a range of possible visual techniques at the disposal of the writers (see Table
5.17 in Chapter Five). The previous discussion on the contextual aspects of this
text illustrates very clearly that these representational features do not occur in
isolation - the messages the visuals attempt to impart to the viewers are influenced

222
and constrained by what has been written and/or drawn before by The Economist
magazine writers, as well as other myriad sources of reportage and analysis which
the reader/viewers may have experienced in the days and months prior to the
publication of the article. Again, the Mountains text is therefore very much a time
and context-bound text.

7.2.1 Visual Message Elements in The Sketch Caricature

The sketch caricature in the Mountains text is a typical example of the kind of
naturalistic visual that is commonly utilised in leader articles in the Finance
department of The Economist magazine. Starting firstly with an analysis of the
sketch caricature in terms of the Identification of the represented participants, or
by asking the question Who or what is in the visual frame?, and Who or what are
they interacting with?, we find that the most salient represented participants are
easily identifiable to the viewer via the use of verbal labelling, or through
caricatured representations of the faces and features of two well-known public
figures, well-known because they have been referred to and pictured in previous
articles dealing with Lloyd’s problems. Accordingly, this visual displays a very
large round object which is easily denoted as a boulder, but which is also intended
to be interpreted as representing the Lloyd’s insurance market, an interpretation
that is verbally realised by the label “Lloyd’s”. A further connotative
interpretation of the boulder is possible, based on a knowledge of the situational
setting of the text. The boulder identified as “Lloyd’s” can also be interpreted as a
visual metaphor for Lloyd’s current problems.

There are also the paired bodies and faces of two men, the Chief Executive, David
Middleton (foreground), and the Chairman, David Rowland (background), both
realised by the use of caricatured facial approximations. Another important
represented participant is a section of the mountain-side on which the other
participants find themselves, specifically the ledge on the upper slope. This
participant is realised by the use of a profile view of a steep, ridged mountain-side
and ledge to the right of the two men, and the use variegated black shading giving
the profile a granite-like or rock effect. The men are portrayed as Actors who are
interacting with the boulder and the ledge/upper slope of the mountain, both of

223
which are represented participants which can be interpreted as acting as Goal (that
which is acted upon). The aim of the interaction of these represented participants
seems to be for the two men to at least solve the immediate problem of the boulder
by pushing it up to the ledge, which is acting as an interim resting place (the next
slope is the next task). There are therefore four important represented participants,
Lloyd’s, the boulder, the two men, and the upper ledge/slope of the mountain-side,
each of which will be glossed as the VMEs “Lloyd’s”, “Rowland & Middleton
(the two financiers)”, “the Boulder (Lloyd’s problems), and “Upper ledge/slope
(the solution)”.

In terms of the Activity, or asking the question What action is taking place?, we
can see that the action taking place is the action of the two financiers attempting to
deal in some way with the problem posed by a significantly larger, heavy round
boulder which is on the side of this steep mountain, and under the influence of
gravity. This can be interpreted as them either pushing it up, or trying to hold it
back. The size of this boulder in relation to the mountain-side and the two men
acting on it makes it a visual feature with visual ‘weight’, which according to
Arnheim (1988:229) is “the dynamic power in an object by virtue of its
conspicuousness, size, shape, location etc.”. To the viewer, the sight of two men
of smaller stature interacting in some way with a much larger, round boulder on
the side of a very steep mountain conveys a powerful message of struggle, and
suggests that either opposing or restraining the boulder’s natural action could
perhaps be an impossible task. The viewers’ appreciation and sense of this
difficulty is also intensified by the caricaturisation of the boulder to the point that
it is almost perfectly round (making it easier to roll down or to lose control), that
the two men are very close to a ledge, a point where they can perhaps rest
(although notice that there is room only for the boulder, not the men), and that
they are faced with a steeper, almost vertical mountain-side as the next obstacle
(giving the viewers a very clear message about the impossibility of any future
tasks, should the two men somehow overcome the present one).

The most probable interpretation here is that the caricatured Middleton and
Rowland are in fact attempting to push the Lloyd’s boulder up the mountain, an

224
action that is strongly realised visually by the direction of their eyes, which is
upwards and from left to right, looking upwards towards the completion of some
kind of a task. The visual technique used here to realise this is known as ‘visual
lines on the stage’, where an actor’s glance or look can give spatial direction to the
audience’s perception of the action (Arnheim 1974: 28). The positioning and
shape of their bodies also produces this sense of direction and goal-oriented
behaviour, an effect realised by the vectors which can be drawn as a result of the
parallelism produced by their arms, lower legs and again, gaze. The task they are
engaged in may be denoted as pushing the boulder up the mountain slope.
However, the visual metaphor here is that they are trying to deal in some way with
Lloyd’s problems (the boulder), and shows that the men as Actors are interacting
with the boulder on the upper ledge/slope. In this depiction the boulder is a
represented participant acting as Goal, which in SFL terms refers to the “one to
which the process is extended”, or that which is acted upon in some way (Halliday
1994:110). The upper ledge/slope on the hand is a represented participant acting
as Range, which in the SFL sense refers to “an entity which exists independently
of the process but which indicates the domain over which the process takes place”
(op.cit:146). The overall impression conveyed here is one that suggests that
success as a result of this action on the mountainside implies a solution to Lloyd’s
problems.

There is also the impression of effort and strain created visually for the viewer
through their body position and their interaction with the more visually salient
interaction between the heavy round boulder and the mountain-side. This sense is
further realised and reinforced by the puffs of dust and dislodged rock debris
behind the men’s feet, which is suggestive of their stepping forward under great
strain, and perhaps potential danger of slippage and loss of control. There is
therefore one important action being performed here, that of two men
simultaneously attempting to ascend while pushing a boulder up a mountain-side.
This action will be glossed as the VME “climbing - pushing up”.

225
Figure 7.2 A reverse view of the sketch caricature (left - right reading path)

226
These interpretations are supported by the effect that is produced when the visual
is reversed, as seen in Figure 7.2, which shows a reverse view of the sketch
caricature image. The action in the original visual gives the viewer the sense that
the two men are attempting to push the boulder upwards from left to right, while
the reversed version conveys the impression that the two men are attempting to
restrain the boulder from rolling downwards from left to right. Both involve the
sense of effort and strain. The common denominator here in visual terms is the
left-right orientation or reading path to which the generally English-speaking and
mostly Western readers of The Economist magazine are acculturated — an eye
movement which typically follows a left-to-right direction when viewing or
reading (see the discussion of left to right reading path in Chapter Five). This is an
important aspect of visual literacy which is culturally-based and is related to a
sense of visual expectancy.

The Circumstances, or the answer to Where, who with and by what means are the
activities being carried out?, is also interesting. The two men and the boulder as
represented participants in this visual are of course not just placed anywhere; they
are placed or located in a physical situation, and that is most obviously on the side
of a single-ledge mountain-side. While the whole mountain-side is also a salient
represented participant, it is also an aspect of the setting which places the other
interacting participants in a particular location. This combines with the other
features of the visual to give an overall effect of two men struggling with a
boulder on a mountain. It is not just simply a mountain however; it is also a visual
metaphor of the path that the Lloyd’s institution has taken (below the men’s feet),
where it is now (where the boulder touches the mountain), and what would be its
most likely immediate and distant future (the ledge and the steeper slope). The
mountain-side thus connotes Lloyd’s path, and provides a visual narrative of
Lloyd’s past, present and future.

Another aspect of this setting is the fact that this kind of naturalistic sketch
caricature has a highly idealised setting, in that the backdrop of the scene is
completely devoid of any other details which may distract the viewer from the
main action taking place (vegetation, other people, houses etc.). This de-

227
contextualisation of the scene forces the viewers’ attention onto the action taking
place, onto who is involved, and where it is taking place. There is therefore a
circumstance in terms of a setting given which informs the viewer where the
action is taking place, and gives a physical context for this action in terms of what
has, what is, and what will happen — this Circumstance: locative (setting) will be
glossed as the VME “the Mountain (Lloyd’s path - time/place)”.

The Attributes, or the question What are the qualities and characteristics of the
participants?, is concerned with any important quality, character, characteristic, or
property attributed as belonging to a represented participant. In the sketch
caricature the viewers’ appreciation of the difficulty, great strain and perhaps
inherent danger the two men are experiencing is intensified by the boulder being
almost perfectly round (easier to roll down), and the quality of effort and strain
that is reinforced visually through the two men’s body position, and the puffs of
dust and dislodged rock debris behind the men’s feet. A further attribute which
identifies the two men as belonging to the Finance industry in some capacity is
provided by the pin-striped suits they are wearing; this is commonly viewed and
represented as the uniform of those involved with “The City”, or the financial
heart of London — this is of course where the famous Lloyd’s of London
headquarters building is located in London (Clarke 1995:1). All these attributes
are relatively significant, but supplementary features of the represented
participants which work to strengthen the dominant impressions the viewers have
of this scene; as such they may be glossed as the supplementary VMEs
“effort/strain in action”, and “financiers”.

Given these interpretations, the narrative message or focus of this visual can be
summarised as a situation where the two top financial managers of Lloyd’s (the
two financiers) have a very great problem with the Lloyd’s market which is
decidedly difficult to control, restrain, or reverse (the boulder). Should the two
men manage to overcome this present problem, they can perhaps look forward to a
brief but potentially dangerous respite (the narrow ledge), but then they are faced
with a further, almost impossible task (the extreme slope of the mountain-side)
which they alone have to complete, a representation which is suggestive of the

228
need for outside intervention or assistance. When the already-discussed features of
the situational setting of the text are also taken into account, this interpretation is
further supported. For example, the two men are readily identified as Rowland and
Middleton because their images have been used in many articles on Lloyd’s
problems, and the difficult situation portrayed also links with the common
knowledge that Lloyd’s is having severe problems and has an uncertain future.
The whole mountain-side is thus a circumstantial visual metaphor for where
Lloyd’s has been (the lower slope), where it is now (where the boulder is touching
the mountain), and its possible future path and difficulties (the higher slope, ledge
and even higher slope). This draws directly on the Mountains text’s intertextual
history and the readers’ familiarity with the issues.

This point about the readers’ familiarity with the issues relates to the contextual
aspects of this text, specifically the authors’ conceptions of their audience. As
mentioned in Chapter Six, the interview with the Ms. Penny Garret of The
Economist magazine London office revealed that the magazine assumes their
audience ranges from professional economists, business people, to the general
reader who has attained a certain level of education and sophistication. This was
supported by Luhman’s (1998) comments about audience in his Economist
magazine review. The sketch caricature in the Mountains text quite obviously
depicts a scene which draws upon this level of assumed background knowledge on
the part of the readers — the view of a boulder being pushed up a mountain quite
obviously refers to the ancient Greek myth of Sisyphus. As Biedermann (1992:5)
points out, this is one of the stories expressing the punishment in Hades for famed
evildoers:
[such as] the “Rock of Sisyphus” (Sisyphus, who built the city of
Corinth, tried to trick Hades, the God of the underworld, and was
punished by having to roll a boulder uphill, only to have it slip
from his grasp and roll back down again: the torment of eternal
frustration.

229
Questions Visual Message Visual Visual Realisations
Asked Elements Transitivity
Roles
Verbally indicated labelling of “Lloyd’s”
Who or what? Lloyd’s Participant boulder.

Sketch caricature of two business-suited,

middle-aged men identified and
Who or what? Rowland & Middleton Participant (Actors) differentiated by their caricatured facial
(the two financiers) features and accessories (striped financial
suits, Middleton wears glasses).

Sketch caricature of large boulder, shaded to

represent rock but caricatured in that very
Who or what? Boulder Participant (Goal) few boulders are naturally smooth and
(Lloyd’s problems) rounded. Is being acted upon by the two
men. This is a visual metaphor for Lloyd’s
current problems.

Sketch caricature of the upper ledge/slope of

the mountain, which is acting as the range
for the activity of the two men with the
Who or what? Upper ledge/slope Participant (Range) boulder. The ledge/slope exists
(the solution) independently of the process but indicates
the domain over which the action of the two
men takes place. Success in this action on the
mountainside implies a solution to Lloyd’s
problems.

Sketch caricature of two men in

pushing/climbing stances. They are
expending effort and straining which is
realised through their body position in
relation to a more visually salient (heavy)
What action is Climbing - pushing up Process object. Their limbs form parallel vectors in
taking place? (enacting solutions with an upward direction, which matches the
effort) vectors from their eyes to the mountain side
or top, indicating the direction or goal of
their possible action. This is a visual
metaphor for the actions being taken to deal
with Lloyd’s problems, or the solution.

Sketch caricature of a profile of a mountain-

side, shaded to give a granite-like effect, and
Where, who Mountain Circumstance: locative slightly caricatured to convey differing
with, or by what (Lloyd’s path - time/place) (setting) gradients and a ledge. The slope is a visual
means? . metaphor for Lloyd’s path: past, present,
future.

Sketch caricature of the round boulder and

What are the Struggle and effort the puffs of dust and debris, all combined to
qualities and connote struggle, effort and difficulty.
characteristics and Attributes
of the
participants? Financial/banking Striped financial suits connoting membership
executives. in the banking industry.

Table 7.1 The sketch caricature Visual Message Elements (VMEs)

230
The readers are obviously assumed to be familiar with this myth, and to see the
ironic association in the message that the Lloyd’s of London institution is perhaps
an evildoer which will suffer “the torment of eternal frustration” — the two
Lloyd’s men represent Sisyphus (which is perhaps a little unfair since they are the
rescuers, not the original evildoers) and the boulder represents Lloyd’s problems.
This sketch caricature is a statement about what is and has been happening; but it
is also a projection of what the creators of the visual see about the chances of
success for Lloyd’s with its problems, issues which will be discussed in the
following sections.

This general narrative message can be read in this sketch caricature visual by
virtue of the interplay between its VMEs. These, along with their semantic
participant categories and realisations are summarised in Table 7.1.

7.2.2 Visual Message Elements in the Line Graphs

An examination of the line graphs in terms of Identification of the represented
participants, or by asking the questions What are the participants in the visual
frame? and What are they interacting with?, reveals some interesting features. The
two line graphs enclosed in the one frame in the Mountains text are, like the
sketch caricatures, typical of the kind utilised in the leader articles of the Finance
department of The Economist magazine. In examining the visual ideational
features of these line graphs, again, one has to look at what is within the visual
frame, the action of the graphic elements, and the kinds of mathematical visual
relations utilised to present the visual ideational content of the image (these can be
either the derivative [rate of change], the comparative or the proportional
mathematical relations).

The line graphs both deal with the same general topic as the verbal aspect of the
text, and which have been covered in the articles published in the previous two
years. These are the problems that Lloyd’s of London is having, and the business
plan developed to deal with them in the interim. This is verbally indicated by the
sub-headings which refer to active names, and the number of syndicates with open
years. There are also two other represented participants which act together

231
mathematically — these are the (x,y) axes of the two graphs representing a set of
independent and dependent variables related to each other in a derivative (rate of
change) mathematical relationship. The independent (x) variable in both graphic
instances is chronological in nature (time in years), and the dependent (y) variable
is numerical (numbers of names and syndicates). These two variables, the
independent and the dependent, interact with each other in various proportions to
produce two line graphs which can themselves also be considered to be
represented participants. These however will be treated as processes, or
represented actions, because even though they are denoted as two graphic lines on
the page, they do in fact connote the representation of a process, which has been
characterised as “the rate of change” or derivative mathematical relation, and
which suggests variability and differing rates of change. As we shall see below,
this has a further general connotation, that of increasing losses for Lloyd’s. Based
on this, four important represented participants can be identified in this visual,
which will be glossed as the VMEs “Active Names”, “Syndicates with open
years”, “0 - 35,000 (number)”, and “1982 - 1993 (time)”. The VME for “0 -
35,000 (number)” is a conflation of the number of names (in thousands), and the
number of syndicates with open years (in lots of twenty-five).

With regard to the Relational Activity, or What is the mathematical relation

portrayed?, we see that the main message of these two graphs individually and
collectively is to suggest that Lloyd’s is experiencing increasing losses. This is
realised by showing how the dependent variables, the number of active names and
the number of syndicates with open years, have behaved in relation to the
independent variable, which is time. The message, as suggested above, is one of
variability in terms of the derivative (rate of change) mathematical relation.
However, this variability connotes increasing financial loss(es) for Lloyd’s,
because the decrease in names in the years 1988-1993 means that the number of
people who invest their money in Lloyd’s is falling, and the increase in open year
syndicates numbers between the years 1982-1991 means that those who are unable
to settle their commitments in Lloyd’s are increasingly liable to continuous loss
for which Lloyd’s is liable.

232
The basic, general message is therefore money increasingly lost. The left hand line
graph portrays this relationship over the eleven year period from 1982 to the
beginning of 1993 as a steep increase in the number of names investing in the
Lloyd’s market up until 1988 (where it peaked at approximately 32,500 names),
and a steep decline continuing unchanged to the date of the publication of the text
(to a little below 20,000 names). Hence the use of the heading “Slumping…” to
reinforce the message. The right-hand graph portrays the open years to time
relationship only up until 1991, a nine year period. The differing time periods
(eleven and nine years) for the dependent variables is related to the fact that the
accounts for syndicates are reported three years in arrears (to allow ample time for
settlements to be cleared), so the data for early 1993 is not available. Counting
back three years from 1993 means that the data for 1990 is the most recent set of
figures available (information which is given in many of the previous articles on
this issue). The line graph for the period represented shows the number of open
year syndicates rising in the early to mid-eighties, falling for a brief period in
1987-88, but then rising steeply from 1989 on. Despite this variability, the
relationship portrayed is a steadily mounting increase in the number of syndicates
with open years. Hence the use of the heading “… mounting”. There are therefore
two important processes being depicted here, that of two line graphs with a
combined overall graphic focus suggesting increasing losses for Lloyd’s, and two
verbal headings referring to specific graphic foci, those of increases (“slumping”)
and decreases (“mounting”) in dependent variables. These two mathematical
processes will be glossed as the VMEs “Graphic focus: increasing loss”, and
“Slumping … mounting”.

For the Circumstances, or in asking the question Where, what with and by what
means are the relations realised?, the verbally indicated footer “Source: Lloyd’s of
London” works to place the information and the format it is given in firmly within
the sphere of the economic or financial world, and also gives the data an official
imprimatur which indicates the source and certifies its accuracy. There is
therefore a circumstance in terms of a ‘disciplinary’ setting given which informs
the viewer that the quantitative data is from Lloyd’s of London — this
Circumstance: locative

233
Questions Visual Visual Visual Realisations
Asked Message Transitivity Roles
Elements
Verbally indicated visual sub-heading
What are the identifying the quantitative focus of the left-
participants? Active names Participant side line graph. Gives the left-hand visual’s
“topic focus”.

Verbally indicated visual sub-heading

What are the Syndicates with open identifying the quantitative focus of the
participants? years Participant right-side line graph. Gives the right-hand
visual’s “topic focus”.

Two vertical ‘y’ axis’ with points marked

evenly for firstly the number of names
What are the 0 - 35,000 (number) Participant (thousands), and secondly the number (also
participants? verbally indicated) of syndicates with open
years (in lots of twenty-five).

Two horizontal ‘x’ axis’, both beginning

What are the with 1982 and increasing by two year lots,
participants? 1982 - 1993 (time) Participant and both finishing in 1993. Time is also
verbally indicated by “January 1st” in the
sub-heading.

Twin verbally indicated visual headings

What is the referring to vertical movements, and
mathematical Slumping - mounting Process suggesting that the left Line graph is going
relation down, while the right Line graph is still
portrayed? going up. Gives a “point of view” to each
visual.

Two line graphs displaying graphic action,

What is the variations, and the rates of change in the
mathematical Graphic focus: Process dependent variables (active names and open
relation increasing loss year syndicates).
portrayed?

Verbally indicated visual footer. Indicates

Where, what that the data visually represented is from an
with and by Source: Lloyd’s of Circumstance: locative ‘official’ source, in this case from Lloyd’s
what means are London (setting) of London itself.
the relations
realised?

Table 7.2 The line graph Visual Message Elements (VMEs)

234
(setting) will be glossed as the VME “Source: Lloyd’s of London”. A summary of
all these VME features for the line graphs may be seen in Table 7.2.

Now that the VMEs in the naturalistic and mathematical visuals in the Mountains
text have been described and derived, these features will be examined in relation
to how they relate in intersemiotic, lexicosemantic terms to the verbal mode. The
question again is what evidence is there to support the contention that the visual
and verbal aspects of the Mountains text are related to each other in terms of
ideational intersemiotic complementarity? It follows that if they are in the same
compositional page space, and they are both concerned with the same subject
matter, then there should be some evidence or support for some kind of
intersemiotic lexicosemantic relationship. The following sections will examine
these questions by describing how the VMEs are related to each other lexico-
semantically.

7.2.3 The Mountains Text: An Intersemiotic Ideational (Representational)

Analysis
The following discussion examines the Mountains text in terms of the
intersemiotic cohesive relationship between its VMEs and its lexical features.
Starting with the VMEs of both the sketch caricature and line graphs and checking
through the verbal aspect of the Mountains text for semantically-related lexical
items produces several lexical inventories. As explained previously in Chapter
Five, the selection of lexical items is based on the existence of some kind of
semantic relationship to each VME, and that one of these relationships can be one
of the intersemiotic sense relations referred to as Repetition, Synonymy,
Antonymy, Hyponymy, Meronymy and Collocation.

The inventory of those lexical items which are deemed to be semantically related
in some way to the VMEs derived for the Mountains text’s visuals is summarised
in Tables 7.3 (a) and (b) and 7.4 (a) and (b). The procedures for the derivation of
these tables initially involves the division of the Mountains text into consecutive
sentences, starting with the department header, and continuing through to the last
sentence in the article (a total of 67). This division can be seen in Appendix Two.

235
The derived VMEs are then ranged against these sentences in a matrix format, and
the relevant lexical items are entered into the matrix ‘cells’ in accordance with the
appropriate sentence number. As mentioned earlier, this is an adaptation of lexical
string analysis used in text cohesion studies — the fundamental difference here
being that the relationship analysed is between the visual and the verbal modes,
rather than between the lexical items in the verbal as it unfolds.

The reader should note that the VMEs in the sketch caricature which have been
referred to as Attributes: “effort/strain in action” and Attributes: “financiers”, have
each been subsumed under the Process: “Climbing - pushing up (enacting
solutions with effort)” and the Participant: “Rowland & Middleton (the two
financiers)” VMEs respectively. The lexical items relevant to these VMEs have
been included and combined in these categories because they are very closely
related to them. In a sense they work to add additional or supplementary meaning
to either the process portrayed (e.g. the puffs of dust portrayed connoting effort in
the process of pushing up), or to provide more clues as to the identification of the
represented participants, which in this case are the two men (e.g. a viewer would
not only know that they are Middleton and Rowland based on the facial
likenesses, but also because of the supportive attribution provided by the uniform
of a financier or banker, the pin-striped suit). The reader will see these conflations
in the inclusion for example of such lexical items as climbed/scaled (for
“Climbing - pushing up)” and push through/exert control (for “effort/strain in
action”) in the one lexical inventory referred to as “Climbing - pushing up
(enacting solutions with effort)”.

The inventories given in Tables 7.3 (a) and (b), and 7.4 (a) and (b) also include an
identification of the intersemiotic cohesive relations between each VME and each
lexical item. These intersemiotic sense relations have been identified according to
the following symbols:
• Repetition (R) the repetition of the experiential meaning
• Synonymy (S) a similar experiential meaning.
• Antonymy (A) an opposite experiential meaning.
• Hyponymy (H) classification of a general class of something and its sub-
classes.

236
• Meronymy (M) reference to the whole of something and its constituent
parts.
• Collocation (C) words that tend to co-occur in various subject areas.

As explained in Chapter Five, each intersemiotic sense relation describes the

nature of the intersemiotic semantic relationship between the VMEs and each
identified lexical item, a relationship which is expressed diagrammatically in
Figure 7.3:

VISUAL LEXICAL
MESSAGE Í Î ITEM
ELEMENT INTERSEMIOTIC A word or phrase in the
verbal aspect of the text
A represented element of a
visual acting as a Represented
Í SENSE Î which is semantically related
RELATION in some way to the
Participant, Process,
experiential meaning of the
Attribute, or Circumstance Í Î
VME
which can be glossed as a
VME

Figure 7.3 Intersemiotic sense relations

As the inventories in Tables 7.3 (a) and (b) and 7.4 (a) and (b) show, there are
significant and varying numbers of lexical items which relate semantically to the
experiential meaning expressed by the VMEs. The reader would also be aware
that a small number of the lexical items identified as particular intersemiotic sense
relations would seem to be borderline choices in terms of which label should be
assigned to them. This can be seen for example in Table 7.3(a) in the “Climbing-
Pushing up (enacting solutions with effort)” lexical inventory, and in Table 7.4(a)
in the “Slumping … Mounting” lexical inventory. In both these inventories the
lexical items ‘scaled’ (S) and ‘peak’ (C) could be conceivably labelled as ‘scaled’
(R) and ‘peak’ (S or R). The decision about what to label these lexical items in
terms of their intersemiotic relationship to each VME is based on the same steps
that are taken in lexical chain analysis: ‘scaled’ is (S) not (R) because the root
form of “scale” does not occur in the wording of the VME “Climbing-Pushing
up”. This is also the case and for the same reasons with ‘peak’, which is (C) and
not (S or R). Of further note is the fact that although an attempt has been made to
restrict the selection of lexical items to individual or single lexical items, in
accordance with the typical procedures followed in lexical chain analysis, there

237
are however instances of what may termed lexical phrases, or items which in two-
word or more terms could

238
 REPRESENTED PARTICIPANTS PROCESSES CIRCUMSTANCES
S’s Lloyd’s Rowland & Boulder (Lloyd’s Upper ledge/slope Climbing - pushing Mountain
Middleton (the two problems) (solution) up (enacting (Lloyd’s path -
financiers) solutions with effort) time/place)
1 FINANCE (C)
2 climb (R) Mountains (R)
3 Lloyd’s (R) business plan (S)
the market (S)
4 scaled (S) peak (C)
5 Lloyd’s (R)
London's insurance
market (S)
6 the market (S) losses (H) boost (C) last year (M)
Lloyd’s (R) lawsuits (H)
negligence (H)
7 over (C)
8 insurance cycle (C) losses (H)
9 market (S) management team
(C)
David Rowland (R)
chairman (C)
Peter Middleton (R)
chief executive (C)
10 now (M)
11 market’s (S) loss (H)
12 errors and omissions chunk (C)
(C)
(E&O) (C) losses (H)
lawsuits (H)
13 market (S) loss (H)
14
15 losses (H)
losses (H)
16 market (S) losses (H)
17 Lloyd’s (R)
18 market (S) now (M)
under (C)
underwriting (C) down (C)
19 struggled (C)
20 reinsuring (C) losses (H)
21
22 Lloyd’s (R)
underwriting (C)
23 Lloyd’s (R) Rowland (R) business plan (S) background (C)
Middleton (R)
24 Lloyd’s (R) plan (S) future (M)
market board (M) now (M)
25 Middleton (R) losses (H) reforms (H)
26 cost-cutting (H)
stepped up (S)
27 Lloyd’s (R) sacking …staff (H)
agencies (M)
28 member's agencies (M) Mr Middleton (R) push (R)
managing agencies reforms (H)
(M)
29 agencies (M) control (S)
exert (S)
30 member's agencies (M) make economies raised (C) now (M)
(H)
amalgamation (H)
centralisation (H)
31 corporate members higher underwriting spur (C)
(M) market (S) standards (H)
underwriting (C)
32
33 open years (H)
litigation (H)
losses (H)
Table 7.3 (a) Lexico-semantic intersemiotic complementarity - sketch caricature

239
240
 REPRESENTED PARTICIPANTS PROCESSES CIRCUMSTANCES
S’s Lloyd’s Rowland & Boulder (Lloyd’s Upper ledge/slope Climbing - pushing Mountain
Middleton (the two problems) (solution) up (enacting (Lloyd’s path -
financiers) solutions with effort) time/place)
34
35 Rowland (R) thin air (C)
36 open years (H) business plan (S)
37 Centrewrite (M) plan (S)
Lloyd’s (R)
38 Centrewrite (M) claims (H) cope (C) past (M)
market (S) contesting (C)
insurers (M)
39 losses (H)
40 Lloyd’s (R) insulate future (M)
newcomers (H) past (M)
41 ensuring (C) future (M)
falls (A)
42
43 Mr Middleton (R) present a united
front (H)
44 insurers (M) fighting (C)
policy-holders (M)
Lloyd’s (R)
45 Lloyd’s (R) lawsuits (H) business plan (S)
46 managing agencies writs (H)
(M) negligence (H)
47 E&O insurers (M) Mr Middleton (R) minimise the sums
siphoned off (H)
48 insurers (M)
49 E&O insurers (M)
reinsurance (C)
Lloyd’s (R)
E&O (C)
50 litigation (H)
51 Lloyd’s (R) bad publicity (H)
policyholders (M)
52 agencies (M) bankruptcy (H)
53 Lloyd’s (R) lawsuits (H)
54 market (S) down (C)
55 E&O insurers (M) settlement (S) now (M)
Lloyd’s (R)
56 problem (R) solution (R)
losses (H) business plan (S)
57 insurance rates (C) rising (C)
58 Lloyd’s (R) losses (H)
59 underwriting (C) Mr Middleton (R) help (C)
60 losses (H) budget measures help (C) past (M)
(H)
61 market (S) future (M)
new capital (C)
62
63 policyholders (M) Rowland (R) business plan (S)
/Middleton (R)
Lloyd’s (R)
64 two men (S) willingness (C)
65
66 Lloyd’s (R) problem (R) plan (S) head off (C)
litigation (H) solve (R) find a way of meeting
losses (H) (C)
67 high (C)
Table 7.3 (b) Lexico-semantic intersemiotic complementarity - sketch caricature

241
 REPRESENTED PARTICIPANTS PROCESSES CIRCUMSTANCES
S’s Active Names Syndicates with 0 - 35,000 1982 - 1993 Slumping… Graphic focus Source: Lloyd’s of
open years (set of numbers) (time period) mounting -increasing London
loss
1 FINANCE (C)
2 Mountains (R)
climb (S)
3 capital providers (S) Lloyd’s (R)
the market (S)
4 peak (C)
scaled (S)
5 Lloyd’s (R)
London's insurance
market (S)
6 names (R) 2 billion (C) Last year (S) boost (C) losses (R) the market (S)
individuals who 3.3 billion (C) 1989 (M) Lloyd’s (R)
provide Lloyd’s
capital (S)
7 over (C)
8 1990 (M) losses (R) insurance cycle (C)
later years (C) profits (A)
9 names (R) market (S)
10 now (S)
premature (C)
11 1990 (M) loss (R) market’s (S)
3 billion (C) June (M)
1989 (M)
12 names (R) syndicates' (R) 500m-1 billion 1989 (M) losses (R) error and omissions
(C) (C)
names (R) double-counting (E&O) (C)
(C)
13 names (R) loss (R) market (S)
14
15 names (R) syndicates (R) 5000 (M) 1989's (M) losses (R)
names (R) more (C) 1990 (M) losses (R)
few (C)
16 cumulative (C) losses (R) market (S)
17 names (R) Lloyd’s (R)
18 active names (R) 20,000 (M) now (S) down (S) market (S)
8.7 billion (C) 1988 (M) underwriting (C)
40% (C)
19 Names (R) 1989's (M) losses (R)
last year (S)
1990's (M)
20 open years (R) losses (R) reinsuring (C)
syndicate years
(R)
successor year
(R)
21 names (R) open year (R) 84% (C) June (M)
one (M)
figure (C)
100% (C)
22 members (S) open years (R) Lloyd’s (R)
underwriting (C)
23 late April (M) Lloyd’s (R)
24 future (C) Lloyd’s (R)
now (S) market board (M)
25 more (C) last year's (S) losses (R)
since then (C)
26 stepped up (S)
27 Lloyd’s (R)
agencies (M)
28 names (R) syndicates (R) member's agencies
(M)
managing agencies
(M)
29 agencies (M)

Table 7.4 (a) Lexico-semantic intersemiotic complementarity - line graphs

242
 REPRESENTED PARTICIPANTS PROCESSES CIRCUMSTANCES
S’s Active Names Syndicates with 0 - 35,000 1982 - 1993 Slumping… Graphic focus Source: Lloyd’s of
open years (set of numbers) (time period) mounting -increasing London
loss
30 80 (M) now (S) raise (S) member's agencies (M)
31 higher (C) corporate members (M)
market (S)
underwriting (C)
32 names (R)
33 open years (R) three (M) 1990 (M) losses (R)
34 sums (C)
35
36 names’ (R) open years (R)
37 syndicates (R) 1991 (M) Centrewrite (M)
Lloyd’s (R)
38 a lot more (C) past (C) Centrewrite (M)
market (S)
insurers (M)
39 names (R) losses (R)
40 future (C) Lloyd’s (R)
past (C)
41 names (R) some (C) 1990 (M) falls (S)
year (C)
earlier (C)
future (C)
42 capital providers (S) up (S)
names (R)
43 old-year (C)
44 years…open (R) some (C) recent (C) insurers (M)
policy-holders (M)
Lloyd’s (R)
45 names (R) Lloyd’s (R)
46 Members (S) managing agencies (M)
names (R)
47 minimise (C) E&O insurers (M)
sums (C)
48 names’ (R) insurers (M)
49 Not all (C) lose (R) E&O insurers (M)
many (C) reinsurance (C)
1 billion (C) Lloyd’s (R)
all (C) E&O (C)
50
51 names (R) three (M) years (C) Lloyd’s (R)
policyholders (M)
52 many (C) bankruptcy (C) agencies (M)
53 names (R) Some (C) 1982 (M) Lloyd’s (R)
54 down (S) market (S)
55 names (R) now (S) E&O insurers (M)
Lloyd’s (R)
56 third (C) 1990's (M) losses (R)
57 early 1990's (M) rising (S) profitable (A) insurance rates (C)
58 names (R) three (M) years (C) profits (A) Lloyd’s (R)
back-dated (C) losses (R)
1990 (M)
1991 (M)
59 underwriting (C)
60 names (R) past (C) losses (R)
61 future (C) profits (A) market (S)
new capital (C)
62 few (C) profits (A)
63 Names (R) a lot of (C) policyholders (M)
investors (S) Lloyd’s (R)
64 two (M)
65
66 open-year (R) 1990's (M) losses (R) Lloyd’s (R)
67 high (C)

Table 7.4 (b) Lexico-semantic intersemiotic complementarity - line graphs

243
be viewed as single lexical items. Instances include such lexical groupings such
as ‘London’s insurance market’, ‘insurance cycle’, ‘errors and omissions’,
‘managing agencies’, ‘higher underwriting standards’, ‘make economies’,
‘present a united front’, ‘minimise the sums siphoned off’ etc., all of which could
be considered to be single lexical entities, concepts or actions linked
intersemiotically to a relevant VME.

A quantitative summary of the identification, frequency and relative prominence

of the intersemiotic sense relations identified as a result of the analysis of the
VME to lexical items relationship is provided in Table 7.5. It is clear from these
results that in intersemiotic ideational terms, there are numbers of lexical items
identified in the verbal mode which are semantically related with the experiential
meaning expressed by the VMEs (a total of 414 items). There also seem to be
significant patterns revealed in the degree of usage of particular intersemiotic
sense relations in specific inventories. These results will now be discussed in
detail for the sketch caricature and line graphs respectively.

7.2.3.1 The Sketch Caricature

As Tables 7.3 (a) and (b) show, there are four main represented participants in the
sketch caricature which generate four inventories. The derivation and analysis of
these will be discussed below.

Lloyd’s
As Table 7.5 shows (following page), in terms of the total numbers of lexical
items, the greatest number of semantically related lexical items between the verbal
and visual features of the Mountains text occur in the Lloyd’s inventory. This is to
be expected in that the text is about the institution of Lloyd’s of London. This
subject matter can potentially be concerned with many of the issues covered in
previous articles on this topic, such as its background history, its functions, its
component parts, its personnel, its problems, its future, and various other features
which comprise what it means to be considered as a financial institution.
Accordingly, the kinds of lexical items which one would expect to be semantically
related to this

244
Visual Message Element Ant.(A) Syn. (S) Rep. (R) Hyp. (H) Mer. (M) Coll. (C) TOTAL

Lloyd’s 12 19 20 13 64
Rowland & Middleton 1 12 3 16
(the two financiers)
Boulder 2 31 1 34
(Lloyd’s problems)
Upper ledge/slope 10 2 12 24
(solution)
Climbing - pushing up 1 4 2 14 21
(enacting solutions with
effort)
Mountain 1 14 8 23
(Lloyd’s path - time/place)
Active Names 6 26 32
Syndicates with open years 13 1 14
0 - 35,000 (number) 8 29 37
1982 - 1993 (time) 22 24 46
Slumping … mounting 9 1 5 15
Graphic Focus - Increasing 5 18 1 24
Loss
Source: Lloyd’s of London 12 19 20 13 64
Table 7.5 Intersemiotic sense relations in the Mountains text

245
VME would be concerned with these kinds of features, and the inventory
generated reflects this. Firstly, the topic and institution-identifying lexical item
‘Lloyd’s’ is repeated consistently, as are the synonyms for it such as ‘market’, ‘the
market’, and ‘London’s insurance market’ (Lloyd’s is a place where insurance
policies are traded, so it is indeed a market). The text-topic is therefore carried and
reinforced across the modes by the use of intersemiotic repetition of ‘Lloyd’s’ and
the use of these synonymous terms.

The other intersemiotic sense relations utilised in this inventory support the
development of this subject matter, and reinforce the fact that this text is about a
particular financial institution. For example, Lloyd’s is a highly structured
organisation which consists of various people (its ‘market board’,
‘policyholders’), insurance companies (‘insurers’, ‘E&O insurers’,
‘Centrewrite’), and investor organisations (corporate members’, ‘managing
agencies’, ‘member’s agencies’) etc. There is thus a high frequency of
intersemiotic meronymy, which is concerned with part/whole relations (which is
in this case are the functioning parts of Lloyd’s of London). Intersemiotic
collocation is also significant here in that any discussion of a financial institution
such as Lloyd’s invariably requires a discussion of various financial issues and
related areas of concern. The subject area is signalled clearly via the use of
‘FINANCE’ as the department heading and the subsequent usage of such terms as
‘insurance cycle’, ‘errors and omissions’, ‘E&O’, ‘underwriting’, ‘reinsuring’,
and ‘new capita’, all lexical items which could be reasonably expected to co-occur
in a text on a financial topic or a topic about an institution like Lloyd’s of London.

The relatively high occurrence of intersemiotic synonymy and repetition shows

therefore that both the visuals and the verbal aspects of the Mountains text
complement each other in maintaining and supporting the central topic, while the
significant use of meronymy and collocation work interactively to support a
financial discussion which is situated in a much wider, general eco-financial
context. There is thus clear evidence of intersemiotic complementarity in terms of
both modes dealing with the same topic area and pertinent terminology.
Rowland & Middleton (the two financiers)

246
One of the main purposes of the sketch caricature is the identification of
significant represented participants, the characters involved in the action
portrayed, and any salient attributes they may have to assist in this identification.
Both Rowland & Middleton are important characters in the discussion of Lloyd’s
problems and their possible solution - we know that due to their prominence in the
intertextual history of the Mountains text, and by virtue of the frequent use of
intersemiotic repetition of their names, as in ‘David Rowland’, ‘Peter Middleton’,
‘Mr. Middleton’, and ‘Middleton’, the intersemiotic synonymy of ‘two men’, and
the use of lexical items which intersemiotically collocate with the idea of these
two men in control of the represented participant boulder labelled as ‘Lloyd’s’
through ‘management team’, ‘chairman’, and ‘chief executive’ (it could be argued
in fact that both ‘chairman’, and ‘chief executive’ may be interpreted as
intersemiotic synonyms rather than as intersemiotic collocations — that is a
reasonable interpretation to make, being perhaps more a matter of the potential
reader’s background knowledge of just who Rowland and Middleton are and their
positions in Lloyd’s). A further point of note is that there are no lexical items
which semantically link to the represented attributes of the two men (the pin-
striped suits connoting financiers); these attributes are acting in a supportive role
in visual terms, making sure perhaps that if viewers don’t recognise the two men
as Rowland and Middleton immediately, they can see that the two men acting with
the Lloyd’s boulder at least should be there as people concerned with financial
issues. The two modes thus intersemiotically complement each other in terms of
the main actors involved in the Lloyd’s crisis.

Boulder (problems)
The boulder, as already mentioned, is a visual metaphor for Lloyd’s problems
which the two men have to deal with. These are identified and mentioned
throughout the verbal aspect of the text as the decreasing numbers of new names
(or less active names), the increasing number of open syndicates, the potential for
litigation, and the poor performance of the various investor agencies. All of these
have been the subject of presentation and discussion in the previous articles
examined in the analysis of the Mountains text’s intertextual history. Intersemiotic
complementarity between the visual representation of these problems via the

247
boulder and the verbal reference to them is realised via the intersemiotic repetition
of ‘problem’ (the boulder’s connotative meaning), and the extensive use of
intersemiotic hyponymy of kinds or types of problems, as realised through the use
of ‘lawsuits’, ‘writs’, and ‘litigation’ (against investor agencies), ‘negligence’ (of
investor agencies), ‘losses’ and ‘bankruptcy’ (of profits), ‘open years’
(increasing), ‘claims’ (from natural disasters), and ‘bad publicity’ (affecting
confidence). There is thus a clear intersemiotic linkage between the visually
represented and verbally discussed problems.

The Upper Ledge/Slope (the solution)

As mentioned previously, the upper ledge and slope of the mountain, when
interpreted in terms of their role in the interaction between the two men and the
mountain side, can be seen as a visual metaphor for a solution to Lloyd’s problems
i.e. the road to solving Lloyd’s problems involves (at least in part) success at
pushing this boulder up the mountain-side to the ledge and then perhaps further
upwards — if they get up to the ledge Lloyd’s will survive for the moment, but
they still have to “surmount” the remaining problems (the upper slope). The ledge
and the upper slope is a visual metaphor for the solution to Lloyd’s current
problems in that reaching this particular part of the mountain connotes an
effective, successful settlement, while slipping back down the mountain connotes
failure. The visual representation of Lloyd’s attempts to deal with these problems
is complemented through the intersemiotic synonymy evident in the use of
‘business plan’, ‘plan’ and ‘settlement’, (referring to the rescue package set up to
solve Lloyd’s problems), the intersemiotic repetition of ‘solution’, and the
significant usage of intersemiotic hyponymy in the types of actions proposed as
part of this rescue plan, as in ‘reforms’ (in the previous report), ‘cost-cutting’,
‘sacking … staff’, ‘make economies’, ‘amalgamation’, ‘centralisation’, higher
underwriting standards’, ‘insulate newcomers’ (from the past problems), ‘present
a united front’ (against claimants), ‘minimise the sums siphoned off’, and ‘budget
measures’. These are all types of solutions which have been discussed in previous
articles in The Economist magazine, and which can be subsumed under the super-
ordinate classifier “solution”, as visually connoted by the ledge and upper slope of
the mountain.

248
Climbing - pushing up (enacting solutions with effort)
The actions of the two men, that of “Climbing - pushing up”, represents their
attempts at enacting solutions, and as mentioned previously there is a sense
visually created that their actions involve making an effort or struggling with
Lloyd’s problems. The action portrayed is reinforced intersemiotically at the very
beginning of the verbal aspect of the text by the intersemiotic repetition of ‘climb’,
followed by the intersemiotic synonymy of ‘scaled’ and ‘stepped up’. The
supplementary attribute of effort being expended is also reinforced by the
intersemiotic collocation of ‘boost’, ‘struggled’, ‘control’, ‘exert’, ‘spur’, ‘cope’,
‘contesting’, ‘ensuring’, ‘fighting’ and ‘willingness’, all lexical items which could
be reasonably expected to occur to varying degrees in any discussion of a topic
involving the expending of commitment, effort, struggle and strain. The
metaphorical meanings of these two conflated VMEs is important, in that the
lexical items semantically related to the actions portrayed and their manner of
being performed describe aspects of Lloyd’s problems (as in: ‘Names who
struggled to pay’; insurers are fighting policyholders’) and the actions being taken
to solve them (as in: ‘cost-cutting … is being stepped up’; ‘the centre may exert
some control’; ‘spur higher underwriting standards’; willingness to listen’).
These intersemiotic lexical items tend to semantically mirror the represented
actions provided by the sketch caricature both in type and intensity, and they also
refer to subject matter that has been treated in the texts previously published.

Mountain (Lloyd’s path - time/place)

The profile of the mountain denoting a circumstance of setting (where the action
takes place) is not only important in terms of the visual meanings projected by the
two men’s (or Lloyd’s) situation, what they are doing, and how they are doing it.
The visually represented mountain also projects metaphorical meaning as a visual
narrative showing the two men’s (or Lloyd’s) past path, present situation and
possible future. There is therefore a chronological connotation or metaphor
projected, where the side of the mountain and the mountain itself connotes a
narrative of the past, present and future. The mountain as denoted setting is
announced at the very beginning of the verbal aspect of the text by the

249
intersemiotic repetition of ‘Mountain', followed by ‘peak’ (which forms an
intersemiotic collocation with the visually represented mountain), and then
‘chunk’ (which forms an intersemiotic collocation with the visually represented
boulder). The lexical items ‘under’, ‘down’, ‘thin air’, and ‘high’ also form
intersemiotic collocations by being related to the positional and descriptive
aspects of mountains and their physical settings. These work to supplement the
visual setting represented in the sketch caricature. The mountain as a narrative
metaphor for Lloyd’s history and its future prospects however, is also supported
intersemiotically by the time references sprinkled throughout the verbal text. Since
Lloyd’s ‘story’ can be considered in terms of a period of time (past-present-
future), then the intersemiotic relationship between the visual story and the aspects
of the verbal story referred to in the text would seem to be one of intersemiotic
meronymy (the whole being the period of time). The inventory bears this out
through such repeated use lexical items as ‘last year’, ‘now’, ‘future’, and ‘past’,
all which could be considered as references to different parts of the whole time
period. This is supported by the intersemiotic collocational use of ‘background’.

7.2.3.2 The Line Graphs

As Tables 7.4 (a) and (b) show, there are seven main represented participants in
the line graphs which generate seven clear inventories. The derivation and
analysis of these will be discussed below.

Active names and Syndicates with open years

The main focus of the line graphs is the behaviour, over time, of the number of
“Active names”, and the number of “Syndicates with open years”’. These are
important represented participants, an importance reflected by the fact that they
have been referred to, discussed and analysed in the two years prior to the
publication of the Mountains text, and by the degrees to which they are
semantically related to the verbal part of the Mountains text. Both the participants
represented in these line graphs are intricately concerned with the problems that
Lloyd’s has, is having, and may have in the future. For Active names this is
carried and reinforced strongly by using intersemiotic repetition of ‘names’,
‘active names’, and the supplementary use of intersemiotic synonymy with

250
‘capital providers’, ‘individuals who provide Lloyd’s capital’, ‘members’, ‘capital
providers’, and ‘investors’. In the Syndicates with open years intersemiotic
repetition is also used via ‘syndicates’, ‘open years’, ‘syndicate years’, ‘successor
year’, ‘years …open’, and ‘old year’.

1982 - 1993 (time) and 0 - 35,000 (number)

The number and type of intersemiotic sense relations in the inventories generated
for the line graphs’ independent (time) and dependent (number) variables are also
interesting. As mentioned above, the graphs focus on the behaviour, over time, of
the number of “Active names”, and the number of “Syndicates with open years”.
With regard to time, the line graphs in the Mountains text deal with the period
1983 to 1991/3, which may be considered as a closed set of years. As this text is
very much concerned with a discussion of what has happened, what is happening
now, and potentially what may happen in the future, any lexical reference to this
particular time period is an instance of intersemiotic meronymy because it is a
reference to a part of that set of years. The instances of intersemiotic meronymy in
the text such as ‘1989’, ‘1990’, ‘June’, ‘1988’, ‘1990’s’, ‘late April’, ‘1991’, and
‘1982’ are all segments of the time period 1982 to 1993. Discussion centred
around the specific set of years is supported by general intersemiotic collocation
with lexical items such as ‘last year’, ‘later years’, ‘now’, ‘premature’, ’future’,
’since then’, ’past’, ‘earlier’, ‘recent’, and ‘back-dated’.

With regard to the treatment of the dependent variable number, the line graphs in
the Mountains text deal with the set of numbers ranging from zero to 35,000,
which may also be considered as a closed set. The instances of intersemiotic
meronymy such as ‘5000’, ‘20,000’, ‘80’, ‘one’, ‘three’, and ‘two’
intersemiotically complement this, as does the significant usage of intersemiotic
collocation in ‘2 billion’, ‘double-counting’, ‘more’, ‘few’, ‘cumulative’, ‘40%’,
‘figure’, ‘sums’, ‘a lot more’, ‘some’, ‘minimise’, ‘third’, ‘a lot of’ etc. Instances
of these kinds of lexical items may be expected to accompany any generalised
discussion of money or numerical-related matters. An important issue arises here
however. This concerns for example the inclusion of the lexical items ‘three’
(sentence 33) and ‘two’ (sentence 64) — as the reader can see, this usage in the

251
verbal aspect of the text refers to the three problems Lloyd’s has to deal with
(‘open years, litigation and financing the 1990 losses’) and the two men
(‘Rowland’ and ‘Middleton’), not specifically to the set of numbers referred to in
the VME. This raises an interesting issue which could be the subject of further
more delicate analysis: the selection of lexical items not only from the point of
view of their direct intersemiotic semantic relationship to the VMEs, but also in
terms of an analysis of the semantic roles they play in the clause, and whether the
clausal elements actually relate semantically to the VMEs.

Slumping … … mounting and Graphic Focus - increasing loss

The portrayed action or processes in the line graphs are indicated by the verbally
labelled ‘Slumping … mounting’ graphic headings, both of which reinforce the
intertextual history of the whole text and the underlying message focus of the two
graphs — that of the increasing losses Lloyd’s has been incurring as a result of the
decrease in names and the increase in open-year syndicates. Both these processes
are expressed in the VMEs “Slumping … mounting” and “Graphic focus -
increasing loss”. The general semantic concepts of ‘slumping’ and ‘mounting’ are
signalled immediately via the use of intersemiotic repetition of ‘Mountain’, and
continued throughout the text via the intersemiotic synonymy of the set ‘climb’,
‘scaled’, ‘down’, ‘stepped up’, ‘raise’, ‘falls’, ‘up’, and ‘rising’. Intersemiotic
collocates of ‘slumping’ and ‘mounting’ include such lexical items as ‘peak’,
‘over’, ‘higher’, and ‘high’.

The VME “Graphic focus - increasing loss” is, as explained above, the underlying
focus of the information presented in the two line graphs, and this focus is
consistently reinforced throughout the verbal aspect of the text by the
intersemiotic repetition of the lexical items ‘losses’, and ‘loss’. This is supported
by the strong intersemiotic collocation of ‘bankruptcy’, and ‘losses’ is contrasted
with its logical opposite, the intersemiotic antonymy generated by ‘profits’ and
‘profitable’.

Source: Lloyd’s of London

252
This verbal footer giving the source of the data acts as a Circumstance: locative,
giving the setting for the graphic action. It is also a VME that is concerned with
the main topic area of both the visual and verbal modes in the Mountains text —
the institution of Lloyd’s of London and its problems. It is first and foremost
acting as an identifier of the source of the graphic information displayed; however,
in the verbal aspect of the text there is no specific lexical reference to it as that
source. As the main subject matter of both the modes however, the lexical items
produced in the inventory for this VME mirror those listed in the inventory for the
Lloyd’s inventory for the sketch caricature, and thus will not be repeated again
here.

Looking at these results in terms of the quantitative data shown in Table 7.5, some
interesting comments can be made about the nature of the intersemiotic
complementarity between the visual and verbal modes in the Mountains text.
Firstly, the greatest numbers of instances of intersemiotic complementarity
occurred in the two inventories concerned with the general subject matter of the
text, that of the institution of Lloyds of London. Both the “Lloyd’s” and “Source:
Lloyd’s of London” inventories combined accounted for 31% out of a total of 411
lexical items. This is relatively unsurprising, but it does confirm that both visual
and verbal modes do ‘work together’ on the page, at the very least in terms of the
general subject matter — this intersemiotic complementarity is realised mainly
through a significant usage of intersemiotic repetition and synonymy,
intersemiotic sense relations which perform the function of introducing and
maintaining the topic and subject matter.

The most significant VMEs next were the dependent and independent variables of
the line graphs, the period of time (1982-1993) and the set of numbers (0-35,000).
The former accounted for 11% of the lexical items, while the latter accounted for
9% of the lexical items. Both work in concert with the verbal aspect of the text in

253
Stage I - Analysis of the Visual(s)
1) Examine the visual(s) in terms of viewer address (presence or absence of gaze), levels of involvement
(horizontal angle), power relations (vertical angle), and social distance (size of frame) in the visual
Mood system.

2) Examine the visual(s) in terms of contextualisation (full to absence of background), the degree of
representing or abstracting detail, texture, illumination (light and shade), and colour saturation,
differentiation, or modulation in the visual Modality system.

3) Interpret the meaning of these features in terms of the position in which the viewer is placed: whether
information is offered or questions asked, the position of the viewer in involvement/power/social terms,
and the encoded acceptability of what is represented.

§
Stage II - Analysis of the Verbal
1) Examine the text’s Mood features in terms of the ordering in the main Subject/Finite constituent of the
clause complexes, to ascertain the ways that the readers are being addressed.

2) Examine the Modalisation features: where there is argument about the probability or frequency of a
Proposition, and any comments that may also be made.

3) Look at the ways that the speaker’s attitudes to these Propositions are realised through the use of such
features as attitudinal Epithets.

§
Stage III - Analysis of intersemiotic complementarity
1) Examine the multimodal text in terms of how both the verbal and visual elements address their
reader/viewers in combination (offering information, asking questions?), as well as how mode-based
attitudes are expressed about the truth, credibility, probability, or frequency of their propositions.

2) Interpret the semantic relationship between each mode in terms of whether there is Reinforcement of
Address, Attitudinal Congruence, or Attitudinal Dissonance.

Figure 7.4 Procedural steps in the INTERSEMIOTIC INTERPERSONAL analysis

254
terms of topic maintenance, or in ideational intersemiotic complementarity with
each other in relation to discussing Lloyd’s problems in chronological and
numerical terms. This topic maintenance is further reinforced when the other
VMEs are considered, because they are all concerned with sub-topics of the main
subject matter, and as such are further confirmation of ideational intersemiotic
complementarity between the visual and verbal modes in dealing with the central
topic area, the condition and problems of Lloyd’s of London. Finally, much of
what the Mountains text deals with has been presented and discussed before in
previously published articles, so this text is not an isolated textual instance but is
one of a progression of texts — it is intertextually connected and has a history
which also influences how it may be interpreted intersemiotically.

7.3 An Intersemiotic Interpersonal Analysis of The Mountains Text

The examination of the interpersonal features of the sketch caricature and the
mathematical line graphs in the Mountains text involves, in line with the social
view taken in this study, a look at the ways that relations between the visual and
the viewer are represented. As explained in the Chapter Five, the ways in which
the producer and viewer of a visual are placed socially in relation to each other is
important because this can affect the topic of the visual, the ways that it is read,
and the ways that it is interpreted. In the socially-determined context of The
Economist magazine, an examination of the ways that the visuals in the Mountains
text represent interpersonal features in the differing visual coding orientations will
require an analysis of the ways that they address the viewers (in terms of visual
MOOD “speech acts”), the degrees of involvement and power relations encoded
between them (in terms of visual point of view), the social distance that is
produced, and the attitudes that are presented (in terms of visual Modality). The
ways that these visual meanings are related intersemiotically to the verbal aspect
of the text can then be examined to test the proposition that interpersonal
meanings in both modes in the Mountains text are related through intersemiotic
reinforcement of address (MOOD), and through intersemiotic attitudinal
congruence and attitudinal dissonance (Modality) relations. The actual sequence
of steps taken in this analysis are outlined in the Figure 7.4.
7.3.1 Visual Address (MOOD): Offers and Demands

255
As already discussed in Chapter Five, the four primary speech functions proposed
by Halliday (1995) can also be used to interpret the interpersonal meanings in
visuals in terms of whether these forms of address are offers or demands. Offers
and demands were broken down into:
• offering information (a statement that can be agreed with, acknowledged or
contradicted).
• demanding information (questions which can be answered or disclaimed).
• offering goods and services (offers which can be accepted or rejected).
• demanding goods and services (commands which can be undertaken or
refused).

It was also stated that when considering these forms of address in visual
communication, these categories are not so easily applied — visuals often need
verbal support to make the nature of the speech function clear, as in what looks
like a visual offer of goods and services supported by a verbal contact address in
an advertisement, or the verbal reinforcement provided by a printed question to
complement a questioning facial expression (Kress and van Leeuwen 1990:30).
Halliday’s (1985:70-71) distinction between the PROPOSAL in the exchanging of
goods and services, and the PROPOSITION in the exchanging of information was
also considered, with the point made that while it is an important distinction in
spoken language, in page-based text like that produced by The Economist
magazine it is not goods and services that are exchanged with its readers, but
information (if an offer of goods and services seems to be being made, it is usually
an advertisement, and in the form of information about the offer - the reader can’t
physically pick up the offered goods and services during the actual discourse act
as in a spoken exchange, but can at a later date if the offer is accepted). Given
these points, the analysis of The Economist magazine here will consider how
intersemiotic complementarity in The Economist magazine is realised through the
ways that its propositions are addressed to its readers via questions and statements
in the exchange of information. This is a concern with forms of Visual Address
(“Visual Speech Acts”), or the ways that The Economist magazine addresses its
readers in order to exchange (or offer) information.

In the Mountains text, the sketch caricature and the mathematical line graphs can
be interpreted as offering information via the speech function of making

256
statements. In the sketch caricature the absence of any gaze or facial expressions
towards the viewer suggests that it is simply offering information. There are no
indications of a question being asked (realised by vector drawn from some point
of origin to the viewer’s face), nor gestures which command (realised for example
by a pointed finger forming a vector to the viewer). There are no vectors which
can be drawn from the represented participants directly towards the viewers; all
the vectors are concerned with the represented participants within the visual
frame, and relate to the ways that they are interacting with each other. The sketch
caricature is therefore a portrayal or a scene on which the viewer can look with
really no requirement to react other than to agree with it, or to either acknowledge
or contradict its existence or veracity as an acceptable scene (some elements of
this will be discussed in the section on visual modality). This interpretation is
corroborated by the verbal support provided by the label “Lloyd’s” which simply
labels or identifies the boulder as metaphorically representing the Lloyd’s
insurance market, and the interplay between the horizontal and vertical angles
which affect the level of involvement by the viewers and the power relations
encoded between them and the represented participants respectively (discussed
more fully in the next section).

In the mathematical line graphs, there really is no question that they are offering
information. The represented participants form no other relationship to the
viewer(s) than being simply a display of numbers and graphic lines which indicate
some inter-related information or data. The viewer is not asked anything about the
information, but basically offered information which can be agreed or disagreed
with, or acknowledged or contradicted.

7.3.2 Visual Point of View: Involvement

As pointed out in Chapter Five, the level of involvement by a viewer with a visual
is realised by horizontal angle, which is concerned with the interrelationship
between two frontal planes: the frontal plane of the constructor of the visual and
the frontal plane of the represented participants. These planes can be aligned
parallel to each other, or diverge by forming an angle of varying degrees with each
other. A visual can therefore have either a frontal or oblique point of view, the

257
oblique point of view being a continuum of obliqueness according to the angle of
the divergence. The frontal angle is a statement of inclusion between the
constructor and the visual, while an oblique angle encodes degrees of commitment
to the subject or represented participants, stating to varying degrees of intensity
the level of inclusion. The right-angled or perpendicular oblique angle would be
suggestive of viewing a scene with no involvement at all beyond stating that ‘this
is a scene”.

The sketch caricature in the Mountains text is a naturalistic visual which, in terms
of the horizontal angle, has an extreme oblique point of view. The frontal plane of
the viewer of this visual (which is the same as that taken by its drawers) is
perpendicular to the frontal plane of the most visually salient represented
participants, the two men and the boulder. There is thus a clear absence of
involvement with the scene on the part of the viewer, not by choice (although that
may be the case), but by the lack of involvement coded (or loaded) into the visual
by its constructor(s). It is a scene which shows action happening, and because of
the extreme obliqueness of the angle puts the viewers in the place of those who
are, as it were, watching from the sidelines — it is action played out on a tableaux
which requires nothing of the viewer except to observe the scene. This accords
with the classification of the sketch caricature as a visual offer of information
which can be agreed or disagreed with, acknowledged, or contradicted.

The line graphs in the Mountains text are also viewed from an extreme oblique
point of view, which is really the only way that they can be viewed, since they are
mathematical visuals representing their meanings on a two-dimensional plane. All
viewers are forced to view the information portrayed in an abstract graphic form
as an offer of information with which they can agree or disagree, acknowledge or
contradict. Unlike the naturalistic visuals, into which their constructors can encode
various degrees of involvement through degrees of obliqueness, the mathematical
visuals are because of their scientific/technological coding orientation restricted to
a two-dimensional plane which leaves no other choice for the viewers than to be
an uninvolved observer of a statement of information or data. There is really no
way that this type of visual can include the viewer. Even if some element is

258
introduced giving the mathematical visual some resemblance to a three-
dimensional form and perhaps an opportunity to lessen the severity of the oblique
point of view, as the line graphs do with the use of shadowed boxes giving the
impression of a couple of raised pages on a featureless plane, the two-dimensional
nature of the graphic plane (pages) is still forcing the viewers to view the
information from the most extreme oblique angle. It really makes no sense to look
at the visuals from the point of view of the frontal plane of the represented
participants (the line graphs), since all that could be seen would the edge of the
page on which they have been drawn.

7.3.3 Visual Point of View: Power Relations

The power relations between the viewers and the represented participants in a
visual are encoded in the vertical angle formed between them. This was referred to
in Chapter Five as related to aspects of cinematography where the viewers of film
are positioned to react to the participants in a particular shot according to whether
they are looking down to, up to, or at eye-level with them. This produces three
power positions: a high angle, a low angle and an eye-level angle. The high angle
forces the viewers to look down on the represented participants, which is
suggestive of a superiority to them or of their insignificance, a low angle forces
the viewer to look up to them, which is suggestive of an inferior position to them,
or of their magnificence, and an eye-level angle is suggestive of a sense of
equality between the viewer and the represented participants. Both the sketch
caricature and the mathematical line graphs in the Mountains text form an eye-
level angle between the viewer and the represented participants, and therefore do
not place the represented participants in either a superior or inferior position. This
is in line with the previously mentioned interpretations of the visuals as offering
information, and as offering this information from an oblique point of view, both
of which encode a positioning for no involvement.

7.3.4 Visual Point of View: Social Distance

A further aspect which is specific to naturalistic images (not to mathematical
visuals) is the degree of social distance encoded between the represented
participants and the viewer(s), as realised by the size of frame. As explained in

259
Chapter Five, in television production the size of frame affects how much of the
human body is shown in the visual frame, giving therefore the close up, the
medium shot, and the long shot etc. These different kinds of shots have a parallel
with the varying distances between people when they talk to each other face to
face, where it can be intimate or friendly, or unknown. Contextually-based
distances may involve interaction in specific social and public situations, where a
well-known, familiar or unknown public figure is delivering a speech to an
audience. In the sketch caricature, the degree of social distance between the
represented (human) participants can be characterised as a long to very long shot,
where their full figures are shown and a significant amount of the physical context
in which they are placed is also portrayed. This long to very long shot has the
effect of making the figures portrayed seem to be part of a larger scene, and lends
weight to previous interpretations which suggest that what the sketch caricature is
presenting is a scene which asks the viewer to do very little in terms of
involvement with the represented participants, beyond simply observing the
portrayal of what is going on. In relation to the mathematical line graphs, we can
see that the size of frame differences really do not have any application, since
choosing a long shot or close up shot does not have any impact on how much of
the graphs is shown — if using a close up shot meant that some part of the graph
was missing then the visual would cease to be a valid mathematical representation.

Thus, up to this point we have two visuals which basically address the viewers
with visual statements, and make no demands on them to be involved in any way
beyond being accorded at the very least a neutral or equal status. The viewer is
therefore treated as a peer who is required to either accept or reject the offers of
information made. These offers are not simply made in terms of bald facts and
ideas however; they are made from certain viewpoints which encode various
attitudes.

7.3.5 Visual Modality: Viewer Attitudes

As was shown in Chapter Five, in the visual semiotic code visuals can, like
language, also be interpreted in terms of the truth, credibility, and probability of
what they represent to the viewer(s), and the information they offer can also be

260
affirmed or denied according to whether something is or is not, is real or unreal, as
well as whether other possibilities exist which can express degrees of certainty or
uncertainty (where perhaps something could happen), or of usuality, (where
something might sometimes happen, but not always). The viewer attitudes
encoded in both forms of visuals in the Mountains text will now be considered in
turn.

7.3.5.1 The Attitudes in the Sketch Caricature

The interpretation of the degrees to which a visual is considered to be real or
unreal, credible or incredible, possible or impossible depends in the first instance
on its visual coding orientation. The sketch caricature is one type of visual from a
continuum of naturalistic visuals which are considered to be real/unreal or
possible/impossible depending on the degree of accuracy of their representation of
reality. This continuum was outlined in Chapter Five as a categorisation for
describing the various forms of visuals extent in The Economist magazine in the
naturalistic coding orientation, which consisted of the colour photograph, the
black/white photograph, the sketch drawing, the sketch caricature, and the line
sketch. This visual coding orientation is concerned with the varying degrees to
which naturalistic visuals (and by default their drawers) attempt to portray 'natural'
images, visuals which the members of a particular cultural setting would agree to
be a recognisable form of a representation of reality, as viewed by the human eye.

The sketch caricature in the Mountains text, like many others used in The
Economist magazine, is an attempt to represent familiar objects, entities, scenes,
characters or actions (real or fictional) which are easily recognisable to the
viewers, but are abstracted via caricature. As such it is not a totally accurate
representation, but a stylistic drawing form in which the main features of the
represented participants have been emphasised to present them from particular
attitudinal viewpoints, ones which the drawer(s) hold and wish to convey to any
potential viewer(s). The sketch caricature in the Mountains text gives the
viewer(s) more freedom to agree or disagree with the portrayal, to acknowledge it,
or to contradict it. It is therefore a ‘suggested’ or ‘loaded’ interpretation of reality,
and as such carries a lower modality in terms of its representation of ‘truth’ than

261
photographs or realistic sketch drawings. What is presented is not a true
representation of reality, but a whimsical version of it. The absence of colour and
the abstraction away from many physical details in the sketch caricature which
could have been included, such as additional facial and physical features, and
landscape features such as plants, trees etc., also adds to this relatively low
modality. The emphasis in the sketch caricature is therefore not on accuracy or
realism, but on presenting a subject in such a way that the main features are
caricatured, and so drawing attention to the attitudes that the drawers wish to
express.

The analysis of the ideational features of the sketch caricature shows that it
projects a number of visual messages to its viewer(s), and as discussed above, its
presentation in caricatured form accords varying degrees of credibility to the
represented participants and therefore the visual messages, as well as bestows on it
certain attitudes. These messages of credibility and associated attitudes are
basically centred around:
• the two financiers (representing Rowland and Middleton)
• the boulder (representing Lloyd’s problems)
• the men’s climbing/pushing up actions (representing the two men strenuously
enacting the solutions to Lloyd’s problems)
• the mountain-side (representing where they have gone, where they are now,
and their future path)

The scene presented in the sketch caricature is not a realistic one; one would
hardly see in the real world two be-suited financiers trying to push anything up a
mountain-side, let alone one that is so much larger than them, and on a sharply
steepening slope. It is an impossible situation, which allows the focus of the visual
to be on the attitudes that the constructor of this visual hope to impart to the
viewers. It is an offer of information about a version of suggested reality. A closer
examination of each of these represented participants reveals some interesting
attitudes or dispositions. These are summarised in the following sections.
Middleton & Rowland (the two financiers)
Both Rowland and Middleton are only slightly caricatured in that the pin-stripe
suits they are wearing are believable as financial/banking suits, their bodies are in
proportion, and they are carrying out actions which they both could realistically

262
perform in some situations — that of pushing up something, together. They are
not made to look silly or comical in their general appearance, although the added
detail of the pinstripe pattern of their suits clearly labels them as financial ‘types’,
and may relate to the drawer’s own perceptions of the attitudes his or her potential
viewing audience may hold about people from that profession. In a sense the
attitude to be adopted by the viewer is to be largely neutral in that he or she is
being required to have an attitude of recognition. This attitude is reinforced by the
relatively larger than usual size of their heads, the addition of such features as
glasses, and the sketched facial features which allow them to be more easily
recognised and differentiated.

Boulder (Lloyd’s problems)

The boulder, metaphorically representing Lloyd’s problems, is very highly
caricatured in the sense that real boulders are very rarely perfectly round, and are
very rarely portrayed interacting with much smaller people on the side of a steep
mountain. The viewer here is not meant to take this boulder seriously, but is meant
to associate it with Lloyd’s and its problems — an attitudinal association realised
by the verbal label “Lloyd’s”, a technique used to ensure that the message gets
across to the viewer(s). The viewer is meant also to see these problems as having
large dimensions, which is an attitude realised by the boulder’s visual salience in
relation to the other represented participants. The constructor of this visual is
therefore inculcating in the viewer an attitude that Lloyd’s problems (the boulder)
are enormous, and that the smaller represented participants it is interacting with
have a difficult task ahead to handle or control it.

Climbing-pushing up (with effort)

The actions of the two men in attempting to deal with the enormous, weighty
problem of Lloyd’s are believable human actions, and the body postures portrayed
are also accurate and believable. However, the situation is not a realistic one in
that two financial managers would hardly be engaged in this kind of activity on a
mountain-side, dressed in the ‘uniform’ of bankers, the pin-striped suit. The
message here is the difficulty of the (ascending) action, which involves effort and

263
struggle. The attitude portrayed in the visual is therefore one of difficulty, of
struggling with the huge, unwieldy problems.

The mountain-side (lower slope, ledge, upper slope)

The mountain-side presents the viewer with a highly caricatured representation. It
is not a believable mountain, but it suggests to the viewer a visual metaphor of
what the two men may encounter as they attempt to deal with the unwieldiness of
the Lloyd’s (problems) boulder. The attitude that the viewer is supposed to have in
relation to the scene on this mountain-side is that it is difficult and bordering on
impossible, a difficulty and impossibility realised by the variations in the degrees
of the slope of the mountain. The viewer is asked to understand that although the
two men may have managed to survive their troubles so far, they may be faced
with a further, steeper slope, suggesting a very uncertain future for both Lloyd’s
and perhaps the two men. By using the slope of the mountain-side, the creator of
this visual has loaded in an attitude regarding Lloyd’s and the two men’s chances
of survival. If the men fail in their task of pushing this boulder up the slope (i.e. to
solve Lloyd’s current or immediate problems), the boulder will roll over them, and
Lloyd’s problems will become uncontrollable, running away down the mountain-
side (it could kill them, meaning that they would lose their jobs). If they succeed
in reaching the next ledge (solve the current problems - enact the solutions) they
will then be faced with an even steeper slope which they may attempt to scale, but
may even more likely lead to theirs and Lloyd’s destruction (insurmountable
problems). This is reinforced again if the viewer takes into account that the ledge
has just enough room for the boulder, but not for the two men. The attitude
projected to the viewer(s) is therefore one of doubt about the future, and not one
of optimism, but of pessimism. Even if they solve the current problems, they face
almost certain failure soon after.

7.3.5.2 The Attitudes in the Line Graphs

The line graphs, as mathematical visuals, are on the other hand instances which
are derived from the scientific/technological visual coding orientation. As
discussed in Chapter Five, in this orientation the level of credibility does not rest
on the degree to which something is accurately represented, but on how efficiently

264
it represents data in a quantitative form. What is ‘real’ in this coding orientation is
dependent on how much the visual conforms to the accepted scientific and
mathematical norms for presenting data in a visual format in the various contexts
in which it may occur. This orientation therefore focuses on how phenomena are
counted, weighed and measured, and how these measured features interact with
each other in an abstracted, two dimensional plane. What is ‘real’ can be
interpreted in terms of whether it effectively utilises these accepted norms of
mathematical interpretation of interacting data. There is also a continuum of
scientific/technological acceptability which, depending on the degree of use of
certain kinds of scientific/technological visual techniques, places the viewer(s) in
various attitudinal positions. These attitudes will vary across different contexts
and disciplines, and the means of graphic representation in some of the ‘hard’
sciences like physics and chemistry will differ to those that are acceptable to a
discipline like economics or accounting.

As shown in Chapter Five, the kinds of mathematical visuals used in The

Economist magazine are largely line graphs, bar graphs, pie charts, and tables.
These kinds of visuals vary in the amount of data they show and the ways that are
deemed to be appropriate by the members of the discipline for whom they are
produced. Since The Economist magazine is produced for a wide audience of
professional and non-professional readers in the economic and financial sphere
who are both initiates and non-initiates with mathematical presentation
techniques, mathematical modality in The Economist could be interpreted by
making reference to the continuum of mathematical modality as suggested in
Figure 5.11 in Chapter Five.

The line graphs, according to this modality continuum, could be identified as

being of median modality (placed approximately between type 2 and 3) because
they combine elements of both these descriptions. For example, there is no colour
used within the visual frame, and the visual is decontextualised by the black line
bordering. On the other hand the mathematical relationships shown are the
relatively straight-forward derivative (rate of change) relation between only two
variables (sometimes three or four could be portrayed in the professional

265
literature), and both line graphs within the visual frame are dominated by the
interpretative headline which does not focus on the topic of the graphs, but
interprets them for the viewers (‘slumping’ for the names; ‘mounting’ for the open
year syndicates); thus, the viewers really don’t need to look at the behaviour of the
graphs closely at all.

The line graphs are therefore hardly the kind of mathematical representations
which would be expected to appear in a professional journal on economics. They
may possibly be found in a textbook on economics, but this is not very likely
because of the accent on the visually salient interpretative headline ‘Slumping …
mounting’, and due to the simplicity of the data and relationships shown. It is a
mathematical visual which, it could be reasonably suggested, is aimed mainly at
an audience which may or may not be familiar with more complex forms of
graphic presentation, but certainly do not really want the issues discussed using
professional rhetorical and visual methods. This interpretation can be supported by
the view of The Economist magazine’s editorial staff discussed in Chapter Six,
where it was stated that one of the primary purposes in the graphic design process
was to attract and keep the subscribing reader interested; mathematical visuals
were to support the information in the article, but with the proviso that
information that is presented in too complex a form will not fulfil the primary aim.

As shown in the analysis of the line graphs visual’s ideational features, there are a
number of visual messages projected at its viewer(s), and their presentation in
mathematical form accords varying degrees of credibility to the represented
participants and hence the visual messages, as well as endows them with certain
attitudes. These messages of credibility and endowed attitudes are basically
centred around:
• graphic focus - the idea of increasing losses for Lloyd’s (decreases in names
means less capital, increasing open year syndicates means names lose money).
• “Slumping … mounting”: the visual headline which synonymises the
increasing losses.
• Active names: the sub-heading identifying the quantitative focus of the left
graph.
• Syndicates with open years: the sub-heading identifying the quantitative focus
of the right graph.
• Time: the ‘x’ axes of both graphs.
266
• Quantity (number): the ‘y’ axes of both graphs.
• Source: Lloyd’s of London: the footer identifying the source of the data.

What the line graphs visual is essentially doing is presenting, in largely neutral
terms, information about the behaviour of two important aspects of the Lloyd’s of
London insurance market. As an instance of a mathematical visual derived from
the scientific/technological coding orientation, its main focus is to present
information in a mathematical format to inform the viewer(s) about the interaction
between its represented participants and therefore the visual messages. As already
mentioned, as a believable instance of this visual, it has basically median modality
- it is a credible instance of a mathematical line graph, although not a professional
or academically appropriate instance. Besides the neutral identification and
presentation of mathematical data, there are however three instances from these
VMEs where the visual does address the viewer specifically with a visual message
which in a sense forces the viewer to adopt a specific attitude or disposition
beyond the acceptance of its straightforward report of quantitative information.
These are the use of the interpretative headline “Slumping …… mounting”, the
visual footer “Source: Lloyd’s of London”, and the section of the right-hand line
graph which presents data on the increase in open year syndicates from early 1990
to 1991. A closer examination of each of these represented participants reveals
some interesting attitudes or dispositions, which are summarised below.

Slumping …… mounting
The use of this interpretative headline is quite clearly an attempt to convey, in as
direct a manner as possible, the topic focus of the graphic behaviour, which is
“increasing losses” (of money), and the direct causal relationship of the names and
open year syndicates. What is interesting in interpersonal terms here however is
that the choice of these words in the verbal headers to describe these quantitative
movements gives the viewers a much stronger message. They are decidedly
emotive words which ask the viewers to adopt a certain feeling or disposition to
the increasing losses implied in the line graphs. A check of a thesaurus supports
this point (Random House Webster’s 1992). The word ‘slumping’ is synonymised
by making reference to a dying man, personal problems, being tired, and an
economic slowdown, while ‘mounting’ is explicated by reference to climbing a

267
ladder, a death toll, riding a camel, and guns on warships. The former is almost
totally concerned with negative emotions, while the latter is a little less, though it
could be, especially when preceded by such a word as ‘slumping’ in a headline.
This interpretation can be further confirmed when the question is asked why the
graph constructors did not choose more neutral, less emotive words like
‘Decreasing… increasing’, or ‘Falling … rising’ for the line graph headings. The
viewer is thus being directly addressed with an offer of information which is
loaded with an attitude suggesting a sense of foreboding, or deepening concern
about the future.

Source: Lloyd’s of London

The function of this verbal footer is clear in that it is telling the viewer(s) that the
information is derived from the Lloyds’ of London data bank. It is asking the
viewer(s) to adopt an attitude of acceptance of the data as being authentic and
accurate. It is giving the information contained within the visual frame (not the
individual line graphs themselves) an ‘official’ seal of approval, so it is assigning
a higher modality to this content — the viewer(s) should believe what they are
told. The viewer(s) are informed that the information portrayed is within the
sphere of the economic or financial world (in ideational terms), and in
interpersonal terms are also manipulated by the footer giving the data an official
imprimatur which not only indicates the source of the data, but assures its
accuracy.

Line graph data on open year syndicates from early 1990 to 1991
The occurrence of this part of the open year syndicates line graph is interesting in
that its content relates directly to the above two attitudes. In brief, this graph is
showing data which has not been calculated at the time of the publication of this
issue of The Economist. This point is directly concerned with the already
discussed practice of the Lloyd’s Corporation closing the accounts on syndicates
three years after they officially close to allow outstanding claims to be settled —
since this text was published in March, 1993, three years back is March 1990, yet
this line graph shows data up until early 1991. Also, the sentences 11 to 15 in the
verbal aspect of the text refer specifically and discuss the fact that the 1990 figures

268
will be announced in June 1993. In the light of the official imprimatur given by
the ‘Source: Lloyd’s of London’ footer, and the use of an emotive, interpretative
headline, the reader is being asked to accept the data portrayed as being accurate,
believable, when in fact a portion of it is a projection of what can be expected.
The viewer is asked to agree with this encoded view of the future.

Thus, the line graphs are making visual statements to the viewers by simply
reporting the data, and the viewers are asked to adopt an attitude of acceptance of
them as believable instances of mathematical reportage. However, the viewers are
also asked, through embedded verbal means, to adopt the attitudes inherent in the
headlines, that of concern about, and a negative impression of the future. Here the
headline in a sense ‘anchors’ the attitudinal meaning of the mathematical modes,
and could be seen as an instance of what Barthes (1977 ) was referring to in his
discussion of ‘anchorage’ in image-text relations, where the ‘terror of uncertain
signs’ are somehow fixed by the verbal.

7.3.6 The Mountains Text: An Interpersonal Intersemiotic Complementarity

Analysis
As discussed in the previous sections, the Mountains text visuals address their
viewers by making visual statements. They also indicate and project to them a
range of attitudes about what is represented. This section will be concerned with
ascertaining the ways that the verbal mode intersemiotically complements the
visual forms of address and visual attitudes presented in an attempt to show the
ways that interpersonal intersemiotic complementarity can be realised.

7.3.6.1 The Analysis of Intersemiotic Address in the Mountains Text

In Chapter Five the explanation of MOOD in the SFL grammar model pointed out
that in language the role in the exchange and the nature of the commodity
exchanged can be combined into the four primary speech functions of Offer,
Command, Statement and Question, which can be matched by a set of appropriate
responses: accepting the offer, carrying out the command, acknowledging a
statement, and answering a question (Halliday 1994:68-69). It also discussed how

269
when information is exchanged in an interaction between a speaker/listener and a
writer/reader, as it commonly is in The Economist magazine, it is the MOOD
element in the clause which carries the components of information that are
bandied back and forth in the exchange and which realise the speech function. It
was also shown how both the speech role and whatever is to be exchanged are
expressed in lexicogrammatical terms through the choices made in the MOOD
structure of the clause, in terms of the ordering of the Subject, Finite, Predicators,
and various aspects of the Residue.

In the discussion of Visual Address in The Economist magazine above, it was seen
that the main form of address is the exchange of information between the
magazine and its viewer/readers. This exchange of information can be done by
visually asking questions or by visually making statements, and it was found that
the only form of visual address was the offer of statements of information. In
language, the order of the elements for both these forms of address is significant,
in that the order Subject before Finite realises the ‘declarative’ (statement), and
the order Finite before Subject realises the ‘interrogative’ (question) in the
exchange of information. It is proposed in this section that an examination of the
verbal aspect of the Mountains text in terms of this ordering may confirm the
proposition that interpersonal meanings in both modes in the Mountains text are
related through intersemiotic reinforcement of address.

An analysis of the Subject/Finite relationship in the clauses in the Mountains text

is presented in the following Tables 7.6 (a) and (b). These tables show that all of
the clauses are declarative in MOOD, and that they are realising the unmarked
speech

270
 Sketch Caricature - Line Graphs - Visual
Visual Offer of Offer of Information
Information (statement) (statement)
Cl’s Subject/Finite/Predicator Positioning MOOD ACT Intersemiotic Intersemiotic
Complementarity Complementarity
1 Mountains still to climb none Stat. Rein. of Add. Rein. of Add.
2 April's business plan for Lloyd's / does / not Dec. Stat. Rein. of Add. Rein. of Add.
satisfy
3 the market/ could / die Dec. Stat. Rein. of Add. Rein. of Add.
4 ONE peak scaled / often / reveals Dec. Stat. Rein. of Add. Rein. of Add.
5 it / is Dec. Stat. Rein. of Add. Rein. of Add.
6 Last year / was Dec. Stat. Rein. of Add. Rein. of Add.
7 the market / reported Dec. Stat. Rein. of Add. Rein. of Add.
8 lawsuits alleging negligence / mushroomed Dec. Stat. Rein. of Add. Rein. of Add.
9 names / were / outraged by Dec. Stat. Rein. of Add. Rein. of Add.
10 some / said Dec. Stat. Rein. of Add. Rein. of Add.
11 Losses for 1990 / would / be Dec. Stat. Rein. of Add. Rein. of Add.
12 the insurance cycle / turned Dec. Stat. Rein. of Add. Rein. of Add.
13 later years / would / bring Dec. Stat. Rein. of Add. Rein. of Add.
14 the new management team of David Rowland as Dec. Stat. Rein. of Add. Rein. of Add.
chairman and Peter Middleton as chief executive /
would / improve
15 (elipsed Subject) / placate Dec. Stat. Rein. of Add. Rein. of Add.
16 optimism / now / seems Dec. Stat. Rein. of Add. Rein. of Add.
17 It / has / become Dec. Stat. Rein. of Add. Rein. of Add.
18 the market's loss for 1990, (elipsed Subject) / will / Dec. Stat. Rein. of Add. Rein. of Add.
be
19 (elipsed Subject) / probably / close to Dec. Stat. Rein. of Add. Rein. of Add.
20 a chunk of this - possibly £500m-l billion - / is Dec. Stat. Rein. of Add. Rein. of Add.
21 it / represents Dec. Stat. Rein. of Add. Rein. of Add.
22 Both of these / involve Dec. Stat. Rein. of Add. Rein. of Add.
23 (elipsed Subject) / not Dec. Stat. Rein. of Add. Rein. of Add.
24 the money / still has / to be Dec. Stat. Rein. of Add. Rein. of Add.
25 1989's losses / were / concentrated Dec. Stat. Rein. of Add. Rein. of Add.
26 the 1990 losses / will / be Dec. Stat. Rein. of Add. Rein. of Add.
27 few names / will / avoid Dec. Stat. Rein. of Add. Rein. of Add.
28 The cumulative effect of huge losses / is / Dec. Stat. Rein. of Add. Rein. of Add.
undermining
29 names / resign Dec. Stat. Rein. of Add. Rein. of Add.
30 (elipsed Subject) / go bust Dec. Stat. Rein. of Add. Rein. of Add.
31 Lloyd's capacity / has / shrunk Dec. Stat. Rein. of Add. Rein. of Add.
32 The market now / has Dec. Stat. Rein. of Add. Rein. of Add.
33 Names[[who struggled to pay1989's losses last Dec. Stat. Rein. of Add. Rein. of Add.
year]] / will / find
34 most / are / trapped Dec. Stat. Rein. of Add. Rein. of Add.
35 84% of names / have Dec. Stat. Rein. of Add. Rein. of Add.
36 that figure / will / be Dec. Stat. Rein. of Add. Rein. of Add.
37 Members with open years / can / stop Dec. Stat. Rein. of Add. Rein. of Add.
38 they / cannot / leave Dec. Stat. Rein. of Add. Rein. of Add.
39 It / is Dec. Stat. Rein. of Add. Rein. of Add.
40 (elipsed Subject) which / will / be unveiled Dec. Stat. Rein. of Add. Rein. of Add.
41 The plan, / [[which / will / set …]] is / now Dec. Stat. Rein. of Add. Rein. of Add.
being discussed
42 its proposals / could / be Dec. Stat. Rein. of Add. Rein. of Add.
43 the reforms / suggested Dec. Stat. Rein. of Add. Rein. of Add.
44 Cost-cutting, for instance, / is / being Dec. Stat. Rein. of Add. Rein. of Add.
45 the huge losses revealed since then / have / Dec. Stat. Rein. of Add. Rein. of Add.
softened
46 Lloyd's / is / sacking Dec. Stat. Rein. of Add. Rein. of Add.
47 agencies / are / following Dec. Stat. Rein. of Add. Rein. of Add.
48 Mr Middleton / thinks Dec. Stat. Rein. of Add. Rein. of Add.
49 the members' agencies [[that look after names]] and Dec. Stat. Rein. of Add. Rein. of Add.
the managing agencies [[that run syndicates]] / are
50 The centre / may / exert Dec. Stat. Rein. of Add. Rein. of Add.
51 the agencies / all / trade Dec. Stat. Rein. of Add. Rein. of Add.
52 He / wants / to raise Dec. Stat. Rein. of Add. Rein. of Add.
Table 7.6 (a) Addressing the reader/viewer - Interpersonal Intersemiotic MOOD.

271
 Sketch Caricature - Line Graphs - Visual
Visual Offer of Offer of Information
Information (statement) (statement)
Cl’s Subject/Finite/Predicator Positioning MOOD ACT Intersemiotic Intersemiotic
Complementarity Complementarity
53 (elipsed Subject) / to / make Dec. Stat. Rein. of Add. Rein. of Add.
54 He / is Dec. Stat. Rein. of Add. Rein. of Add.
55 this / will / be welcomed Dec. Stat. Rein. of Add. Rein. of Add.
56 it / leaves Dec. Stat. Rein. of Add. Rein. of Add.
57 The main reason [[ that these are so tricky ]] / is Dec. Stat. Rein. of Add. Rein. of Add.
58 Mr. Rowland / likes / to say Dec. Stat. Rein. of Add. Rein. of Add.
59 The business plan / will / certainly discuss Dec. Stat. Rein. of Add. Rein. of Add.
60 it / has / to Dec. Stat. Rein. of Add. Rein. of Add.
61 finding an exit route / has / become Dec. Stat. Rein. of Add. Rein. of Add.
62 The plan / may / suggest Dec. Stat. Rein. of Add. Rein. of Add.
63 Centrewrite / will / need Dec. Stat. Rein. of Add. Rein. of Add.
64 names [[who are relatively free of these burdens]] / Dec. Stat. Rein. of Add. Rein. of Add.
will / protest
65 Those with an eye to the future / say Dec. Stat. Rein. of Add. Rein. of Add.
66 nobody, individual or incorporated / will / join Dec. Stat. Rein. of Add. Rein. of Add.
67 Some / have / suggested leaving Dec. Stat. Rein. of Add. Rein. of Add.
68 (elipsed Subject) / ensuring Dec. Stat. Rein. of Add. Rein. of Add.
69 capital-providers / can / find Dec. Stat. Rein. of Add. Rein. of Add.
70 such a move / could / deter Dec. Stat. Rein. of Add. Rein. of Add.
71 Mr Middleton / sees Dec. Stat. Rein. of Add. Rein. of Add.
72 that / may / not do Dec. Stat. Rein. of Add. Rein. of Add.
73 insurers / are / fighting Dec. Stat. Rein. of Add. Rein. of Add.
74 some recent decisions / have / been helpful Dec. Stat. Rein. of Add. Rein. of Add.
75 The business plan / will / have less to say Dec. Stat. Rein. of Add. Rein. of Add.
76 Members' and managing agencies / have / been Dec. Stat. Rein. of Add. Rein. of Add.
deluged
77 they / are Dec. Stat. Rein. of Add. Rein. of Add.
78 Mr Middleton / points / out Dec. Stat. Rein. of Add. Rein. of Add.
79 prospective litigants / could / settle Dec. Stat. Rein. of Add. Rein. of Add.
80 it / would / minimise Dec. Stat. Rein. of Add. Rein. of Add.
81 the insurers / are / not keen to settle Dec. Stat. Rein. of Add. Rein. of Add.
82 Not all the claims / are Dec. Stat. Rein. of Add. Rein. of Add.
83 many E&O insurers / have Dec. Stat. Rein. of Add. Rein. of Add.
84 the E&O pot of about £1 billion / is Dec. Stat. Rein. of Add. Rein. of Add.
85 Letting the litigation run / is Dec. Stat. Rein. of Add. Rein. of Add.
86 It / means Dec. Stat. Rein. of Add. Rein. of Add.
87 (elipsed Subject) which / might / put off Dec. Stat. Rein. of Add. Rein. of Add.
88 It / could / drive Dec. Stat. Rein. of Add. Rein. of Add.
89 Some names / reckon Dec. Stat. Rein. of Add. Rein. of Add.
90 Lloyd's itself / could / be held Dec. Stat. Rein. of Add. Rein. of Add.
91 it / were / shown Dec. Stat. Rein. of Add. Rein. of Add.
92 they / are / right Dec. Stat. Rein. of Add. Rein. of Add.
93 (elipsed Subject) that / could / close down Dec. Stat. Rein. of Add. Rein. of Add.
94 the angry names / argue Dec. Stat. Rein. of Add. Rein. of Add.
95 A bond issue / is Dec. Stat. Rein. of Add. Rein. of Add.
96 Rising insurance rates / should / make Dec. Stat. Rein. of Add. Rein. of Add.
97 names / cannot / touch Dec. Stat. Rein. of Add. Rein. of Add.
98 they / must / meet Dec. Stat. Rein. of Add. Rein. of Add.
99 Mr Middleton / would / like to help Dec. Stat. Rein. of Add. Rein. of Add.
100 any borrowing to pay for past losses / could / run Dec. Stat. Rein. of Add. Rein. of Add.
101 some / see Dec. Stat. Rein. of Add. Rein. of Add.
102 Mortgaging future profits / would / make Dec. Stat. Rein. of Add. Rein. of Add.
103 profits / may / not, in the event, materialise; Dec. Stat. Rein. of Add. Rein. of Add.
104 few people / expected Dec. Stat. Rein. of Add. Rein. of Add.
105 Names, policy-holders and prospective investors / Dec. Stat. Rein. of Add. Rein. of Add.
are / pinning
106 The two men / win Dec. Stat. Rein. of Add. Rein. of Add.
107 good intentions / are / not Dec. Stat. Rein. of Add. Rein. of Add.
108 the plan / cannot / solve Dec. Stat. Rein. of Add. Rein. of Add.
109 (elipsed Subject/Finite) / help / head off Dec. Stat. Rein. of Add. Rein. of Add.
110 (elipsed Subject/Finite) / find Dec. Stat. Rein. of Add. Rein. of Add.
111 Lloyd's / may / not Dec. Stat. Rein. of Add. Rein. of Add.
112 The stakes / are Dec. Stat. Rein. of Add. Rein. of Add.
Table 7.6 (b) Addressing the reader/viewer - Interpersonal Intersemiotic MOOD.

272
function of making a statement. There are no clausal examples of interrogative or
imperative MOOD, and thus no evidence of questions asked (even rhetorically) or
requests made, and no orders given to the readers. Given that both the visuals
address the viewers in this way, we therefore have intersemiotic complementarity
realised in terms of the ways that both the visual and verbal modes address the
viewer/readers. Both the modes make statements which the viewer/readers can
agree or disagree with, acknowledge or contradict, and both intersemiotically
complement each other via the relation referred to as reinforcement of address.

This interpretation of the Mountains text as being a multimodal or composite offer

of information consisting of visual and verbal statements is supported when you
look at the level of involvement required by the viewer-reader. In the verbal
aspect of the text there are no attempts to directly involve the reader as the other
interactant in this exchange of information. An examination of the text reveals for
example no forms of direct address, such as reference to the reader as ‘the reader’,
or the use of the second person pronoun ‘you’. Further, there are no instances of
the first person plural pronouns of ‘we/us/ours’ used in the inclusive sense to draw
the reader into the discussion, as in references such as: ‘As we saw in ….’, or even,
‘As we can see in the chart ….’ which could draw the reader into a multimodal
exchange of information. All singular and plural personal pronouns are in the third
person, as in the use of ‘he/it/its’ and ‘they’, and they are used throughout the text
in referring to the situation, institutions and people discussed.

The visuals intersemiotically complement this verbal level of involvement (and

vice versa). As was pointed out in the previous sections, the sketch caricature in
the Mountains text is a naturalistic visual which, in terms of the horizontal angle,
has an extreme oblique point of view. The frontal plane of the viewer of this visual
is perpendicular to the frontal plane of the most visually salient represented
participants, the two men and the boulder. There is thus a strong absence of
involvement with this scene on the part of the viewer, not by choice (although that
may be the case), but by the lack of involvement coded (or loaded) into the visual
by its constructor(s). It is a scene which shows action happening, and because of
the obliqueness of the angle puts the viewers in the place of those who are, as it

273
were, watching a scene from a movie —the viewer is required to do nothing
except to observe the scene and accept or reject the action represented. The sketch
caricature is a visual offer of information which can be agreed or disagreed with,
acknowledged, or contradicted.

In addition, the line graphs in the Mountains text are also viewed from an oblique
point of view. All viewers are forced to view the information portrayed in an
abstract graphic form as an offer of information with which they can agree or
disagree, acknowledge or contradict. Unlike the naturalistic visuals, into which
their constructors can encode various degrees of involvement through degrees of
obliqueness, the mathematical visuals are because of their mathematical coding
orientation restricted to a two-dimensional plane which leaves no other choice for
the viewers than to be an uninvolved observer of a presentation of information or
data.

Therefore, in the Mountains text we have both the verbal and visual modes
addressing the viewer/reader in the same way - they both make offers of
information by making statements which can be can be agreed or disagreed with,
acknowledged, or contradicted. The reader/viewers are not explicitly referred to,
and are assumed in many ways to be simply the receivers of these multimodal
statements. Therefore, in terms of address, and involvement there is evidence to
confirm the proposition that the intersemiotic complementarity between the visual
and verbal modes in the Mountains text is realised by intersemiotic reinforcement
of address.

7.3.6.2 The Analysis of Intersemiotic Attitudinal Congruence and Attitudinal

Dissonance in the Mountains Text (Visual-Verbal Modality)
In language the truth or credibility of what is represented by a speaker or writer is
expressed through the use of modality at the clause level, and the polarity between
the affirmation and denial of this offered information is expressed in terms of
whether something is or is not, or whether it is real or unreal (via the Finite
element expressing polarity). In between these two extremes of certainty there are
other possibilities which express degrees of certainty or uncertainty, where

274
perhaps something could happen, or of usuality, where something might
sometimes happen, but not always (Halliday 1994:88-92)

As pointed out in the analysis of the visuals in the Mountains text, the line graphs,
while not of ‘technical’ quality in terms of professional publications in the
financial realm, are an accurate representation and report of what has actually
happened, and as such would be considered as a collection of definite visual
statements of what is or what is not in quantitative terms. An examination of the
verbal aspect of the Mountains text also reveals a number of definite statements
about aspects of the information that the line graphs present. These are presented
in Table 7.7.

S’s Firm Statement Visual Message Element

6 Last year was awful: the market reported losses of over £2 billion Graphic Focus - Increasing
($3.3 billion) for 1989, … Losses
12 … a chunk of this [already-mentioned loss] … is a form of double- Graphic Focus - Increasing
counting, Losses
15 … whereas 1989's losses were concentrated on the 5,000 names Graphic Focus - Increasing
Losses
16 The cumulative effect of huge losses is undermining the market. Graphic Focus - Increasing
Losses
17 As names resign or go bust, Lloyd's capacity has shrunk. Graphic Focus - Increasing
Losses
20 And most [already-mentioned names] are trapped in "open years" … Graphic Focus - Increasing
Losses
23 It [already-mentioned losses etc.] is an inauspicious background Graphic Focus - Increasing
Losses
Table 7.7 Factual statements in the Mountains text

What Table 7.7 shows is that interpersonal intersemiotic complementarity is

realised by the attitudinal congruence relation between the definite visual
statements of what has happened (in terms of names and open year syndicates),
and the specific references to these events in the verbal aspect of the text.. In the
verbal mode statements regarding actual events are made initially about the recent
past, covering the major problems that have arisen with Lloyd’s losses (sentence
6). This is followed by statements about the nature of the increasing losses
incurred (sentences 12, 15, 16, 20). All of these statements are mirrored visually
by the two line graphs’ general graphic message of increasing losses, and the
effects of the quantitatively represented behaviour over time of decreasing
numbers of names and increasing numbers of open-year syndicates.

275
It follows from this that if the line graphs as definite visual statements are related
intersemiotically through the attitudinal congruence relation to the verbal aspect of
the text, then it is possible that this may also be the case with the sketch caricature
— it is possible that this visual mode and the verbal mode project the same or
similar attitudes in terms of the ways that they attempt to temper their statements.
It was pointed out earlier that the sketch caricature is basically addressing the
viewers in terms of visual statements, but that these were a ‘suggested’ or ‘loaded’
interpretation of reality, and as such carry a lower modality in terms of a
representation of ‘truth’ than photographs or realistic sketch drawings. In this
visual what is presented is not a true representation of reality, but a whimsical
version of it which projects various attitudes in terms of its believability and
possibility. It is not stating what is or is not, or true or false as the line graphs
generally are, but is projecting to the viewers a range of attitudes about the
situation at Lloyd’s which can be placed somewhere in between.

As discussed in Chapter Five, Halliday (1994:88-92) distinguishes three degrees

of modality in the SFL model which relate to three degrees of confidence, or lack
of confidence about what a speaker feels towards the truth or otherwise of a
proposition. These can be summarised as:
• Low modality is concerned with the possibility that something may be true, and
may be realised by modal auxiliaries may, might, and could, as well as by
possible (Adjective), possibly, perhaps (Adverb) and possibility (Noun).

• Median modality is concerned with the probability that something may be true,
based on a reasonable inference about what could be reasonably expected to
happen, with the implicit assumption that it is possible for the event to not
occur. This may be realised by modal auxiliaries should, and ought, as well as
by probable and likely (Adjective), probably and presumably (Adverb), and
probability and likelihood (Noun).

• High modality is concerned with the degree of certainty that something may be
true; it is not a categorical assertion, but an expression of diminished certainty.
This may be realised by modal auxiliaries will and must, as well as by certain
(Adjective), certainly and surely (Adverb), and certainty (Noun).

An analysis of the verbal aspect of the Mountains text for the usage of these kinds

276
 Rowland & Middleton
23 It is an inauspicious background for Lloyd's first-ever business plan, which will be unveiled by Messrs
Rowland and Middleton in late April.
Boulder (Lloyd’s problems)
11 ... the market's loss for 1990, to be announced in June, will be bigger than that for 1989 ...
15 ... the 1990 losses will be more widely spread; few names will avoid them.
19 Names who struggled to pay 1989's losses last year will find it even harder to meet 1990's.
21 Already 84% of names have at least one open year; by June that figure will be close to 100%.
39 And names who are relatively free of these burdens will protest against any tacit mutualisation of losses.
40 Those with an eye to the future say that Lloyd's must insulate newcomers from the cost of the past -
otherwise nobody, individual or incorporated, will join.
58 But names cannot touch those profits ..... ; meanwhile, they must meet losses for 1990 (and perhaps 1991).
The Mountainside (future)
10 Such optimism now seems premature.
11 ... the market's loss for 1990, ... will be bigger than that for 1989, probably close to £3 billion.
12 Admittedly, a chunk of this [loss] - possibly £500m-l billion - is a form of double-counting ...
42 By showing that capital-providers can find the exit blocked, however, such a move could deter new names
from signing up.
44 But that [concentrating old-year claims and reserves in a single body] may not do much to close years that
are open .........
47 Mr Middleton points out that, if prospective litigants could settle with the E&O insurers, … it would
minimise the sums siphoned off by lawyers.
51 It means at least three years of bad publicity for Lloyd's, which might put off not only new names but also
new policyholders.
52 It [litigation] could drive many agencies into bankruptcy.
57 Rising insurance rates should make the early 1990s profitable.
60 But any borrowing to pay for past losses could run foul of the trade department's solvency rules ...
62 And profits may not, in the event, materialise;
66 If the plan cannot solve the open-year problem, ...... Lloyd's may not survive.

Climbing - pushing up (the solution)

23 It is an inauspicious background for Lloyd's first-ever business plan, which will be unveiled by Messrs
Rowland and Middleton in late April.
24 The plan, which will set Lloyd's future course, is now being discussed by its market board.
32 Most of this [solutions] will be welcomed by names.
45 The business plan will have less to say about the lawsuits lodged by Lloyd's own names.
36 The business plan will certainly discuss open years;
37 The plan may suggest a bigger job for Centrewrite, the Lloyd's-owned insurance company ...
25 According to Mr Middleton, its proposals could be more radical than the reforms suggested in last year's
taskforce report ....
56 A bond issue is also talked of as a possible solution for the third problem, financing 1990's losses - which
the business plan may not even discuss.
Lloyd’s
38 But Centrewrite will need a lot more capital from the market
40 Those with an eye to the future say that Lloyd's must insulate newcomers from the cost of the past …
29 The centre may exert some control ....
3 If April's business plan for Lloyd's does not satisfy both its capital providers and its clients, the market
could die.
51 It means at least three years of bad publicity for Lloyd's, which might put off not only new names but also
new policyholders.
53 Some names reckon that, .... Lloyd's itself could be held liable if it were shown to have acted in bad faith.
54 If they are right, that could close down the whole market.

Table 7.8 (a) Modality features of the sketch caricature

277
of modality features according to the VMEs in the sketch caricature and the line
graphs is presented in Tables 7.8 (a) and (b).

Active names
19 Names who struggled to pay 1989's losses last year will find it even harder to meet 1990's.
21 Already 84% of names have at least one open year; by June that figure will be close to 100%.
32 Most of this [solutions] will be welcomed by names.
39 And names who are relatively free of these burdens will protest against any tacit mutualisation of losses.
40 Those with an eye to the future say that Lloyd's must insulate newcomers from the cost of the past -
otherwise nobody, individual or incorporated, will join.
58 But names cannot touch those profits ..... ; meanwhile, they must meet losses for 1990 (and perhaps 1991).

Slumping … … mounting
11 ... the market's loss for 1990, to be announced in June, will be bigger than that for 1989 ...
21 Already 84% of names have at least one open year; by June that figure will be close to 100%.

Graphic focus - increasing losses

15 ... the 1990 losses will be more widely spread; few names will avoid them.
19 Names who struggled to pay 1989's losses last year will find it even harder to meet 1990's.
11 ... the market's loss for 1990, ... will be bigger than that for 1989, probably close to £3 billion.
12 Admittedly, a chunk of this [loss] - possibly £500m-l billion - is a form of double-counting ...

Table 7.8 (b) Modality features of the line graphs

As pointed out in Section 7.1 on visual modality in the Mountains text, both the
sketch caricature and line graphs carry visual modality features which work to
mitigate the focus of their attitudes — the sketch caricature was interpreted as
being an instance of suggested reality by virtue of its caricaturisation, and the line
graphs were interpreted as being definite quantitative statements about past
happenings, but of median modality in the mathematical continuum because of the
degree of detail shown and the use of the interpretive headlines ‘Slumping …
mounting’. What these analyses show is that the treatment of the represented
VMEs in both the sketch caricature and the line graphs involves varying levels of
modality, producing intersemiotic complementarity in terms of attitudinal
congruence.

As Tables 7.8 (a) and (b) show, there are some general patterns which suggest
varying degrees of attitudinal congruence between the two modes in the
Mountains text. What is immediately apparent is that the VMEs representing
Lloyd’s problems (the Boulder), the business plan (the action of Climbing-pushing
up), and the two financiers (Rowland & Middleton) in the sketch caricature, and
those representing the names (Active Names), and the idea of increasing losses
(Graphic Focus and Slumping … mounting) in the line graphs, make significant
278
usage of high modality (the use of will, must, certain, certainly, surely, and
certainty) to project to the readers high degrees of certainty with their
propositions. This accords with the sketch presentation which has as its premise
the shared assumptions by the reader/viewers and the drawer that Lloyd’s has
problems to deal with, and the graphical presentation of factual information to
show how the numbers of names are increasing and that this also means increasing
losses. The situation and problems portrayed are real ones, and so are accorded a
higher modality.

On the other hand, the treatment of the VMEs concerned with discussions about
what will happen in the future (The Mountain-side) shows a very significant usage
of low modality (the use of may, might, could, possible, possibly, perhaps and
possibility) to project to the readers the possibility that something may be true, as
opposed to the relative certainty attached to the shared-understanding about
Lloyd’s problems. This is synonymised by the sketch presentation of the upper
mountain-side, which is a visual metaphor for Lloyd’s future. What can be seen
here is that the drawers obviously consider Lloyd’s future to be bleak, with further
‘Mountains to climb’ — however the fact that the scene that has been drawn is
caricatured gives it the sense that it is a suggested future, and this is mirrored by
the low modality accorded this aspect of the text’s treatment of the same issue, as
the penultimate sentence (66) makes clear in ‘Lloyd’s may not survive’.

Thus, in terms of the degrees of attitudinal congruence realising intersemiotic

complementarity between the two modes in the Mountains text, we find that there
are three general areas. The first is the attitudinal congruence between the definite
statements made by the line graphs and the discussion of them in the verbal aspect
of the text. The second is the attitudinal congruence between most of the VMEs in
the sketch caricature and the line graphs and the high modality displayed in their
treatment, and the third is the attitudinal congruence between the VME dealing
with Lloyd’s future in the sketch caricature and the relatively lower modality in

279
Sentence Rowland & Middleton Boulder Climbing up - The Mountainside Slumping …
(Lloyd’s problems) pushing up (future not mounting
(difficulty) optimistic) (concern about future)
1-7
8 (losses) smaller
9 (Rowland as) chairman
(Middleton as) chief executive
10
11 (loss) bigger
12
13
14
15
16 huge (losses)
17
18
19 harder
20 (losses) too big unpredictable
21
22
23 Messrs (Rowland and inauspicious inauspicious
Middleton) (background) (background)
24
25 Mr. (Middleton) huge (losses)
26
27
28 Mr. (Middleton)
29
30
31
32
33 big (problems)
34 huge (sums)
35 Mr. (Rowland)
36
37 bigger (job)
38
39
40
41
42
43 Mr. (Middleton)
44
45
46
47 Mr. (Middleton)
48
49 (E&O pot) too
small
50
51
52
53
54
55
56
57
58
59 Mr. (Middleton)
60
61
62
63 Rowland/Middleton
64 The two men
65-67

Table 7.9 Attitudinal intersemiotic complementarity in the sketch caricature

280
evidence in the verbal aspect of the text. A fourth aspect could also be added here,
and that is the lack of attitudinal dissonance, where opposite or ironical attitudes
are displayed between the two modes. Based on this analysis it is clear that both
modes are placed on the page to work in concert, to offer information via both
verbal and visual statements, and to ensure that the ways that it is received and
dealt with by the viewers and readers are synonymous. This of course accords
with the stated aims of the graphic design team at The Economist magazine, where
it was stated in Chapter Six that images are generally used to attract the reader to
the written article, and mathematical visuals are to support the propositions in it.
Of course the use of irony across modes is one of the more popular ways used by
graphic designers to attract readers to articles in some magazines, but in The
Economist magazine, at least in terms of its treatment of a financial issue like
Lloyd’s, propositional and attitudinal congruence is used consistently.

7.3.6.3 The Analysis of Intersemiotic Attitudinal Congruence and Attitudinal

Dissonance in the Mountains Text (Attitudinal Attributes and Adjectives).
An examination for further evidence of attitudinal intersemiotic complementarity
between the visual and verbal modes in the Mountains text reveals a number of
features which lend further support to these conclusions of attitudinal congruity.
The two most significant instances of intersemiotic complementarity in these
terms are concerned with the visually represented men, represented as the VMEs
“Rowland & Middleton (the two financiers)”, and “the Boulder (Lloyd’s
problems)”. The intersemiotic attitudinal features relevant to these represented
participants are realised by the use of forms of address and attitudinal adjectives in
the verbal aspect of the text which cohere in a synonymous way with the visual
attitudinal message elements. An inventory of the intersemiotic attitudinal
adjective features revealed by an analysis of the Mountains text are summarised in
Table 7.9.

The first and most obvious feature here is the nature of the references or forms of
address used for Middleton and Rowland in the verbal aspect of the text, and how
this is complemented by their represented visual attributes As noted previously,
caricaturisation is often used to portray represented participants in a comical or

281
whimsical way, and that both these men are portrayed only in such a way as to
facilitate recognition by the viewers. As such they are not the main focus of the
message of the sketch caricature. They are treated with perhaps a modicum of
respect — they are portrayed in pin-striped suits which accord them some kind of
‘official’ status, and even though they are portrayed as having a difficult
immediate and future task, they are not portrayed as incompetent or as unable to
cope. This portrayal of their attributes is mirrored in the forms of address used for
identifying them in the verbal aspect of the text. Both men are introduced initially
in terms of their positions in Lloyd’s through ‘David Rowland as chairman and
Peter Middleton as chief executive’, both of which are positions of power and
responsibility (sentence 9). They are then referred to throughout the rest of the text
by the respectful titles of ‘Messrs’ and ‘Mr’ (sentences 23, 25, 28, 35, 43, 47, and
59), thus according them some respectful hedging on the part of the writers
(perhaps this may change if the business plan they have produced fails). What is
interesting in this respect also is that in the very last section of the article, where
the writers are commenting on Lloyd’s prospects and drawing conclusions from
the preceding analyses, the respectful titles are dropped, Middleton and Rowland
are mentioned by their last names only (sentence 63), and are finally and simply
referred to as ‘The two men’ (sentence 64).

The other significant instance of intersemiotic complementarity in terms of

attitudinal features are the references made to the magnitude of Lloyd’s problems,
as represented by the visually salient boulder, and by the use of adjectives
referring to Lloyds’ problems in general, the size of its problems with increasing
losses, and the need for raising enough sums of money. The losses are initially
referred to in straightforward comparative terms as being ‘smaller’ and ‘bigger’
than what they are compared to - this occurs in the first eleven sentences of the
text, the section which is basically concerned with orienting the reader by
reporting the background situation (the discourse structure of the Mountains text
is presented in the next section). However, once the background is reported, and
the analysis of Lloyd’s problems starts, the attitudes to the losses change in nature
to the use of much stronger adjectives such as losses which are ‘huge’ (sentence
16), and ‘too big’ (sentence 20). The problems in general that Lloyds is having are

282
simply big (sentence 33), and the sums of money needed to help are ‘huge’
(sentence 34), or the sources of funds to help are ‘too small’ (sentence 49).

In terms of the attitudes projected by the line graphs and in relationship to the
verbal aspect of the text there is also evidence for intersemiotic complementarity
in attitudinal terms between the modes. This can be seen for example in the
general graphic focus which conveys the sense that Lloyds is experiencing
increasing losses. Both graphs portray this loss in numerical form, and the
headings provide an interpretation in terms of the fact that these losses are
continuing and increasing. As mentioned already in relation to Lloyd’s problems,
the losses are initially referred to in straightforward comparative terms as being
‘smaller’ (sentence 8) and ‘bigger’ (sentence 11) than what they are compared to,
and they become losses which are huge (sentence 16), and then too big (sentence
20). So there is a visual-verbal Attitudinal Reinforcement in terms of the sense of
experiencing losses and the variation in their magnitude. This interpretation is
reinforced when one considers the fact that the graphs are drawn up in such a way
as to emphasise this sense of being huge, and then too big. As already mentioned,
the expression of the rate of change of a variable over time is precisely the
function of this type of line graph, and the degrees of change (i.e. magnitude) can
be varied by manipulating the one or both axial scales. In this case the scales of
the graphs are compressed to make the slope of the graphic lines more acute than
if they would be if the year scales were more expanded, with the result that the
year-on-year rate of change seems to be greater and the sense of the magnitude of
the changes accentuated.

The occurrence of these two significant uses of support for the attitudes expressed
in the visuals lends weight to the interpretations of the visuals in terms of their
main visual message focus — that of the problems that the management of
Lloyd’s are having and will have with the magnitude of its financial problems. In
both the modes the two men are referred to in reasonably respectful terms, and the
attitudes projected regarding the magnitude of the problems that Lloyds is having
are also attitudinally congruent. Thus we have, in interpersonal terms, further
evidence to support the central proposition of this study, that both the verbal and

283
visual modes in the Mountains text intersemiotically complement each other to
produce a coherent multimodal text through reinforcement of address and
attitudinal congruence.

7.4 An Intersemiotic Compositional Analysis of The Mountains Text

An examination of the intersemiotic compositional features of the Mountains text
involves an examination of those features of layout which work to realise
intersemiotic complementarity between the visual and verbal modes. It will be
argued that these various elements or features have not been placed on the page(s)
randomly, but for various integrative purposes, the most important of which are to
convey to the readers a sense of unity, of co-operation, and of consistency in terms
of the information presented in the total message. In the Mountains text the ways
that the sketch caricature and the mathematical line graphs interact with each other
and with the verbal aspect of the text requires making observations about these
elements of the text’s visual-verbal layout, as well as the text’s inter-visual
features. There are a number of important visual-verbal compositional aspects in
the Mountains text which show that both the modes work together to produce this
coherent and unified visual-verbal message to the readers. These are concerned
with where the visual and verbal elements are spatially organised on the page in
terms of their horizontal/vertical, left/right, or central placement, the size of the
visual and verbal elements relative to each other, and how the visuals and the type
interact with each other in taking up page space. The compositional features of the
Mountains text will be examined according to the following compositional
relations, which were outlined in Chapter Five:
• Information valuation on the page
• Salience on the page
• Degrees of framing of elements on the page
• Inter-visual synonymy
• The importance of potential reading paths

Each of these relations are concerned with some of the major compositional
principles already discussed in Chapters Four and Five, such as visual salience and
perspective, the use of balance or balancing centres and relative placement on the
page, the directionality produced by vectors, reading paths and possible attendant

284
Stage I - Analysis of the Visual(s) on the page
1) Examine the visual(s) on the page in relation to information value (vertical, horizontal axes, and the
balancing centre & margins).

2) Examine the visual(s) on the page in relation to salience (foregrounding & backgrounding. relative sizing.
contrasts in tones and differences in sharpness of focus).

3) Examine the visual(s) on the page in relation to framing (clear spaces or actual frame lines, contrasts in
tones and/or colours, and structural elements which divide visual space).

4) Examine the visual(s) on the page in relation to vectors producing an hierarchy of elements via
arrangements of abstract graphic elements, and the similarities and differences in size or volume in co-
occurring images.

§
Stage II - Analysis of the Verbal (body copy)
1) Examine the body copy on the page in relation to information value (top/bottom placement: visual ‘weight’
assigned to headings, sub-headings, blurbs, boldface lead-ins, by-lines, & captions; left/right placement:
columns, margins, line spacing, line length; and centre & margins: balance between margins & ‘live
matter’).

2) Examine the body copy on the page in relation to salience (relative sizing of the type face; contrasts in tonal
values: lightface vs. extra-bold, contrasts in type: traditional vs. decorative typefaces, roman vs. italics and
various forms of outlining, shadowing, and ornamental or swash characters; and contrasts in colours: highly
saturated vs. softer muted.).

3) Examine the body copy on the page in relation to framing (frame lines, empty space, borders; contrasts in
tonal values and/or colours; and elements which divide/control visual space: runarounds, bleeding).

§
Stage III - Analysis of intersemiotic complementarity
1) Examine the ways the modes are mapped to realise a coherent layout or composition by information
valuation on the page: left/right, top/bottom placement, and balance.

2) Examine the ways the modes are mapped to realise a coherent layout or composition by salience on the
page: foregrounding and-backgrounding, relative sizing, tonal contrasts, and variations in focus.

3) Examine the ways the modes are mapped to realise a coherent layout or composition by degree of
framing of elements on the page: clear spaces or actual frame lines, contrasts in tones and/or colours, and
visual-verbal bleeding though runarounds, margins and gutters.

4) Examine the ways the modes are mapped to realise a coherent layout or composition by Inter-visual
synonymy: the degrees of semblance in form across visual modes.

5) Examine the ways the modes are mapped to realise a coherent layout or composition by potential reading
paths.

Figure 7.5 Procedural steps in the INTERSEMIOTIC COMPOSITIONAL analysis

285
narrativisation, and the sense of division produced by visual framing, dividing
lines, runarounds and other graphological conventions. These compositional
intersemiotic relations can also be examined in terms of the Mountains text’s
discourse structure by investigating how both modes utilise given-new relations
and the ways this relates to the chronological development of the issues covered in
the article. Before dealing with these features however, they should be prefaced by
an examination of the interaction between intertextuality features and
compositional aspects of the Mountains text’s placement in the layout of the
whole magazine — this relates to the placement of the Finance department in
relation to the whole magazine, and each of the Finance articles in relation to the
Finance department itself. The actual sequence of steps taken in this analysis are
outlined in the Figure 7.5.

7.4.1 Intertextuality and Composition in the Mountains Text

As mentioned previously in the contextual description in Chapter Six, the
Mountains text has not been produced in isolation, but carries with it the
compositional influences of the preceding texts which deal with this specific as
well as other more general issues in The Economist magazine. Not only has the
Mountains text’s intertextual history left its mark in terms of ideational and
interpersonal meanings, but also in terms of various 'coded' layout conventions or
sequences that are the result of the layout and design practices that The Economist
magazine graphic design staff typically follow. This is not only in relation to the
ways they have organised the two modes in the Mountains text itself, but also in
relation to the text’s position relative to the other texts in the Finance department,
the Finance department to other departments in the magazine, and the
compositional features of the whole magazine itself.

The contextual description given in Chapter Six showed that the compositional
choices which the graphic design team make in producing an issue of The
Economist magazine are in large part also influenced by those features of the
context of culture which relate specifically to various journalistic, publishing,
economic, political, artistic and graphic design, and general western cultural
mores (Garrett 1994). This description also demonstrated that The Economist

286
magazine typically consists of a number of departments or sections, one of which
is the Finance department, and that these departments share a consistent format in
terms of typology and layout. For example, the start of these departments in every
issue is always instantiated compositionally by the use of a large bolded and
capitalised heading — the Finance department in this case starts with the heading
“FINANCE”. This same word, in smaller, bold type, is also placed on the top
corner of each subsequent page (on the left of the left-hand page, on the right of
the right-hand page) to signal that each of the pages on which it is placed is part of
the same department. The ending of this department is signalled by the large
bolded heading of the subsequent department, using the same typographical form
as that used to signal the beginning of a department. Each of the articles within the
Finance department are different in size and length, whether they include visuals
or not, but each has a visually salient bolded heading, and an ending signalled by a
column-wide dark line with a small black square. In the Mountains text, as already
seen, the text begins with the large bolded heading ‘Mountains still to climb’, and
ends with the black line and square. In the particular issue in which the Mountains
text occurs, this division of the articles by bolded heading and column-wide line is
further accentuated by the following article being framed or boxed with double-
borders, thus making a clear distinction compositionally between the leading
article and the following shorter articles.

7.4.2 Intersemiotic Compositional Features in the Mountains text

There are a number of important intersemiotic compositional aspects in the
Mountains text which demonstrate that both the modes work together to project a
coherent visual-verbal bundle of meaning to the readers. As mentioned above,
these are concerned with information valuation on the page, salience on the page,
degrees of framing of elements on the page, inter-visual synonymy, and the
importance of potential reading paths. Each of these will be discussed in the
following sections.

A consideration of Information valuation on the page is interesting in that the

vertical or top/bottom placement of the visuals on the page in the Mountains text
shows that the sketch caricature is placed in the lower status half of the page, but

287
this loss of vertical prominence is compensated by its placement in a primary
position in relation to the spine, and the fact of its visual salience in terms of size
and colour saturation (a casual reader flipping through the magazine will more
easily see the sketch caricature first and may be attracted to the article because of
the nature of the sketch or the inherent humour of the situation portrayed; this
accords with the ‘attract’ function of sketch caricatures, which is to grab the
reader’s attention).

On the other hand, the line graph visual is placed in the centre of the top-half of
the page, a primary position in terms of visual weight, and as a divider or
balancing centre between the two halves of the type. Despite the fact that one of
the functions of composition is to be engaged in a “striving for unity” (Arnheim
1988:133), the line graph seems to divide the verbal aspect of the text, to force it
to go around the visual frame. This “invasion” of the page space is accentuated by
the dividing lines of the graphic borders and the use of the runaround. However,
one of the functions of a central placement is that perceptually, for the viewer,
central placement also acts as a “stabiliser of weight”, where visual elements
“located in the central area or on a centrally located axis gained in power” and
helps the objects outside the centre zone to be “united and stabilised when they are
grouped around the balancing centre” (133). Thus, the “invasion” of the verbal
page space is stabilised by the central placement of the visual, and the sense of
compositional unity and inter-modal coherence is maintained.

The relative sizes of the visuals (their visual salience) compared to the verbal
aspect of the text in relation to the space taken up by each on the page, referred to
as Salience on the page, also illustrates how important the elements are to the
compositional makeup of the text. As mentioned in Chapter Five, White
(1982:127) has asserted that in compositional terms size is an indicator of visual
importance, so a graphic designer should “Signal the Big Idea of the story in the
Big Picture - and make that big picture as big and as dominant as possible”. In the
Mountains text the size of the sketch caricature in relation to the amount of space
taken up by the type shows that it is a very significant part of the page and has an
important part to play in the multimodal transmission of the thesis of the text. This

288
signification is accentuated by the fact that the bottom right-hand corner of the
page is dominated by the colour saturation of the thick blackness of the mountain,
and the vertical and horizontal edges of the corner frame extend well over 50% of
the distance of the page borders. The line graph is also visually salient in terms of
the surrounding type, taking up a significant proportion of the available space in
the top half of the page. This prominence is emphasised via the use of shadowed
boxes for the whole visual frame and the individual line graphs, giving a three-
dimensional sense that they are being raised to prominence from the surface of the
surrounding white space on the page, with the effect of foregrounding them for the
viewer.

In terms of the Degrees of framing of elements on the page, the first and most
obvious feature is the fact that both the visuals in the Mountains text compete to
varying degrees with the verbal aspect of the text for page space. With the sketch
caricature this sense of competition is realised by the fact that there is no uniform
division between the verbal (type) space and the visual space of the sketch
caricature. If there were a typical clear division in framing terms, there would
most probably be a line border, or an orderly (linear) clear space around the visual
to delineate it clearly. Here however, the sketch caricature forces the type to
conform to its shape, a technique referred to as a runaround and purposely used
by graphic designers for effect. White claims it is often seen as a dangerous
technique in design terms however, because whenever it is used there must be
some sacrifice in legibility (White 1982:106-7). This also creates problems for the
typesetter in that the type has to be arranged around the visual in such a way that
the reader’s reading path is not too disrupted, so that there is a clear even space
between the visual and the type, and so that the arrangement of the elements on
the page in relation to each other does not create a sense of disunity, or a sense of
too much disturbance or disharmony.

The dangers in using the runaround can be minimised by adapting the type in
some way (usually by changes in font, size or bolding), but in the Mountains text
this is not done. The severity of this “battle” for space created by the variable
runaround on the page between the sketch caricature and the type is however

289
compensated for by the use of the visual as a page corner framing device. The
bottom right corner of the sketch caricature acts as a continuation of the page
frame, thus lessening the sense of conflict and giving it a greater sense of
inclusion with the type, and a feeling of flow and of compositional unity with the
whole page. The bottom right hand corner of the visual actually replaces the type
and conveys the visual sense that it is part of the visual frame. Despite this
competition for space, the overall effect for the reader is of compositional
complementarity, that the sketch caricature forms an integral part of the page, and
the two modes in a sense “melt” or blend with each other. In this way, there is an
intersemiotic compositional co-operation which conveys to the reader a sense of
visual unity, and lends compositional support to the intersemiotic semantic
relations in the text.

The same can be said about the visual-verbal compositional interaction between
the type and the line graphs. Here the type is again forced to conform to the visual
via the use of a linear, evenly spaced runaround. However, the disruption to the
type caused by this technique is minimised by the effect created by a more
conventional linear border used to cordon off or frame the visual in a clearer and
more severe way. There is some contention for page space, but the more
conventional runaround in combination with the linear border allows the reader to
retain a sense of uniformity and complementarity between the two semiotic
modes.

An examination of the visual to visual interface in the Mountains text shows some
interesting features in terms of how the two different visual coding orientations,
naturalistic and mathematical, complement each other in supporting the realisation
of the intersemiotic complementarity between the visual and verbal modes. The
intersemiotic relation relevant here has been referred to in Chapter Five as Inter-
visual synonymy, a relation which is concerned specifically with the degrees of
semblance in form across visual modes that work to present a kind of cross-modal
harmony, or a compositional intersemiotic complementarity. This
complementarity can occur whether the visuals are from the same or different
coding orientation — in either case there is the potential for some semblance in

290
form, shape or colour across the page which can work to mirror the meanings
expressed by both visuals.

This kind of inter-visual synonymy can be seen between the sketch caricature and
the line graphs, where there are subtle visual harmonies created by the axes they
contain. In both visuals, the information that is to be conveyed is above and to the
left of the point of origin from which the actual and perceived “x, y” axes can be
derived. This axial similarity across modes conveys to the viewer a sense of visual
harmony in that the focus of the information they are both presenting is a left to
right, climbing/ascending, rising/falling action, and a sense of energy, dynamism
and variability. These are realised by the intersemiotic correlations between the
upward slope and the rising graph lines, the ledge and the graphic peak (two
pausing places), and the downward slope and the falling graph lines.

This compositional complementarity is even further reinforced in the sketch

caricature almost exactly mirroring the right-hand line graph’s movement — the
sketch caricature represents a movement from left to right up a steep(ening) slope,
there is the ledge or peak, and there is a further potential action rising up a steeper
slope. The semantic thrust of the verbal headlines ‘Slumping ……mounting’ lend
verbal support also, where the sketch caricature shows someone mounting a slope,
and the right-hand line graph shows the numbers of something also rising. The
content-focus of both the visuals is supported as well through the use of the verbal
labelling of ‘Lloyd’s’ in the sketch caricature, and the labelling of the source of the
graphic data in the line graphs as being derived from ‘Lloyd’s of London’. There is
therefore not only a directional and configuration-based synonymy between the
visuals, but also lexical repetition and synonymy to reinforce the intersemiotic
complementarity in compositional terms they realise.

The importance of potential reading paths is related to the ways that the page
space is approached, or the directions which readers of a multimodal text take
when they interact with an article. As discussed in Chapter Five, in the SFL
model, the textual metafunction is realised in those grammatical and discourse
features which serve to organise text, and these resources provide readers with

291
cues as to where they are and where they are going. In English, what is expressed
first, or is in the left hand position in the reading path, is very often what carries
information which may be retrieved from the context, may be considered given, or
understood as known by the reader. This Given-New structure is mapped onto the
thematic structure of a text, providing a “composite texture to the discourse and
thereby relat[ing] it to its environment” (Halliday 1994:299). In English the first
position in a clause expresses an important and distinct kind of meaning; it signals
what the message is about. This is a kind of signpost for the development of the
text, and is referred to in SFL as the THEME, or “what the message is concerned
with: the point of departure for what the speaker is going to say” (op.cit.38). The
rest of the clause, or the new information, is referred to as the RHEME. The
semantic relationship between Given-New information structure and thematic
structure in language occurs as a result of a speaker choosing a THEME from
within what is Given, and then positioning the important subject matter of the
New somewhere in the RHEME (op.cit.299).

In the visual semiotic system however, the left to right orientation of the reading
path in English and the organisation of meaning into a Given-New contrast is also
of importance in organising the ways that meanings are represented to the viewers,
but in ways that are of course peculiar to the visual mode. As explained in Chapter
Five, in a visual semiotic there are strong culturally-based conventions for
interpreting how visuals can be read, one of these being the left to right reading
path orientation and the visual expectancy that this can produce in viewers. The
discussion which follows will attempt to examine how this sense of expectancy in
visuals can be reinforced intersemiotically by the discourse organisation in the
verbal text, producing compositional intersemiotic complementarity.

Interpreting visual-verbal intersemiotic relations in the Mountains text in the light

of the Given-New organisation, and assuming that the visuals should and most

292
Given/ Narrative Sketch Caricature Names Line Graph Syndicates Line Discourse Stage
New Stage Graph

The slope below the The movement of the The movement of the Reporting the background.
boulder represents the graph over the years graph over the years 1982 Predictions/comment on what has occurred with this topic in the past,
GIVEN PAST past (the path already 1982 to January, 1993. A to January, 1990. A and leading up to the current situation. It reports on Lloyd’s problems: its
travelled in terms of report on past data. report on past data. losses, decreasing names, lawsuits and the effects on the Lloyd’s market.
losses, names and open Makes predictions about the release of new loss figures in June 1993,
years). and the effect of these losses on the number of open year syndicates.

The point where the The point on the graph The point on the graph Reporting the current situation. A report on what is happening now as a
boulder is actually which shows the number which shows the number result of the background events discussed in the preceding section. What
GIVEN PRESENT touching the mountain of names as at March 27, of open years as at March is happening now is the unveiling of a business plan by the new
represents the present 1993, the date of the 27, 1993 (note the system management team of Rowland and Middleton to attempt to deal with the
situation, that of Mountains text of reporting accounts already mentioned problems.
struggling with current publication. back-dated 3 years).
problems.

The slope to the right of No information. The movement of the Problematising the situation. A discussion, analysis and comment on
the boulder, the ledge graph over the years from three problem areas which The Economist magazine feels should be
and the even steeper June 1990 to early 1991. addressed in order for the current situation to resolve itself. These
slope represents the An obvious prediction problems are treated and analysed in order: the number of open years,
NEW FUTURE future path (a prediction about the graph’s future litigation by names’ syndicates, and financing the 1990 losses.
by the writers about the movement since no data
nature of the future is available due to the Concluding and commenting.
problems confronting system of back-dated An opinion expressed about the current situation (the two managers’
Lloyd’s). reporting. willingness to listen to names, policy-holders and prospective investors),
a modulated predictive and summarising conclusion (Lloyd’s may not
survive if it does not meet the three problems discussed), and a final
coda-comment (the stakes are very high - i.e. Lloyd’s survival).

Table 7.10 Discourse staging and visual narrative staging

293
likely would be read with a left to right reading path by its potential viewers, we
find a clear link between the narrativisation inherent in the visuals and the text’s
organisation at the level of discourse. The action portrayed in both the sketch
caricature and the line graphs is not simply projected as action in isolation, but is
in fact action as part of a narrative which through visual means projects past,
present and future meanings. This narrativisation in the sketch caricature is seen in
the profile of the actions of the two men attempting to ascend the mountain-side
under a heavy burden, where the Given is the represented past path they have
already trodden (to the left of the boulder and below their feet) and the represented
current position of the boulder (the point where it touches the mountain). The New
however is the represented upper slope (the expected future or path they have to
tread (all the slope to the right of the boulder).

This narrativisation can also be seen in the line graphs where the action in the
graphic lines’ movement up and down along the vertical plane is to be interpreted
according to the passage of time (these movements and what they represent in
quantitative terms are mostly Given in that they are reporting the past and the
current situation, or that which should be understood as fact — the already-
mentioned quantitative prediction with the open year syndicates figures can be
treated as New). The narratives portrayed in these visuals are realised in visual
terms by the already-discussed vectors produced by the represented participants,
the culturally-based left to right reading paths, as well as the supplementary
support provided by the effects of visual salience and balancing centres produced
by the visuals themselves. The discourse and visual narrative staging in
accordance with this Given-New and past-present-future staging can be seen in
Table 7.10.

Table 7.10 summarises the interactive, intersemiotic complementarity relationship

the narrative messages of the visual modes have with the unfolding stages of the
Mountains text’s verbal discourse structure. What this table shows clearly is that
the division into past, present and future time periods or scenes in the two visuals,
as well as their status as Given or New information, is intersemiotically
complemented by the stages of the discourse structure of the Mountains text. In

294
the same ways that both the visuals report on the past in terms of what has
happened with Lloyd’s and its problems in recent years, so too does the verbal
aspect of the text, which sets up the analysis of Lloyd’s problems leading up to the
current situation by referring to the major problem areas it has been dealing with:
decreasing names, lawsuits, the effects of these on the market. The present
situation of Lloyd’s is projected in visual and verbal terms also, in that the line
graphs show current data, and the sketch caricature represents the two men and the
boulder at a certain point on the mountain-side. This is mirrored in the verbal
aspect of the text by a discussion of the rescue plan and the people involved with
it. Both the past circumstances and the current situation reportage are to be
considered as Given in that their elements have been covered in the texts
published in the two years prior to the publication of the Mountains text — again
this is evidence to support the importance of the text’s intertextual history as an
influence on the text’s current form. The New or future information is also
presented by the sketch caricature (the upper slope) and to a degree the syndicates
line graph (data over the years 1990-1991), and this is complemented by the
verbal analysis of the current situation, and the opinions and predictive
conclusions that are expressed and which are based on that analysis.

7.5 Discussion of Results and Summary

The analysis carried out in this and Chapter Six of this study has attempted to
show that a metafunctional interpretation of the visual-verbal intersemiotic
relations in a multimodal text would prove fruitful in interpreting the ways that
these intersemiotic relations are realised. It has also been proposed that both the
verbal and visual modes project their meanings in concert via semantic relations
which realise intersemiotic complementarity, and that these meanings are
dialectically related to various contextual features of the Mountains text, as well
as aspects of The Economist magazine’s context of culture.

The analysis in Chapter Six of the context of the Mountains text and The
Economist magazine as an eco-financial publishing institution revealed that apart
from being situated in the broader UK and world economic, financial and eco-
political environment, a specific multimodal text like the Mountains text is also

295
situated in a context of creation which includes the ways that the artistic
conventions of the authors’ community, their conception of their audience, and
their individual preoccupations, are built into the text. The analysis of aspects of
The Economist Style Book, The Numbers Guide, and the interview with the head
of the graphic design team at The Economist’s headquarters illustrated the ways
that The Economist’s own publications and the graphic design team’s writing and
graphic design policies can influence the nature of the end-product, a multimodal
text — important factors found were concerned with the team’s conceptions of the
magazine’s page-based medium, their ideas about relative primacy of the visual
and verbal modes, the general subject-matter the magazine covers, the
recommended typographic and writing/stylistic conventions, and various
institutional preferences in terms of visual and verbal typology. The analysis and
in-house publications also gave certain indications about The Economist magazine
staff’s perception and attitudes towards its readership, and how they interpret their
roles as a writers/publishers dealing with economic, financial and eco-political
issues.

Chapter Six also demonstrated the ways that the multimodal Mountains text is a
product of its environment, and that it functions in that environment by being both
activated by the context in which it occurs, and simultaneously working to
construe this context. The Mountains text was also seen as a multimodal text that
has been influenced by or is the product of other texts, especially those which
have appeared before it. This text’s intertextual history is therefore important,
because the many articles published beforehand in The Economist deal with
relevant issues that have developed over time, and are related in varying ways to
the Mountains text. This includes the text’s subject matter and the issue addressed,
the attitudes expressed towards this issue, and the ways that the magazine has
produced them in visual and verbal terms. It was suggested that these contextual
features could be used to inform an intersemiotic interpretation of the multimodal
Mountains text in terms of three main hypotheses.

Firstly, it was hypothesised that intersemiotic complementarity will obtain when

the ideational meanings in both modes are related lexico-semantically through the

296
intersemiotic sense relations of repetition, synonymy, antonymy, meronymy,
meronymy, and collocation.

In the analysis of the intersemiotic ideational features of the Mountains text, it was
found that the experiential meanings (glossed as VMEs) embodied in the
processes, participants, circumstances, and attributes presented visually in the
sketch caricature and line graphs were semantically complemented by the verbal
mode. Firstly, it was shown that the visually represented processes of “Climbing -
pushing up (enacting solutions with effort)”, “Graphic focus - increasing loss”,
and to a lesser degree “Slumping … mounting” were repeated, synonymised, and
collocationally referred to by a range of identified lexical items in the verbal
aspect of the text.

Secondly, it was also clearly shown that the visually represented participants were
to varying degrees repeated, synonymised, and collocationally referred to by
various lexical items, as in the significant repetition and synonymising of the
institution of “Lloyd’s”, “Source: Lloyd’s of London” and “Active names”, and
the repetition of the names of the two financiers, “Rowland and Middleton”.
Further, there were lexical items which were semantically related to the visuals’
represented participants meronymously by referring to the part of the whole of
some phenomenon, as in lexical items referring to parts of the Lloyd’s of London
institution, particular parts of the time period under discussion, and to a lesser
degree to numerical figures which formed part of the sets of numbers graphically
portrayed. There were also other lexical items which were semantically related to
the visuals’ represented participants hyponymously, by being referred to as a sub-
type of some category of phenomenon, as in the treatment of the types of problems
that Lloyd’s was having (the “boulder”), or the different types of solutions that the
rescue plan would have to cover (the “upper ledge/slope”).

Finally, it was shown that the visual features represented as circumstances (the
whole “Mountain”), and the supplementary attributes (“struggle and effort” and
“financial/banking executives”) were also, to varying degrees, semantically
related to the verbal aspect of the Mountains text by being repeated, synonymised,

297
and collocated. An interesting aspect is that the visual feature represented as
circumstances (locative: the whole “Mountain” as a chronological view) is also
semantically related to the verbal aspect of the Mountains text by being
meronymised, in that there are a number of references to parts of the time period
which describes Lloyd’s past, present and future. In intersemiotic ideational terms
therefore, the analysis of the Mountains text shows that both visual and verbal
modes do ‘work together’ on the page, at the very least in terms of the general
subject matter, and that this intersemiotic complementarity is realised mainly
through the significant usage of intersemiotic repetition, synonymy and
collocation, intersemiotic sense relations which perform the function of
introducing and maintaining the topic and subject matter. This usage is supported
to a lesser degree by the intersemiotic sense relations of meronymy and
hyponymy, both of which seem to relate in a supplementary way to the
discussions and analysis of the main subject matter of the text.

Secondly, it was also hypothesised that intersemiotic complementarity will obtain

when the interpersonal meanings in both modes are related through intersemiotic
reinforcement of address, and through intersemiotic attitudinal congruence and
attitudinal dissonance (modality) relations.

In the analysis of the intersemiotic interpersonal features of the Mountains text, it

was found that the readers and viewers were addressed by both modes in the same
the ways, and that this was realised by the intersemiotic interpersonal relation
referred to as reinforcement of address. In the visual mode, both the sketch
caricature and the line graphs were interpreted as making a range of visual
statements. In the sketch caricature the absence of any gaze or facial expressions
towards the viewer indicating a question is being asked (no vectors can be drawn
from some point of origin to the viewer’s face), or gestures which command (no
vectors towards the viewer), suggested that it is simply offering information to the
viewers. It was also found that these visual features were supported by the ways
that the sketch caricature utilised mechanisms to realise little or no involvement
(horizontal angles), neutral or equal power relations (vertical angles), and degrees
of social distance (size of frame) between the viewers and the image. These

298
findings were also applicable to the line graphs (except size of frame - realising
social distance) — these were also interpreted as offering information, since the
represented (quantitative) participants form no other relationship to the viewer(s)
than being simply a display of numbers and graphic lines interacting with each
other to present information or data which can be agreed or disagreed with, or
acknowledged or contradicted. It was found that the verbal mode reinforced this
visual form of address by consisting entirely of verbal statements about the subject
matter discussed, and that this could be seen in the MOOD structure of the
clauses, all of which utilised the order Subject/Finite, the order which realises the
Declarative and the speech function of making statements.

The analysis also found that the mode-based statements with which the readers
and viewers were addressed were characterised by the intersemiotic interpersonal
relation referred to as attitudinal congruence. Each of the VMEs which were
analysed in the ideational analysis were examined here in terms of the kinds of
attitudes portrayed in both the visuals, and also how they were treated attitudinally
in terms of Modality features in the verbal aspect of the text. Three general
intersemiotic features were found — the first is the attitudinal congruence between
the definite statements made by the line graphs and the discussion of them as
established fact in the verbal aspect of the text; the second is the attitudinal
congruence between most of the VMEs in the sketch caricature and the line graphs
and the high modality displayed in their treatment in the verbal aspect of the text,
and the third is the attitudinal congruence between the VME dealing with Lloyd’s
future in the sketch caricature and the relatively lower modality in evidence in the
verbal aspect of the text.

The variations in the modality were linked to the nature of the visual coding
orientations utilised in the text, in that the higher modality shown for the line
graphs is because they are mathematical statements which are backed by an
‘official’ imprimatur (“Source: Lloyd’s of London”) and thus are able to make
firm reportage statements. The sketch caricature, though a whimsical
representation, does show a scene which metaphorically refers to a realistic
situation of which the viewers and readers of The Economist magazine on this

299
issue would be well aware — the problems of Lloyd’s and the rescue package
being organised by Rowland and Middleton. It is a ‘believable’ situation in that
sense, which is caricatured for reasons of attraction rather than to cast doubt on the
veracity of its propositional content. The relatively lower modality accorded the
caricatured representation of Lloyd’s future and the references to it in the verbal
aspect of the text is of course due to the fact that the sketch caricature portrays an
almost impossible future (the upper mountain-side) for the rescue plan — this is
matched by the verbal references to the plan, which suggest that its success is not
so assured.

The analysis also revealed that attitudinal congruence between the visuals and the
verbal aspect of the text is realised intersemiotically through the attitudinal
attributes and adjectives assigned or used with various represented participants.
The intersemiotic attitudinal features relevant here are the use of forms of address
and attitudinal adjectives in the verbal aspect of the text which cohere in a
synonymous way with the visual attitudinal message elements. The first feature
examined was the nature of the references or forms of address used for Middleton
and Rowland in the verbal aspect of the text, and how this complements their
represented visual attributes. It was found that both men are portrayed only in
such a way as to facilitate recognition by the viewers, and as such are treated with
a modicum of respect by being portrayed in the pin-striped suits which accord
them some kind of ‘official’ status — they are not portrayed as being incompetent
or silly, or in a disjunctive way by being out of the ‘uniform’ of the finance
industry. The analysis here revealed that this portrayal of their attributes is
mirrored in the forms of address used for identifying them in the verbal aspect of
the text.

The other significant instance of intersemiotic complementarity in terms of

attitudinal features were the references made to the magnitude of Lloyd’s
problems, as represented by the visually salient boulder, and by the use of
adjectives referring to Lloyds’ problems in general, the size of its problems with
increasing losses, and the need for raising enough sums of money. The occurrence
of these two significant uses of support for the attitudes expressed in the visuals

300
lends weight to the interpretations of the visuals in terms of their main visual
message focus — that of the problems that the management of Lloyd’s are having
and will have in the future with the magnitude of its financial problems. Thus, in
both the modes the two men are represented and referred to in reasonably
respectful terms, and the attitudes projected regarding the magnitude of the
problems that Lloyds is having are also attitudinally congruent. Thus we have, in
interpersonal terms, further evidence to support the central proposition of this
study, that both the verbal and visual modes in the Mountains text
intersemiotically complement each other to produce a coherent multimodal text
through reinforcement of address and attitudinal congruence.

The lack of attitudinal dissonance was also briefly discussed, in that opposite or
ironical attitudes displayed between the two modes were not in clear evidence. It
was suggested that this analysis shows that both modes are placed on the page to
work together synonymously in terms of their propositional content, to offer
information via both verbal and visual statements, and to ensure that the ways that
it is received and dealt with by the viewers and readers are in agreement. This was
interpreted as being in accord with the findings of the analysis of The Economist
magazine’s contextual features, as shown in the expressed aims of the graphic
design team at the magazine; i.e. images are generally designed and used to attract
the potential viewer/reader in some way, and mathematical visuals are used to
support the informational content contained in the article (although this comment
should be mediated by the fact that line graphs like those used in the Mountains
text do contain an element of editorialising, as the graphic headings “Slumping ...
mounting” show).

Finally, it was hypothesised that intersemiotic complementarity will obtain when

the compositional meanings are integrated by the compositional relations of
information value, salience, visual framing, visual synonymy, and potential
reading paths.

The analysis of the compositional aspects of the Mountains text revealed that it is
in reality a complex interplay between the visual and verbal modes in terms of

301
such compositional relations as information valuation on the page, salience on the
page, and degrees of framing of elements on the page, as well as supplementary
influences provided by inter-visual synonymy, and the importance of potential
reading paths.

The examination of information valuation on the page showed that the vertical or
top/bottom placement of the visuals in the Mountains text is an important
compositional influence — the sketch caricature for example is placed in the lower
status half of the page, but this is compensated by the primary position it occupies
in relation to the spine, and its visual salience in terms of its size and colour
saturation. The line graph is however placed in the centre of the top-half of the
page, a primary position with visual weight and a divider or balancing centre
between the two halves of the type. The line graphs act as a stabiliser of this
‘weight’, where the visual elements in the central area assist the objects outside
the centre zone to be unified and stabilised by being grouped around a balancing
centre. The central placement of this visual is thus stabilising, and the sense of
compositional unity and intersemiotic coherence across the page space is
maintained.

The relative size, or salience on the page of the sketch caricature and the line
graphs in relation to the amount of space taken up by the type shows that they are
a very significant part of the page and have important parts to play in addressing
the reader/viewers. This is accentuated by the colour saturation of the thick
blackness of the mountain, which helps to catch the eye of the potential
reader/viewer (again, an aspect of the graphic design policy at The Economist),
and the fact that the edges of the corner frame work as part of the page borders.
The line graph is also salient on the page, taking up a significant proportion of the
available space in the top half of the page and forcing the viewer to take notice of
it.

The analysis of the degrees of framing of elements on the page, shows that both
the visuals in the Mountains text compete with the verbal aspect of the text for
page space to varying degrees. The sense of competition of sketch caricature with

302
the verbal aspect of the text is realised by the lack of a uniform division between
the type space and the visual space. The sketch caricature forces the type to
conform to its shape through a runaround which causes a kind of “battle” for
space. This conflict between type and visual is however compensated by the use of
the visual as a page corner framing device. The overall effect for the reader here
therefore is of intersemiotic compositional complementarity, where the sketch
caricature forms an integral part of the mostly verbalised page.

Two further aspects of the intersemiotic compositional complementarity displayed

by this multimodal text were found in the visual to visual interface in the
Mountains text. This showed that the two different visual coding orientations,
naturalistic and mathematical, also complement each other in supporting the
realisation of the intersemiotic complementarity between the visual and verbal
modes. This is concerned with inter-visual synonymy, which is concerned
specifically with the degrees of semblance in form between the sketch caricature
and the line graphs — the analysis revealed that there are subtle visual harmonies
created by the axes both the visuals contain. In both visuals it was found that the
information they attempt to convey is above and to the left of the point of origin
from which the actual and perceived “x, y” axes can be derived, and that this axial
similarity across modes conveys to the viewer a sense of visual harmony which
can be derived from the focus of the information they are both presenting — a left
to right, climbing/ascending, rising/falling action, and a sense of energy,
dynamism and variability. Inter-visual synonymy was also found between the
upward slope and the rising graph lines, the ledge and the graphic peak (two
pausing places), and the downward slope and the falling graph lines. One final
aspect of this is that the compositional complementarity between the visuals is
even further reinforced by the fact that the sketch caricature almost exactly
mirrors the right-hand line graph’s movement — the sketch caricature and line
graphs move upwards from left to right on a steep(ening) slope, there is the ledge
or graphic peak, and there is a further potential action rising up the steeper slope
of the mountain-side and the graph.

303
These compositional intersemiotic relations were also revealed in the
chronological (past-present-future) and Given-New complementarity displayed
between the text’s verbal discourse stages and the visual narrative stages. Here the
importance of potential reading paths is related to the ways that the page space is
approached, or the directions which readers of a multimodal text take when they
interact with an article. The analysis of the Mountains text revealed that there is a
clear link between the narrativisation inherent in the sketch caricature and the
text’s organisation at the level of discourse. The action portrayed in both the
sketch caricature and the line graphs is action as part of a narrative which through
visual means projects past, present and future meanings. This narrativisation in the
sketch caricature and the division into past, present and future time periods or
scenes in both the visuals, combined with their status as Given (past and present
situation) or New (the expected future) information, is intersemiotically
complemented by the stages of the discourse structure of the Mountains text.
Thus, the typical reader/viewer who may start with the ‘attract’ images like a
sketch caricature will be immediately exposed to the narrativisation presented in
terms of the Given and New information; if the reader/viewer then commences to
read on, he or she will find that this visual narrativisation is synonymised by the
unfolding discourse of the verbal aspect of the text. If, in the more unlikely and
unexpected case, a potential reader/viewer starts to read the verbal aspect of the
text first, and then either goes back and forth between the visual and the verbal, or
looks at the visuals after reading the whole of the verbal, the same synonymy
between the visual narrativisation and the Given and New of the discourse will
still hold — it will just have been approached in a different way. There is
therefore still a Given-New intersemiotic complementarity, no matter what
potential reading path the reader/viewer may follow.
Thus, the analysis of these aspects found that in the same ways that both the
visuals report on the past in terms of what has happened with Lloyd’s and its
problems in recent years, so too does the verbal aspect of the text, which sets up
the analysis of Lloyd’s problems leading up to the current situation by referring to
the major problem areas it has been dealing with: decreasing names, lawsuits, the
effects of these on the market. The present situation of Lloyd’s is projected in
visual and verbal terms also, in that the line graphs show current data, and the

304
sketch caricature represents the two men and the boulder at a certain point on the
mountain-side. This is mirrored by the verbal aspect of the text in a discussion of
the rescue plan and the people involved with it. Both the past circumstances and
the current situation reportage are to be considered as Given in that their elements
have been covered in the texts published in the two years prior to the publication
of the Mountains text — again this is evidence to support the importance of the
text’s intertextual history as an influence on the text’s current form. The New or
future information is also presented by the sketch caricature (the upper slope) and
to a degree the syndicates line graph (data over the years 1990-1991), — this is
complemented by the verbal analysis of the current situation, as well as the
opinions and predictive conclusions that are expressed and which are based on
that analysis.

The intersemiotic analysis of the Mountains text in terms the three intersemiotic
metafunctions, the ideational, the interpersonal, and the compositional, has
therefore demonstrated that the intersemiotic complementarity found in this text is
realised via a complex interplay of various intersemiotic relational elements, such
that any reader of this multimodal text would need to interact with this complexity
by being required to simultaneously:
1. Comprehend the mode-based processes, participants, circumstances and
attributes which are represented, as well as the semantic relationships between
them.

2. Respond to the mode-based ways that he or she is being addressed, and the
attitudes that are presented in concert with that address. This may be in terms of
any potential visual or verbal statements being made or questions asked, as well
as a variety of projected attitudes and judgements about the information
presented.

3. Appreciate in visual and verbal terms the coherence between one part of the
article and every other part, as well as the relative information values assigned
to the messages delivered by each mode.

Based on the analysis carried out in this study therefore, the multimodal
Mountains text can be interpreted as a coherent metafunctional construct in that it
is a mode-based complex of simultaneously interacting ideational, interpersonal
and compositional meanings.

305
Chapter 8: Conclusion

8.0 Concluding Summary

It is appropriate at this stage to briefly restate what has been covered in this study.
The development of the analysis of intersemiotic relations in page-based
multimodal text began in Chapter Two with a review of the literature dealing with
the analysis of economics discourse by applied linguists and economists. This
highlighted the nature, focus and findings of various studies of economics
discourse, and evaluated the extent of their treatment of visual modes of
communication. It revealed that the published research by applied linguists and
economists has not attempted or been able in any rigorous way to describe,
explain or account for the role of the visual mode in economics discourse. In
Chapter Three an overview and brief background to the conception of the sign and
meaning in general semiotic theory as derived from Saussure was given, as well as
an overview and review of the main schools of semiotic theory which in the
twentieth century have attempted to examine visual communication from a
linguistic perspective. This involved a brief description of the Prague School and
the Paris School, with some concentration on the work of Roland Barthes from the
Paris School.

An examination and review of the work of those who have attempted to

investigate various visual modes from within the Hallidayan school of social
semiotics was the subject of Chapter Four. This school, which extends Halliday’s
Systemic Functional Linguistic theory to the interpretation of the ways that
various visual modes realise their meanings in social contexts, was introduced as
the theoretical framework that would inform the analysis of the multimodal text
extracted from The Economist magazine. This review firstly discussed the
usefulness and applicability of the two major models available for the analysis of
images — Kress and van Leeuwen’s grammar of images, and O’Toole’s language
of displayed art — to the kinds of visuals commonly utilised in The Economist
magazine. It concluded that Kress and van Leeuwen's 'grammar of images' is
applicable to explaining visual meaning in The Economist magazine, as well as
how it is organised. Although O'Toole's work was interpreted as being of less use

305
in these terms, his rank scale, schematic model was seen as being applicable for
the analysis of naturalistic or mathematical images, and in terms of relating them
to their contexts of situation. This review also demonstrated that although Kress
and van Leeuwen's and O’Toole’s ideas represent the core of an expanding body
of work examining different types of visual modes, notably in general types of
images and in displayed art, within this particular body of work there is a dearth of
research into the nature of the intersemiotic semantic relationship between the
visual and verbal modes in multimodal text, and none at all in relation to
economics discourse (in images or in intersemiotic relations). It was at this point
that the need for a study examining visual-verbal interrelationships in page-based
texts in economics discourse was clearly established.

Chapter Five presented the theoretical foundations for the intersemiotic analysis of
the sample Economist magazine text, the Mountains text, by briefly overviewing
the fundamental assumptions of the SFL model as a social semiotic theory of
communication. This covered the theoretical principles of the model’s
extralinguistic features: contexts of culture and situation, the metafunctions, and
the dialectic relationship between these and the linguistic levels of the
lexicogrammar. Drawing on the work of Kress and van Leeuwen, O’Toole et.al.,
as well the linguistic principles outlined in the SFL model, an analytical
framework for examining any potential intersemiotic complementarity in The
Economist magazine text was developed. This framework presented various ways
that the SFL model could be extended to apply not just to language or to visual
modes, but to the analysis and elucidation of the intersemiotic semantic
relationships between them.

Chapters Six and Seven focussed on a detailed analysis of the Mountains text.
Chapter Six, in line with the context-driven nature of the SFL model, presented
general background information on The Economist magazine as an institution
concerned with economic and financial journalism, its institutional history, and its
editorial policy in terms of the ways that it aims to approach its readers through
both verbal and visual means. It was pointed out that the multimodal Mountains
text is, like all such texts of its type in The Economist magazine, a particular

306
contextual configuration of the variables Field, Tenor and Mode which is situated
in a particular context of creation. Here The Economist magazine’s staff’s writing
and graphic design policies, their perception and attitudes towards their
readership, and how they interpret their roles as writers/publishers dealing with
economic, financial and eco-political issues were examined.

The analysis in this chapter also attempted to relate these broader issues to the
analysis of the Mountains text by examining the more specific features of the
text’s context of situation, as well as aspects of its intertextual history. These
discussions dealt with the subject matter and issue being addressed (Field-related),
the attitudes that were expressed towards this subject matter/issue (Tenor-related),
and the ways that the magazine composed these features in visual and verbal terms
on the page (Mode-related). These extra-visual and extra-linguistic variables were
interpreted as having an important bearing on the production and subsequent
interpretation of the Mountains text.

Chapter Seven analysed the Mountains text by applying the analytical framework
presented and outlined in Chapter Five. This application of the analytical
framework was also carried out in terms of the informational insights garnered
from the contextual analysis in Chapter Six. The Mountains text, which along with
verbal text includes both a sketch caricature and a mathematical visual (two line
graphs), was analysed in terms of the ways that intersemiotic complementarity is
realised in the intersemiotic ideational, interpersonal, and compositional features
between these modes.

In summary then, this study has problematised the visual-verbal semantic interface
in the Mountains text by questioning the functioning of the visual vis a vis the
verbal modes in this form of page-based media. In doing so, it has tested the
Hallidayan claim of the inter-relatedness of systems of meaning, and has explored
the proposition that both the visual and verbal modes, while utilising the meaning-
making features peculiar to their respective semiotic systems, work together in the
particular context described to realise a unified, coherent multimodal text for any
potential viewers/readers. The hypothesis underpinning the focus of this study was

307
that in the Mountains text, the visual and verbal modes complement each other
semantically to produce a single textual phenomenon characterised by
intersemiotic complementarity, and that this intersemiotic relationship obtains
when one or more of the intersemiotic ideational, interpersonal, or compositional
metafunctional meanings occur. Inherent in this hypothesis was the assumption
that the realisation of this intersemiotic complementarity did not imply that the
visual semiotic and verbal semiotic merely co-occur on the page space and are
related by simple combination — rather, it was implied that the visual and verbal
elements on the page can combine semantically to produce a single, coherent
multimodal text that in combination can have the potential to produce a more
elaborate mix of meanings than would occur if the visual or the verbal aspects
appeared in single-mode form.

The analysis of the multimodal Mountains text in terms of these hypotheses

showed that the incidence of intersemiotic complementarity between the visual
and verbal modes is potentially realised in this text in three simultaneously-
occurring areas. Firstly, the visual and verbal ideational meanings were found to
be lexico-semantically related to each other by means of the intersemiotic sense
relations of repetition, synonymy, antonymy, hyponymy, meronymy, and
collocation. Secondly, the visual and verbal interpersonal meanings were found to
be related through intersemiotic reinforcement of address, and intersemiotic
attitudinal congruence, and thirdly the visual and verbal compositional meanings
were found to be integrated through the compositional relations of information
value, salience, and visual framing, as well as the influence of visual synonymy,
and potential reading paths.

In conclusion, three outcomes were expected of this research, all of which have
been realised to varying degrees. The first outcome is that the visual and the
verbal modes in the Mountains text have been found to work together
semantically on the page to produce a coherent multimodal text, and that the
intersemiotic semantic resources which realise this intersemiotic complementarity
have been identified. The motivating question for this study which was originally
asked in Chapter One has therefore been answered.

308
Secondly, the results of this study have provided evidentiary support for Halliday
and Hasan’s (1985:4,10) assertion of the interrelatedness of systems of meaning.
This study has therefore applied and extended the Systemic Functional Linguistic
concept of metafunctions to multimodality in text. It has been an effective
qualitative and descriptive study for extending theory through its application.

Thirdly, the results show that some of the concepts and terminology of existing
linguistic theory and their attendant analytical tools are capable of being adapted
to the task of explaining how semiotic systems interact. Thus, the concept of sense
relations in lexical cohesion as outlined by Halliday (1994:330-334) and Halliday
and Hasan (1985:80-82), and the analytical tool of cohesive chains have proved to
be adaptable and useful for a task of this nature.

8.1 Directions for Future Research

The intersemiotic analysis of the Mountains text in this study has revealed it as a
complex multimodal text in which the visual and verbal modes complement each
other on the page in terms of the intersemiotic ideational, interpersonal, and
compositional meanings that it presents to its viewer/readers. It has been found to
be a rich source of evidence for intersemiotic complementarity. It is however only
one multimodal text; many of the features analysed and discussed would bear
further research to confirm and reinforce these findings. Some of these areas are
suggested in the following sections.

8.1.1 Analyses Focussing on Specific Intersemiotic Metafunctions

Each of the intersemiotic metafunctions covered in the analytical framework used
in this study could be individually targeted to examine other types of page-based
multimodal texts, to ascertain whether the results found in the Mountains text can
be replicated in other texts and contexts. The specific analytical focus of these
studies could examine ideational intersemiotic complementarity between any
visuals’ represented participants, processes and circumstances and the co-
occurring verbal features. Alternatively, they could examine interpersonal
intersemiotic complementarity in terms of the various ways that the

309
reader/viewers are related via address and through the projection of attitudes, or
they could focus on finding patterns in the ways that the visual and verbal modes
(single or multiple), interact in page layout to produce intersemiotic
complementarity.

The first and most obvious types of page-based multimodal text that could be
examined are those found in the economics discourse community. Analyses, either
detailed or focussing on one intersemiotic metafunction, could target the different
types of economics discourse identified in Chapter One: the professional
economics journals, economics textbooks, reports by economic organisations, and
reports and analyses of economic issues by other journalistically-oriented print
media (such as newspapers) to ascertain any intra-community differences. Studies
could focus on finding patterns within the discourse types by analysing a larger
corpus, or they could focus on highlighting the similarities and differences
between exemplars. The results of these comparative studies could be especially
useful in educational contexts, since differences revealed in the relative
importance of the visual and verbal in different contexts may assist educators in
tracing the sources of students’ conceptual confusion, or could allow instructors to
tap into the ways that the two modes complement each other so that they can
enrich or even evaluate students’ understanding. A specific example may illustrate
this — in Australian senior high schools it is common to evaluate economics
students via essay questions which use some kind of naturalistic or mathematical
visual as a prompt or stimulus for the actual multiple choice or essay question.
This requires the students to be able to read the visual and to use its represented
features to answer the particular question set. Some understanding on the part of
both the instructors and students of the ways to ‘unpack’ a visual’s message
content, the ways this message is addressed to the viewer, and the attitudes
embodied in the visual, may mean that the students could use this ‘visual literacy’
to answer the specific question set, and could allow them to demonstrate in
writing a deeper understanding of the economic concepts and issues being
addressed by the question. Also, since in these written answers the students are
often expected to utilise diagrams (usually mathematical) to demonstrate their
understanding or provide a proof, an understanding of the ways that the visual and

310
verbal modes complement each other on the page may allow students to develop
more effective answers. A further future benefit is that this understanding could
develop students abilities with the ever-increasing use of computer-based
multimodal texts.

The range of potentially analysable texts could be extended to other multimodal

texts in other commercial arenas and disciplines which commonly combine visual
and verbal modes in their page-based presentations: the most notable here are the
various commercially-available magazines (sports, fashion, computers,
photography etc.), and textbooks in the social sciences (geography, history etc.),
and the hard sciences (biology, chemistry, physics etc.). In this latter area
O’Toole’s work on rank scale in visual representation could be useful in clarifying
the ways that visuals of various complexities represent their meanings at different
levels of complexity, and this could then be related to any co-occurring verbal
component. For example, in Geography and History textbooks at both high school
and undergraduate levels, detailed diagrams representing complex and inter-
related data or concepts are common. Analyses of the ways that the visuals
commonly used in these contexts ‘layer’ or ‘embed’ their information can show
the most common structuring methods used, and reveal content areas that are
confusing for students. Again, developing the skills to ‘unpack’ these visuals by
focussing on their ideational, interpersonal and compositional features can help
students’ understanding and develop their discipline-based skills (which may also
be transferable).

Research could also focus on the co-occurrence of image and verbal text in
advertising, work which would add to and complement the considerable body of
work already published by Barthes (1977), Fiske (1982), Dyer (1982), Danesi
(1994), Vestergaard & Schrøder (1985), et. al. An analysis by Royce (1998b) of a
single-page advertisement from The Economist magazine (which included a
monochrome photograph and accompanying verbal text) showed that while the
photograph (of a young woman) did not seem on the surface to be very complex, it
was in fact an image of multiple layering in ideational and interpersonal terms,

311
and clearly worked with the verbal mode to produce an intersemiotically coherent
multimodal text.

A further area where this study’s analytical framework could be applied is the
intersemiotic relationship between spoken language and the use of accompanying
gesture, or even the analysis of movement portrayed on visual media (TV, movies,
the theatre) and the intersemiotic relationships between the spoken word and the
portrayed actions. The study of movement in visual media is an area of increasing
interest amongst those working with non-page-based multimodality, as the work
by Mercer (1998 in press) on “reconceptualising television realism” illustrates.
Working with the Social Semiotics paradigm offered by Kress and van Leeuwen,
Mercer explores the debate over realism in film and media studies with an analysis
of morning television programs in the UK. He interprets television as a
conjunction of various semiotics: written and spoken language, colour, sound,
visual texture, camera movement, framing, editing etc., and posits that a “text’s
meaning comes out through a reading of all the modes” and that “every text is
positioned with a relationship to social reality and therefore every text establishes
some kind of realism” (op.cit.). Mercer examines in metafunctional terms two
sample texts from morning programming, analyses their conceptions of their
audience and the realisms they create through the representational resources
utilised by the television medium, and shows that both texts create realisms with
very different conceptions of their audiences.

8.1.2 Intersemiotic Complementarity between the Visuals and the Verbal

Discourse
Staging
It is clear that in this study there are areas of the Mountains text which could have
also been examined in some depth, areas which could reveal further evidence of
intersemiotic complementarity between the visual and verbal modes. One of these
areas has already been touched upon in supporting the analysis of intersemiotic
compositional in Chapter Seven. In this analysis, the role of potential reading
paths was examined in terms of the text’s Given/New (past-present-future)
structuring in the sketch caricature and line graphs, and how this related to
elements of the

312
Discourse Unit Paraphrase Sentences

Identifying the a bolded, capitalised section heading provided to situate the text in the 1
department / topic area FINANCE section of The Economist magazine, and providing an initial
orientation to the reader by indicating that the following information will be
financial in nature

Attracting the reader a large font, lower case bolded article headline which is aimed at grabbing the 2
reader’s attention via the use of a familiar catch-phrase, a verbal pun, or a bald
statement. This text’s article or topic heading is “Mountains still to climb”

Stating the thesis one bolded sentence blurb explaining in more detail what the attract headline 3
omits, presenting the thesis or main message of the article. Here it is suggesting
that if Lloyd’s does not solve its problems, then it could face extinction
(bankruptcy).

Orienting the reader two sentences which, in support of the thesis, orient the reader to the article’s 4-5
“slant” or point of view. While not making as strong a case as the thesis it
suggests that further problems are looming despite the progress already made
i.e. “One peak scaled often reveals another”, and applies that opinion to the
subject of the text, Lloyd’s of London.

Reporting the three paragraphs consisting of a total of nineteen sentences which report, make 6-22
background predictions and comment on what has occurred with this topic in the past, and
leading up to the current situation. It reports on Lloyd’s problems: its losses,
decreasing names, lawsuits and the effects on the Lloyd’s market. It makes
predictions about the release of new loss figures in June 1993, and the effect of
these losses on the number of syndicates that are open years.

Reporting the current two paragraphs consisting of a total of 9 sentences which report what is 23-31
situation happening now as a result of the background events discussed in the preceding
section. What is happening now is the unveiling of a business plan by the new
management team of Rowland and Middleton to attempt to deal with the already
mentioned problems.

Problematising the seven paragraphs consisting of thirty-one sentences which involve a 32-62
situation presentation of and discussion/analysis/comment on three problem areas which
The Economist magazine feels should be addressed in order for the current
situation to resolve itself. These problems are treated and analysed in order: the
number of open years, litigation by names’ syndicates, and financing the 1990
losses.

Concluding and a single paragraph consisting of five sentences which express an opinion about 63-67
commenting. the current situation (the two managers’ willingness to listen to names, policy-
holders and prospective investors), provide a modulated predictive summarising
conclusion (Lloyd’s may not survive if it does not meet the three problems
discussed), and inserts a final coda-comment (the stakes are very high - i.e.
Lloyd’s survival).

Table 8.1 Schematic structure in the Mountains text

313
discourse structure of the verbal aspect of the text. The Mountains text could also
have been analysed in ideational metafunctional terms at the level of its schematic
structure, to ascertain the ways that both visual and verbal modes project and
allow for the unfolding of their meanings in stages.

The schematic structure of the Mountains text is presented in Table 8.1, and has
been derived in accordance with Martin’s (1992) usage i.e. the socially-
determined steps or stages that people must go through in various written or
spoken interactions.

The division of the Mountains text into these staged, functional constituents has
been carried out in accordance with the steps suggested by Eggins (1994:37-38),
who suggests that the criteria for the division of a text into a series of stages/parts
can be done according to the functioning of the different constituents in relation to
the purpose of the whole text. She writes,
in assigning labels, the aim is to describe what the stage is doing,
relative to the whole, in terms as specific to the genre as can be
found. “Empty” functional labels such as Beginning, Middle, End,
or Introduction, Body Conclusion should be avoided since they are
not genre-specific (all genres have Beginnings, Middles and Ends).
Instead, to find labels, ask, for example, “what is being done in the
beginning of this text?” or “what is being done in the body of an
essay that is different from what is being done in the body of a
transactional genre?”, etc. (op.cit:38)

In Chapter Seven, only those elements of the schematic structure which related to
narrative staging were examined, but here the ways that both the VMEs and the
full schematic structure relate intersemiotically across the page are examined.

As the following brief discussion shows, evidence of intersemiotic

complementarity can be found through an analysis of the intersemiotic
relationship between the ideational VMEs and the schematic structure of the
Mountains text. The analysis of the VMEs represented by the caricatured
mountain-side, boulder and two men in the sketch caricature shows that they may
be viewed in composite terms as a visual metaphor which relates to specific parts

314
of the schematic structure of the Mountains text. The represented participants and
the interaction between them in this visual presents a clear message that solving a
current difficulty often does not mean that all is solved, in that there could be
greater problems ahead. This is realised by two of the represented participants
performing an action on a steep mountain-side ending with a ledge (a place of
rest), but also an even steeper slope (suggesting the impossibility of the task)
looming up ahead. This composite visual message is verbally reinforced and
repeated by the headline attract ‘Mountains still to climb’. Further, in the sketch
caricature the most visually salient represented participant is the mountain, and
this salience is reinforced by the first word a browsing reader will meet:
‘Mountain’. Mountains can be seen as obstacles, something to be climbed, so
there is evidence here of a clear attempt to cue the reader into the way that the
topic of the article will be treated. There is thus a clear intersemiotic semantic link
between the headline and the sketch caricature in working co-operatively towards
attracting the reader to the article and its topic. This is also, as stated earlier in the
contextual discussions, one of the objectives of the writers and graphic designers
at The Economist magazine.

The composite action on the mountain also relates semantically to both the stating
the thesis and orienting the reader stages in the discourse structure. The thesis is
basically the message that if Lloyd’s does not solve its current and possible future
problems in terms of its capital providers (the names) and the people it offers
insurance to (its clients), then it could face extinction (bankruptcy). This message
is paralleled by the action between the represented participants in the sketch
caricature, but especially so when the second, upper slope of the mountain-side is
considered. Advancing up the slope that the two men are depicted as trying to
climb could well be seen as a manageable task, especially since there is a ledge to
rest on. However, once this slope is overcome there is a further, steeper slope to
deal with, one that according to a common sense understanding of gravity and
physics would seem to be impossible to climb, and possibly life-threatening. Here
the attitudes of the writers are visually suggested, and this interpretation is
strengthened by the verbal support provided by ‘If April’s business plan for
Lloyd’s does not …… the market could die’. This attitude is immediately and

315
further reinforced by the orientation given to the reader via the two sentences
suggesting that once the present difficulties are overcome there may be more,
perhaps impossible tasks ahead, as in ‘One peak scaled often reveals another’
followed by ‘So it is at Lloyd’s, London’s insurance market’. These sentences
require the reader to apply this loaded interpretation to the topic at hand. The
reader is also given closure in the conclusion and comment stage, where the issue
of Lloyd’s survival is addressed and verbally confirmed through ‘If the plan
cannot solve the ……Lloyd’s may not survive’.

Thus, the sketch caricature basically presents the reader with the thesis that the
two financial executives, David Rowland and David Middleton, have a difficult
situation in attempting to deal with Lloyd’s current problems. They have a
difficult road ahead of them too - they may solve or alleviate these current
difficulties, but they may have to deal with greater, even insurmountable problems
in the future, problems which could be fatal to them and to Lloyd’s. These
messages, carried visually by the sketch caricature, are repeated in the attraction,
thesis statement, orientation and conclusion stages of the schematic structure of
the Mountains text, suggesting that the function of sketch caricatures in the
Finance department of The Economist magazine may be to co-operate visually
with the beginning and concluding stages of the text. Further research could
explore the proposition that the visual-verbal modes in a corpus of Finance
department texts operate in concert to present to the reader intersemiotically
coherent texts in terms of how each one deals with its topic, thesis, point of view
and conclusions.

The intersemiotic complementarity between the line graphs and the schematic
structure of the Mountains text may be explained by referring to the lexical
inventories generated for each of their VMEs. The line graphs deal with VMEs
which are specific issues within the general topic of Lloyd’s problems; these are
the loss of names, the increase in open year syndicates, and how both have
behaved over time. What is clear is that in each of the VME inventories concerned
with names, syndicates, time, and number there is a consistent use of semantically
related lexical items throughout reporting the background, reporting the current

316
situation, and problematising the situation stages. In all four VME inventories
there is only one mention of these issues in the attraction, thesis statement, and
orientation discourse stages, and that is a synonym for names, “capital providers”.
The reason for this pattern in the inventories can be traced to the topics of the data
which the VMEs represent, and because mathematical visuals, by their nature,
deal with specific data or relationships, and in this text these specific topics are
discussed in relation to the wider issue, which is Lloyd’s problems and future.

At the discourse level therefore, there is evidence for suggesting that naturalistic
visuals like sketch caricatures tend to visually represent generalised situations and
issues, and as a result have a stronger intersemiotic, discoursal relationship to the
stages of a text where the aim is to grab the reader’s attention and to orient him or
her to the general or overall point of view espoused, while the mathematical
visuals like line graphs tend to deal with more specific data and therefore are more
strongly related to the stages of the discourse which discuss and analyse the
specific issues they visually represent (this does not imply that the relations are
fixed in an either/or dichotomy, but rather refers to a tendency in terms of the
strength of the relation — obviously this is a matter of degree, as both visuals in a
text like the Mountains text can potentially relate to all the discourse stages;
further research is necessary to confirm this). This general-specific perception
accords with the statements made by Ms. Penny Garrett regarding the graphic
design team’s views that images’ function is to attract, and the mathematical line
graphs’ function is to provide informational support. These points may be
summarised in Table 8.2.

Visual Type Intersemiotically Related Discourse Stages

• Attracting the reader
Naturalistic Visuals -
Sketch Caricature • Stating the thesis
(General focus) • Orienting the reader

• Reporting the background

Mathematical Visuals - • Reporting the current situation

Line graphs (Specific • Problematising the situation
focus)
• Concluding and commenting

317
 Table 8.2 Visual to verbal schematic structure intersemiotic complementarity

A further, potentially fruitful area of investigation also could be to examine the

ways that certain VME-related lexical items may tend to cluster at various stages
of a multimodal text as its schematic structure unfolds — an analysis could reveal
how the selection of various lexical items relates to the various stages of the
schematic structure of the text and how this relates to the any potential narratives
represented in the visuals.

8.1.3 Intra-visual Compositional Intersemiotic Complementarity in the

Mountains Text
The visual-verbal relationship within visuals is also an area in which intersemiotic
complementarity can be realised. Often, as is the case with the Mountains text,
visuals have included within the visual frame verbal support for the messages they
are attempting to project. Thus, in terms of the intra-visual compositional aspects
of the Mountains text, there are a number of important features which help to
facilitate the projection of the various visual meanings and relationships. For
example, vectors between the represented participants realise a similarity
compositional relationship. This can be seen within the sketch caricature, where
the two men pushing a large boulder up the mountain slope are related to each
other in visual terms by the vectors formed by the directions of their eyes, the
parallel placement of their arms and legs, and the parallelism produced by the
slope of the first stage of the mountainside, which is also parallel to the vectors
formed by the men’s body positioning. This is further reinforced by the fact that
the verbal labelling of the boulder shows the name ‘Lloyd’s’ skewed off the plane
parallel to the mountainside slope. Writing is normally horizontal to a plane, so
any skewing of this expectation gives a sense of movement, an effect heightened
by the left-side of the word being higher than the right-side, and is suggestive of a
forward, rolling movement. This suggested forward movement arises from the
culturally-based expectation by viewers for circular movement to normally be
clockwise, and that a left to right movement indicates forward movement (which
is also related to reading paths). Thus, the vectors formed by the men’s bodies in
combination with the upward left to right slope of the mountainside, the skewing
of the verbal label ‘Lloyd’s’ from the parallel plane, and the expectation for

318
clockwise movement, all co-operate compositionally to present an integrated and
unified visual message.

In the line graphs this visual-verbal, intra-visual compositional coherence is more

obvious and pronounced. The two line graphs are framed by a single shadowed
box/border, and individually highlighted by the use of shadowed boxes and clear,
linear white margins around each. As such they are both separated, but they are
also within the one outer visual frame, so they could be counted as being one part
of a single visual with two separate, distinct aspects. Framing is therefore used to
both simultaneously include and exclude, and since each of the graphs within the
outer visual frame occupy a little less than half of the available space within it,
neither is more visually salient than the other. This compositional similarity and
the intersemiotic complementarity it realises is further strengthened by the fact
that both the line graphs are designed to represent the same mathematical
relations. In other words, they are both mathematical line graphs showing the
derivative (rate of change) relationship between interacting variables, they both
display their data in the top left to hand quadrants of their axes, and they both
utilise the same “x” independent variable (time) and start the range of that variable
in the same year. This intra-visual complementarity is reinforced by the effect
produced by the fact that they both share the same verbal, graphic headline format
(reversed white on black), they share graphic headings that are consistent in font
size and typeface, and they both share the same footer indicating the source of the
data. Further, it seems that both the ‘y’ axes of the graphs are calibrated so that the
line graphs displayed roughly approximate each other in size and range, despite
the fact that the numbers involved are completely different — it could be
reasonably argued that these graphs were constructed in such a way as to
accentuate the differences in shape and direction of the two graphic lines, and to
complement the symmetry of the two graphic headings, a point which is discussed
in more depth below.

These kinds of intra-visual compositional features could be explored in more

depth in future work because they don’t seem to be specific to the Mountains text,
but are consistent with published and accepted guidelines in relation to the use of

319
headings in charts and graphs — they both use headings which relate semantically
to the import of the data they portray. As White explains,
The readers should never be forced to draw their own conclusions
from a chart: it is best to bring out the significance of the figures or
trends or conclusions or interrelationships - by means of a good
headline (not a label headline, but a sentence which says something)
…… it is much better to indicate the significance of the chart or
graph in a non-label title. (1982,178).

A brief analysis of these graphic headlines shows that they do relate to the data
they portray semantically and typographically. They relate semantically in that the
words ‘Slumping’ and ‘mounting’ could be seen as collocates of each other
because they both belong to a lexical set in which they could be reasonably
expected to occur or co-occur, that of vertical, up-down, rising-falling movement.
A check of a thesaurus confirms this, where the root form “slump” produced
synonyms such as ‘fall’, ‘drop’, ‘slip’, ‘give way’, ‘sag’, ‘tumble’, ‘dip’, ‘plunge’,
‘decline’;’ fall off’, and antonyms such as ‘rise’, ‘upsurge’, and ‘upturn’. The root
form “mount” produced synonyms such as ‘ascend’, ‘climb’, ‘climb up’, ‘scale’,
‘get over’, ‘climb over’, ‘rise’, ‘go up’, ‘increase’, ‘grow’, ‘soar’, ‘swell’, ‘surge’,
and antonyms such as ‘drop’, ‘descend’, ‘fall’, ‘decline’, ‘lessen’, ‘decrease’, and
‘diminish’ (Random House 1992).

Both the line graphs also connect their two headlines typographically. Despite the
fact that the two headlines are divided by the enhanced framing around their
graphs, they are skewered together in meaning by a single headline broken by
leaders (a series of dots). This has the effect of unifying the message-focus of the
two graphs, with the aim to eliminate as much work as possible for the reader. The
use of the skewering effect by the leaders is a device which can produce a sense of
expectation in the reader — as he or she reads from left to right, the leader “leads
the reader on” to the next piece of information, an effect akin to someone pausing
for effect in a speech.

A further device used to show that the two line graphs are compositionally related
to each other is the use of the verbal statement that the source of the data is from
‘Lloyd’s of London’. Its placement in the lower status, bottom left-hand corner of

320
the outer visual frame agrees with White’s (1982:179) contention that the source
of the raw data should be placed in an inconspicuous place; however it also serves
the function of indicating to the viewer that all that is presented within this outer
frame refers to and supports the information contained in the graphs. This
inconspicuous placement of the source is supported by the relative visual salience
of the other verbal instances within the inner visual frames; these visual VMEs
provided by verbal means can be ranked in order of importance for the viewers
according to font size and therefore salience as firstly the graphic headlines, then
the two graphic heading labels, and finally the source of the data.

8.2 Educational Applications of this Study

This linguistic study has attempted to test the claim of the interrelatedness of
systems of meaning, and to assess the applicability of the SFL theory to
multimodality in a page-based text. This kind of study is useful in extending
theory through its application to real text, but it may also be useful in informing
educational practice in various contexts, as in:
• many school subjects such as Geography, Economics, the general Sciences,
History, Domestic Sciences, etc. These utilise a multiplicity of different kinds
of visuals (diagrams, photographs, illustrations, schematic drawings, tables,
charts and graphs etc.) in combination with the verbal text to present, elaborate
or explicate core concepts, and as pointed out before, to evaluate students’
knowledge and understanding.

• school readers in ESL or EFL contexts where the stories in graded readers and
books for English reading classes are often enhanced by visuals. For example,
almost every second page in the readers used in Japanese high schools have
some kind of visual which depicts a character or aspect of the unfolding story.
Teachers generally concentrate exclusively on getting the students to cope with
the verbal text, and rely on accompanying translations to facilitate that task.
The visuals are typically ignored or treated as ‘fillers’ by the teachers. An
appreciation of the fact that the visuals which co-occur with the verbal in these
stories are often an expression of similar or correlated meanings could, for
example, provide teachers with an extra resource to help the students
understand the plot (since most visuals in these books, based on this writer’s
experience, do seem to mirror the stages of the unfolding plot).

• courses in media studies dealing with newspapers, magazines and desktop

publishing, which at all levels have verbal/visual presentations. These courses
often focus on the content or subject matter that the various print or electronic
media project, and tend towards the analysis of the verbal or the visual mode in
isolation — a focus on the ways that this content is projected intersemiotically

321
 may allow students to see differences in the communicative power of
multimodal meaning as opposed to single-mode presentations of meaning, and
to conduct multimodal critical and stylistic analyses.

• industrial and work-related environments which rely heavily on co-occurring

visual and verbal documents such as forms to fill in, safety explanations and
requirements, and instructions for assembly and usage. An appreciation of the
fact that visual means of presenting meaning do work in concert with the verbal
expressions would enable educators in these fields to draw upon an extra
dimension (the intersemiotic) to present and especially evaluate the efficacy of
procedural explanations; an understanding of the areas where the visual
complements the verbal would be especially important in areas of public and
work-related safety, perhaps leading to more efficient presentations.

In the area of second language education, there also seems to be a need for a
greater appreciation of multimodal meaning and a concomitant need for teaching
methodologies to take it into account. An examination of some prominent
examples of textbooks attempting to deal with other modes beside the verbal (the
visual mainly) provides some insight into the reasons for this perceived need. The
textbook by Maley, Duff, and Grellet (1980) is a typical example. This consists of
a range of different kinds of images and drawings designed to stimulate
speculative discussion in the language classroom, with the rationale that it will
stimulate creativity in language use. The images are a collection of photos and
images which are to be used to stimulate students’ thinking and to therefore
encourage them to talk more. Any accompanying verbal text, as the following
extract shows, usually takes the form of instructions to the student to either work
alone or in a group, and a series of statements about the accompanying visual
which aim to involve the student with the visual in some way:

(A b/w picture of what looks like a flying saucer on some steps).

Work on your own at first. Then compare what you’ve done with
your partner.
1. You are present, an element of danger is involved. What do you
do?
2. You were present. Something happened. You had to send a
telegram. Write this telegram quickly ……. (op.cit:48-49).

In this example the visual works only as a stimulus to generate verbal language. In
other exercises the students are asked to answer specific questions about the
visual, which are then used to generate written exercises or discussion. This book

322
can be very useful for developing the students’ visual interpretation skills, but it
could perhaps be made more effective if the questions given below were to be
accompanied by the bracketed additions:
(A b/w picture of a young man and woman).
Work on your own first. Then compare what you’ve done with your
partner(s).
1. Where does the scene take place? [what tells you it is certain
place?]. In an office? At home? In a school? [what tells you it is
one of these places and not the other - what is there or not
there?]
2. Why is she yawning? [what is happening, what are they both
doing?] etc. (op.cit:20)

These could help them understand the how and why of their interpretation by
focussing on who or what is represented and the processes that can be seen in the
visual frame. The skills broken down and developed here could then be extended
by the addition of texts in which the visuals are accompanied by some verbal text
which relates or refers to them in some way. This text for example could be an
image of a museum piece which has an accompanying identifying, descriptive or
explicative verbal text. Questions could then be developed around what both the
modes are presenting and where there are similarities or differences in content or
attitude. The effect of this could be to train students to treat co-occurring modes as
complementary sources of meaning, rather than as isolates.

Another textbook example is that by Wright (1989), a work which contains a

range of ideas and activities showing how to use pictures in the language learning
classroom. However, this book is typical in that it doesn't base its treatment on a
view or theory of communication which informs its treatment of the visual mode
— it treats visuals in isolation, divorced from their co-occurrence with the verbal
mode. Its treatment of the visuals in this way is to a degree effective, and it is
effective if the visuals used commonly occur in isolation, but as this study has
shown this really only treats half the story, since visuals very commonly occur in
combination with some written text, and they occur in a variety of wider contexts
from which additional meaning can be extracted.

323
This publication by Wright is part of a series of books in applied linguistics which
are aimed at TESOL teachers, and like most books in this area, it focusses mainly
on using the visual only so far as to supplement the learning of verbal language. In
his introduction Wright states that visuals can be used “as a reference and stimulus
in order to promote five very different language teaching emphases” (op.cit:4-5),
which he lists, with examples, as being structures (e.g. teaching tenses),
vocabulary (e.g. topic-based words), functions (e.g. making a polite request),
situations (e.g. a street scene), and skills (e.g. listening, reading etc.). Using
pictures in this way is not necessarily inefficient considering the already-
mentioned dominance of the verbal mode in English cultures. However, this
supplementary view of the visual mode often leads teachers to ignore the
importance of the visual as a complete and coherent source of meaning in its own
right (Kress and van Leeuwen’s argument), that commonly works with the verbal
when it co-occurs on the page. This state of affairs can be traced to the fact that
historically, research in applied linguistics has not yet developed the tools to deal
with the ways that other semiotic codes project their meanings, especially when
they co-occur and are obviously an important part of a particular multimodal text.

As this study has shown this situation can be addressed by the usage of a theory of
communication which can account for the intersemiotic relations which most
obviously arise when visual and verbal co-occur in one text, and provides a
common terminology which can work to reduce confusion. Here the SFL model
and its concept of metafunctions can perhaps act as a unifying point.

8.2.1 A Suggested Methodology for Using Visuals in Educational Contexts

Since every image can be addressed in terms of what it presents, who it is
presenting to, and how it is presenting, the concept of the ideational, the
interpersonal, and the compositional metafunctions can act as pedagogical
resources which can allow teachers to work out ways to get their students to
extract just what the visuals are trying to ‘say’. This can then be the basis for a
range of other techniques which can concentrate on various verbal language skills.

324
To do this, teachers could use metafunctional concepts and insights to organise
activities which involve the students asking questions of the visuals, and then
interpreting them for developing various language skills. The richest source of
information can obviously be derived from those questions which focus on the
ideational aspects of a visual, and since many subjects are concerned with
information, its arrangement, and its relationship to other information, most
classroom activities could be centred around extracting just what the visuals are
trying to say to the viewers in terms of their ideational, informational content. The
VMEs derived from these questions could then become the focus of speaking,
writing and reading activities (and indirectly listening). The kinds of questions
that could be asked can be seen in Table 8.3.

Metafunction Questions
• Who, What, Where, Why, How?
• Who or what is involved in any activity? (Identification of
participants).
• What action is taking place, events, states, types of behaviour?
(Activity & Processes).
Ideational • Where [locative], who with [accompaniment], and by what means are
the activities being carried out [means]? (Circumstances).
• What are the qualities and characteristics of the participants?
(Attributes).
• What is it asking of you? (Visual address)
• How does it make you feel about what you see? (Address/Attitudinal
projection)
Interpersonal • Do you believe this is a real scene? Is it possible or probable? (Visual
modality)
• If there are other visuals on the page, are there any similarities or
differences between them? (Visual synonymy).
Compositional • What relationship do they have to other visuals in the reading or on the
same page? (Visual synonymy).
Table 8.3 Asking metafunctional questions of a visual

A range of learning/teaching activities can be developed based on the

information obtained through asking a visual questions. Some of these activities,
along with potential benefits for students, involve developing their reading
readiness, reading comprehension, their writing skills, and their speaking and
listening skills.

In terms of reading readiness, asking questions can activate the students’

background knowledge (content schema) of the text’s topic area, their knowledge
of it through personal experience, opinion, and hearsay. This can be allied to

325
developing strategies for dealing with the reduction of ‘text-shock’, by allowing
the students to ease themselves into a reading to get some idea of what to expect
in terms of who is doing what to whom, why and how. This will allow for an
expectancy to be set up in the students’ minds — the process of reading the text
will then either give them a confirmation of their interpretation of the story, or in
rare cases a different story to what they expect from looking at the visual. A
further area is vocabulary development; the interpretation of a visual will
necessarily involve encounters with new words, which the students can
immediately associate with a visual representation. This allows for a cognitive
association to be set up, facilitating vocabulary learning. Reading skills such as
skimming and sight-recognition of vocabulary development can also be
engendered. By skimming through a reading and identifying words that relate to
the visual in terms of whether they be participants, processes and circumstances
can also help students with pre-reading vocabulary development — in terms of an
activity, the students individually or in groups could start with the visual, extract
the main words or phrases that the visuals generate in their minds, find their
meanings, and then find words/synonyms in the reading.

In terms of activities for reading comprehension, students’ understanding of the

plot of a story could be enhanced. If there is a sequence of visuals in a short story,
as is common in many graded readers and abridged versions of novels used in
schools, the students could be asked to start their reading of the novel by looking
at the visuals only in their story sequence, and then interpret them before they start
to read the actual story. Also, the sequence of visuals in a short story could be
extracted, reproduced and jumbled — the students could then be asked to put them
in an order that they think is appropriate. If this approximates the story, well and
good — if not, it can set up discussion of why or why not. Then they could
actually read the story. This kind of activity can work to clue the students into the
fact that visuals are often used as an entree into a reading (as an attract), and that
this can work well in understanding a plot, since visuals in novels often mirror the
stages of the story, and usually occur on or opposite a page which is dealing with
the same point in the story.

326
In relation to the students’ writing development, especially in the area of narrative
writing, a sequence of pictures could be extracted from a reading that the students
are required to read, and using these decontextualised images could construct their
own story individually or in groups, then write the story in class or as a journal or
as the basis for a class story magazine. This story-writing or class magazine
production could be used as a writing process activity where drafting and
redrafting is carried out in consultation with teachers, or in peer-editing groups.
This kind of activity is relevant to developing in students an understanding of
story/narrative structures: the image sequences and the writing activities which
arise from them could be used to introduce the students to various genres
(narrative, description), based on the ways that visuals are organised. The
sequence of this kind of classroom activity could be to start with visuals which
can readily tell a story, either one story only or a number of possible plots. Then
the teacher could require the students to ask the visuals questions to extract who
are the main characters, what are they doing and with whom, why are they doing
it, and how, etc. Then the students could try to place the pictures in some order,
which could be organised into their own spoken (oral skills development) and then
into a written story.

The students’ speaking and listening skills could also be developed through all the
above activities, which provide ample opportunities for students to speak to the
teacher and to peers. The reading readiness activities for example could also be
used for reporting back to the class, for giving short speeches, explaining,
describing etc., and the students of course have to listen to these student
presentations, so the development of listening skills is another positive result.
Images can also be used for evaluating speaking skills — this could involve
showing a student a picture from a story already read and asking him or her to talk
about it. This will test both production and understanding of the reading’s content,
and could also be used in a class-wide evaluation to see which group understands
a story’s content and sequence best, and if required, allotting various grades.

8.3 Final Comments

327
In this chapter attention has been drawn to areas where the propositions explored
in this study can be further confirmed through more in-depth application of the
analyses, as well as to areas of further research which can highlight other ways
that multimodal texts like the Mountains text can realise intersemiotic
complementarity, such as at the level of the text’s schematic structure and intra-
visually. It has also suggested that the analytical framework developed for the
explication of the Mountains text has possible application to pedagogical contexts.

The focus of this study has been on using the SFL model for the analysis of
multimodal text, and as such forms a continuation and extension of the interest
shown by the many semioticians who have followed Saussure’s semiological
program in the analysis of non-linguistic signs (including the Prague School of the
1930’s and 1940’s, the Paris School of the 1960’s and 1970’s, and the more recent
Social Semiotics School). This study also represents an evaluation and extension
of the work of this latter school, specifically in regard to the ways that various
visuals organise and project their meanings in The Economist magazine, and to the
ways that these visuals may interact with verbal text. Finally, this study has tested
the claim made by Halliday and Hasan (1985:4) of the inter-relatedness of systems
of meaning, showing that the interfaces between the visual and the verbal semiotic
systems in a multimodal text are in fact interrelated and work together to produce
a coherent multimodal text. As such this study has extended the application of
SFL theoretical principles by showing that it is an appropriate paradigm for
explaining many aspects of multimodality.

As pointed out in Chapter One, the motivation for this study lay in a question
arising from this writer’s involvement in various educational contexts relating to
the discipline of economics — this question related to a perception that whilst the
various forms of published economics discourse all utilise some form of visual
communication, the professional, academically-oriented literature clearly directs
the readers to refer to the various graphs, tables, charts and diagrams utilised,
while the eco-journalistic print media only rarely directs or asks the reader to
interact in some specific way with them. The questions explored in this study,
“what is the function of the visual mode when used in conjunction with the verbal

328
in economically-oriented journalistic print media, and what connections are
construed when these two page-based modes co-occur?” have been answered by
showing that a multimodal text like the Mountains text is a single textual
phenomenon characterised by intersemiotic complementarity, in which both
modes work together to project various page-based meanings. The analytical
framework presented in this study has shown that the connections construed in
this intersemiotic relationship can be elucidated through an investigation of the
intersemiotic ideational, interpersonal, or compositional metafunctional meanings.
It can therefore be concluded that the realisations of the visual and verbal semiotic
systems in this multimodal text do not merely co-occur on the page space and
relate to each other by virtue of simple combination; rather, they synergistically
combine on the page to produce a coherent multimodal textual phenomenon.

329
BIBLIOGRAPHY

Ades, D. (1986) Photomontage. London: Thames and Hudson.

Allen, P and R. Pholsward. (1988) “The development of a diagnostic reading test
for EAP students of economics”, in M. L. Tickoo (ed.) ESP: State of the art.
Anthology Series 21, SEAMO Regional Language Centre, Singapore, 58-
73.
Anuatti, F. (1991) "Workshop of the realist group on Keynes: Keyne's rational
theory of persuasion", paper given at The University Of Birmingham
International Seminar on The New Rhetoric and Discourse Analysis: the
case of economics, April 15-17.
Arnheim, R. (1969) Visual Thinking. Berkeley and Los Angeles: University of
California Press.
Arnheim, R.(1974) Art and Visual Perception. Berkeley and Los Angeles:
University of California Press.
Arnheim, R.(1988) The Power of the Centre: A Study of Composition in the Visual
Arts, Berkeley and Los Angeles: University of California Press.
Arnheim, R.(1992) To the Rescue of Art: Twenty-Six Essays. Berkeley and Los
Angeles: University of California Press.
Asher, R. E. editor-in-chief (1994) The Encyclopaedia of Language and
Linguistics. Oxford: Pergamon Press.
Backhouse, R. (1993) "The debate over Milton Friedman’s theoretical framework:
an economist's view", in W. Henderson, A. Dudley-Evans and R.
Backhouse. (eds.) Economics and Language. London: Routledge.
Bakhtin, M. M. (1986) Speech Genres and Other Late Essays. Austin: University
of Texas Press.
Barry, P. (1995) Beginning Theory: An introduction to literary and cultural
theory. Manchester: Manchester University Press.
Barthes, R. (1964/67) Elements of Semiology. London: Cape.
Barthes, R. (1977) Image, Music, Text. New York: The Noonday Press.
Bazerman, C. (1993) “Money talks: the rhetorical project of The Wealth of
Nations”, in W. Henderson, A. Dudley-Evans and R. Backhouse. (eds.)
Economics and Language. London: Routledge.
Becher, T. (1981) "Towards a definition of disciplinary cultures", Studies in
Higher Education, Vol.6:2, 109-122.
Becher, T. (1987) "Disciplinary discourse", Studies in Higher Education,
Vol.12:3, 261-274.
Benson, J. and W. Greaves, (1980) "Field of discourse: theory and application",
Applied Linguistics, Vol.11:1, 45-55.
Berger, J. (1972) Ways of Seeing. London: Penguin Books.
Biedermann, H. (1992) Dictionary of Symbolism: Cultural icons and the meanings
behind them. New York: Facts on File, Inc.
Bloor, M. and T. Bloor. (1993) “How economists modify propositions”, in W.
Henderson, T. Dudley-Evans and R. Backhouse. (eds.) Economics and
Language. London: Routledge.
Bloor, T. and M. Pindi. (1990) “Schematic structure in economics forecasts”, in A.
Dudley-Evans and W. Henderson (eds.) The Language of Economics: The
analysis of economics discourse. ELT Documents 134, London: Modern
English Publications and The British Council, 55-66.

330
Bowcher, W. (1994) “A brief overview of systemic functional linguistics and an
analysis of text”. Obirin Review of International Studies, No.6, 154-177.
Bramki, D. and R. Williams. (1984) “Lexical familiarisation in economics text,
and its pedagogical implications in reading comprehension”, Reading in a
Foreign Language, Vol.2:1, 169-181.
Brett, M. (1991) How to Read the Financial Pages: A simple guide to the way
money works and the jargon 3rd Ed. London: Century Business.
Brown, V. (1993) “Decanonising discourses: textual analysis and the history of
economic thought”, in W. Henderson, A. Dudley-Evans and R. Backhouse.
(eds.) Economics and Language. London: Routledge.
Brown, G. and Yule, G. (1983) Discourse Analysis. New York: Cambridge
University Press.
Butt, D., Fahey, R., Spinks, S. and Yallop, C. (1995) Using Functional Grammar:
An explorer’s guide. Sydney: NCELTR, Macquarie University.
Button, K. (1988) “The writing of economics: a review article”. Economics,
Spring, 25-27.
Caldwell, B. and A Coats. (1984) “Communications: the rhetoric of economics: a
comment on McCloskey”, Journal of Economic Literature, Vol.22, June,
575-578.
Cameron, R. (1991) "Economics and ecology - who is borrowing from whom?",
paper given at The University Of Birmingham International Seminar on The
New Rhetoric and Discourse Analysis: the case of economics, April 15-17.
“Can Your Graphics Board do all this for $369?” (1995) PC Magazine. May 16th,
265.
Cirlot, J. E. (1971) A Dictionary of Symbols. 2nd Ed. London: Routledge.
Clarke, W. M. (1995) How the City of London Works. 4th Ed. London: Sweet and
Maxwell.
Cloran, C. (1994) Rhetorical Units and Decontextualisation: An enquiry into some
relations of context, meaning and grammar. Monographs in Systemic
Linguistics, No. 6. Nottingham: Department of English Studies, University
of Nottingham.
Culyer, A. (1985) "What's wrong with Economics textbooks?", Economics,
Spring, 15-17.
Danesi, M. (1994) Messages and Meanings: An introduction to semiotics.
Toronto: Canadian Scholar’s Press Inc.
Danesi, M. and D. Santeramo. (1992) Introducing Semiotics: An anthology of
readings. Toronto: Canadian Scholar’s Press Inc.
Darnell, A. (1991) "Statistical testing and econometrics: the rhetoric of
insignificance", paper given at The University Of Birmingham International
Seminar on The New Rhetoric and Discourse Analysis: the case of
economics, April 15-17.
De Escorcia, B. (1984) “Team teaching for students of economics: a Colombian
experience”, in R. Williams, J. Swales and J. Kirkham (eds.) Common
Ground: Shared interests in ESP and communication studies. ELT
Document Special, Oxford: Pergamon Press, 135-143.
van Dijk, T. A. (1977) Text and Context. London: Longman.
Dondis, (1973) A Primer of Visual Literacy. Cambridge, Massachusetts: The MIT
Press.

331
Downing, A. and P. Locke (1992) A University Course in English Grammar.
London: Prentice-Hall International.
Dudley-Evans, A. (1987) “Genre analysis and ESP”, ELR Journal, Vol.1, The
University of Birmingham, 1-9.
Dudley-Evans, A. (1993) “The debate over Milton Friedman’s theoretical
framework: an applied linguist's view”, in W. Henderson, A. Dudley-Evans
and R. Backhouse. (eds.) Economics and Language. London: Routledge.
Dudley-Evans, A. and W. Henderson. (1990a) The Language of Economics: The
analysis of economics discourse. ELT Documents 134, London: Modern
English Publications and The British Council.
Dudley-Evans, A. and W. Henderson. (1990b) “The organisation of article
introductions: evidence of change in economics writing”, in A. Dudley-
Evans and W. Henderson (eds.) The Language of Economics: The analysis
of economics discourse. ELT Documents 134, London: Modern English
Publications and The British Council, 67-78.
Dyer, G. (1982) Advertising as Communication. London: Methuen.
Eco, U. (1976) A Theory of Semiotics. Bloomington: Indiana University Press
Eggins, S. (1994) An Introduction to Systemic Functional Linguistics. London:
Pinter Publishers.
Ezell, M. and K. O’Keefe, eds. (1994) Cultural Artifacts and the Production of
Meaning: The page, the image, and the body. Ann Arbor: The University of
Michigan Press.
Favretti, R. (1991) "Epistemology and persuasion in economic discourse", paper
given at The University Of Birmingham International Seminar on The New
Rhetoric and Discourse Analysis: The case of economics, April 15-17.
Fisher, J. (1987) Pedagogical Implications of Thematic Analysis of Economics
Texts. Unpublished MA dissertation, University of Birmingham.
Fisher, J. (1990) “Factors influencing the design of syllabi and support materials
for non-native speakers studying economics”, in A. Dudley-Evans and W.
Henderson (eds.) The Language of Economics: The analysis of economics
discourse. ELT Documents 134, London: Modern English Publications and
The British Council, 84-93.
Fiske, J. (1982) Introduction to Communication Studies. 2nd Ed. London:
Routledge.
Garrett, P. (1994) Interview with Chief Editor of Graphic Design - The Economist
Magazine, London: 17th August.
Gerrard, B.(1993) "The significance of interpretation in economics”, in W.
Henderson, A. Dudley-Evans and R. Backhouse. (eds.) Economics and
Language. London: Routledge.
Gombrich, E. (1960) Art and Illusion: A study of the psychology of pictorial
representation. London: Phaidon Press Ltd.
Gombrich, E. (1982) The Image and the Eye: Further studies in the psychology of
pictorial representation. London: Phaidon Press Ltd.
Gutwinski, W. (1976) Cohesion in Literary texts: a study of some grammatical
and lexical features of English discourse. The Hague: Mouton.
Hall, J. (1994) Illustrated Dictionary of Symbols in Eastern and Western Art.
London: John Murray.
Halliday, M.A.K. (1976) System and Function in Language. (ed. G. R. Kress)
London: Oxford University Press.

332
Halliday, M.A.K. (1978) Language as Social Semiotic. London: Edward Arnold.
Halliday, M.A.K. (1985) An Introduction to Functional Grammar. London:
Edward Arnold.
Halliday, M.A.K. (1987) “Language and the order of nature” in D. Attridge, A.
Durant et.al. (eds.), The linguistics of writing. Manchester: Manchester
University Press, 135-154.
Halliday, M.A.K. (1991) “The Notion of Context in Language Education” in Thao
Lê and M. McCausland (eds.), Language Education: interaction and
development. Proceedings of the International Conference, Vietnam, 30
March - 1 April 1991. Launceston, Tasmania: University of Tasmania, 1-26.
Halliday, M.A.K. (1992) “How do you mean?” in M. Davies and L. Ravelli (eds.),
Advances in Systemic Linguistics: Recent theory and practice. London and
New York: Pinter Publishers, 20-35.
Halliday, M.A.K. (1994) An Introduction to Functional Grammar. 2nd ed. London:
Edward Arnold.
Halliday, M.A.K. and R. Hasan. (1976) Cohesion in English. London: Longman.
Halliday, M.A.K. & R. Hasan, (1985) Language, Context and Text: Aspects of
language in a social-semiotic perspective. Oxford: Oxford University Press.
Hartley, J. (1985) Designing Instructional Text. 2nd ed. London: Kogan Page Ltd.
Hasan, R. (1981) “What’s Going On: A dynamic view of context in language” in
J. E. Copeland and P. W. Davis (eds.) The Seventh LACUS Forum,
Columbia SC: Hornbeam Press, 106-121.
Hasan, R. (1993) “Context for meaning” in James E. Alatis (ed.) Language,
Communication and Social Meaning: Georgetown University Round Table
on Languages and Linguistics 1992. Washington. D.C.: Georgetown
University Press, 79-103.
Hasan, R. (1995) “The conception of context in text” in P. H. Fries and M.
Gregory (eds.) Discourse in Society: Systemic Functional Perspectives.
Meaning and Choice in Language: Studies for Michael Halliday. Vol. L,
Norwood New Jersey: Ablex Publishing Company, 183-296.
Hasan, R. (1996) Ways of Saying, Ways of Meaning: Selected papers of Ruqaiya
Hasan. (eds. C. Cloran, D. Butt and G. Williams) London: Cassell.
Hasan, R. (in press) “Society, Language, and the Mind: The meta-dialogism of
Basil Bernstein’s theory”. in F. Christie (ed.) Pedagogy and the Shaping of
Consciousness: Linguistic and social processes. London: Cassell.
Henderson, W. (1986) "Metaphor in Economics", in M. Coulthard (ed.) Talking
About Text. Discourse Analysis Monograph No.13, University of
Birmingham: English Language Research, 109-127.
Henderson, W. (1993) "The problem of Edgeworth's style", in W. Henderson, A.
Dudley-Evans and R. Backhouse. (eds.) Economics and Language. London:
Routledge.
Henderson, W. and A. Dudley-Evans, eds. (1990) "Introduction: The analysis of
Economics discourse", in A. Dudley-Evans and W. Henderson, The
Language of Economics: The Analysis of Economics Discourse. ELT
Documents 134, London: Modern English Publications and The British
Council, 2-15.
Henderson, W. and Dudley-Evans, A. (1991) "The new rhetoric and discourse
analysis: the case of economics", opening session paper given at The

333
 University Of Birmingham International Seminar on The New Rhetoric and
Discourse Analysis: The case of economics. April 15-17.
Henderson, W. and A. Hewings. (1987a) Reading Economics: How text helps or
hinders. British National Bibliography Research Fund Report 28, London:
The British Library.
Henderson, W. and A. Hewings. (1987b) “The language of economics: the
problems of terminology”. Economics, Winter, 123-127.
Henderson, W. and A. Hewings. (1990) “A language of model building?”, in A.
Dudley-Evans and W. Henderson (eds.) The Language of Economics: The
analysis of economics discourse. ELT Documents 134, London: Modern
English Publications and The British Council, 43-54.
Henderson, W; A. Dudley-Evans and R. Backhouse. (eds.) (1993) Economics and
Language. Routledge, London.
Hewings, A. (1989) Reading for Content: Aspects of problems posed by
economics. Unpublished MPhil: University of Birmingham.
Hewings, A. (1990) “Aspects of the language of economics textbooks”, in A.
Dudley-Evans and W. Henderson (eds.) The Language of Economics: The
analysis of economics discourse. ELT Documents 134, London: Modern
English Publications and The British Council, 29-42.
Hewings, A. and W. Henderson. (1987) “A link between genre and schemata: a
case study of economics text”. ELR Journal, 1, University of Birmingham,
156-175.
Hodge, R. and G. Kress. (1988) Social Semiotics. Cambridge: Polity Press.
Hoey, M. (1983) On the Surface of Discourse. Boston: George Allen and Unwin.
Hollis, R. (1994) Graphic Design: A concise history. London: Thames and
Hudson.
Houghton, D. (1991) “Time perspective and tense choice in selected social cost
benefit analysis case studies”, paper given at The University of Birmingham
International Seminar on The New Rhetoric and Discourse Analysis: The
case of economics. April 15-17.
Houghton, D. and P. King (1990) “What it makes sense to ask: students' and
lecturer's questions in English for development economics”, in A. Dudley-
Evans and W. Henderson (eds.) The Language of Economics: The analysis
of economics discourse. ELT Documents 134, London: Modern English
Publications and The British Council, 95-114.
Jenks, C. ed. (1995) Visual Culture. London: Routledge.
Johns, A. (1980) “Cohesion in written business discourse: some contrasts”. The
ESP Journal, Vol.1:1, 37-43.
Johns, A. (1986) "The Language of Business", Annual Review of Applied
Linguistics, Vol.7, Cambridge University Press, 3-17.
Jordan, R. R. (1978) “Language practice materials for economists”. in A.
Mountford and R. Mackay English For Specific Purposes. Longman, 179-
189.
Jordan, R. R. (1984) “Motivation in ESP: a case study of methods and materials
for economics”. in G. James, The ESP Classroom -Methodology, materials,
expectations. Exeter Linguistic Series 7, University of Exeter, 82-88.
Jordan, R. R. (1990) “He said: quote....unquote”, in A. Dudley-Evans and W.
Henderson (eds.) The Language of Economics: The analysis of economics

334
 discourse. ELT Documents 134, London: Modern English Publications and
The British Council, 79-83.
Kress, G. and R. Hodge (1979) Language as Ideology. London: Routledge and
Kegan Paul.
Kress, G and T. van Leeuwen, (1990) Reading Images. Geelong, Victoria: Deakin
University Press.
Kress, G and T. van Leeuwen, (1996) Reading Images: The Grammar of Visual
Design. London: Routledge.
Larsen, S. E. (1994) “Semiotics”, in R. E Asher,. editor-in-chief, The
Encyclopaedia of Language and Linguistics. Oxford: Pergamon Press, 3821-
3832.
Laseau, P. (1989) Graphic Thinking for Architects and Designers. 2nd ed. New
York: Van Nostrand Reinhold.
Leeds-Hurwitz, W. (1993) Semiotics and Communication: Signs, codes, cultures.
New Jersey: Lawrence Erlbaum Associates.
Lemke, J. L. (1990) Talking Science: Language, Learning and Values. Norwood,
New Jersey: Ablex Publishing Corporation.
Lemke, J. (1995) “Intertextuality and Text Semantics”, in Peter Fries and Michael
Gregory, eds., Discourse in Society: Systemic functional perspectives. Vol.
L in the series Advances in Discourse, Norwood, New Jersey: Ablex
Publishing Corporation,. 85-114.
Luhman, D. (1998) Review of The Economist magazine. Numen Lumen
Homepage. http://www.numen-lumen.com/MgR/TheEconomist.html
Lutz, R. (1949) Graphic Presentation Simplified: A guide to the use of graphic
methods in business. New York: Funk and Wagnalls.
McCloskey, D. (1983) “The rhetoric of economics”. Journal of Economic
Literature Vol.21:481-517.
McCloskey, D. (1984) “Communications: reply to Caldwell and Coats”. Journal
of Economic Literature, Vol.22:579-580.
McCloskey, D. (1986) The Rhetoric of Economics. Brighton: Wheatsheaf Books.
McCloskey, D. (1990) If You're So Smart: The narrative of economic expertise.
Chicago: The University of Chicago Press.
McConnell, C.R. (1982) "Readability formulas as applied to college economics
textbooks", Journal of Reading, Vol.26:1 October, 14-17.
McLuhan, M. (1964) Understanding Media: The extensions of man. New York:
Signet Books.
Maki, U. (1993) "Two philosophies of the rhetoric of economics”, in W.
Henderson, A. Dudley-Evans and R. Backhouse. (eds.) Economics and
Language. London: Routledge.
Maley, A., A. Duff and F. Grellet (1980) The Mind’s Eye: Using pictures
creatively in language learning. Cambridge: Cambridge University Press.
Malinowski, B. (1923) “The problem of meaning in primitive languages”.
Supplement 1 to Ogden, C. and Richards, I. The Meaning of Meaning.
London: Kegan Paul, 296-336.
Malinowski, B. (1935) Coral Gardens and their Magic, Vol.2, London: Allen and
Unwin. Reprinted as The Language of Magic and Gardening (Indiana
University Studies in the History and Theory of Linguistics), Bloomington:
Indiana University Press 1967.

335
Marchese, F. T. ed. (1995) Understanding Images: Finding meaning in digital
imagery. Santa Clara, California: TELOS - The Electronic Library of
Science, Springer-Verlag Publishers.
Martin, J., M. Matthiessen and C. Painter, (1997) Working with Functional
Grammar. London: Arnold.
Mason, M. (1990) “Dancing on air: analysis of a passage from an economics
textbook”, in A. Dudley-Evans and W. Henderson (eds.) The Language of
Economics: The analysis of economics discourse. ELT Documents 134,
London: Modern English Publications and The British Council, 16-28.
Mason, M. (1991) “How language makes economics possible”, paper given at The
University of Birmingham International Seminar on The New Rhetoric and
Discourse Analysis: The case of economics. April 15-17.
Matthiessen C. (1993) “Register in the Round: diversity in a unified theory of
register analysis”, in M. Ghadessy (ed.) Register Analysis: Theory and
practice. London and New York: Pinter Publishers, 221-292.
Matthiessen C. (1995) Lexicogrammatical Cartography: English Systems. Tokyo:
International Language Science Publishers.
Mauranen, A. (1993) “Contrastive ESP rhetoric: metatext in Finnish-English
economics texts”. English for Specific Purposes Vol.12:1, 3-22.
Maybury, M. T. ed. (1993) Intelligent Multimedia Interfaces. London: AAAI
Press/The MIT Press.
Mead, R. and A. D. Lilley (1975) “The use of visual materials in teaching English
to economics students”. English Language Teaching Journal, Vol.29:2, 151-
156.
Mead, R. and W. Henderson (1983) “Conditional form and meaning in economics
text”. The ESP Journal, Vol.2:2, 139-160.
Mercer, P. (1998 in press) "Reconceptualising Television Realism: a framework
for analysis", in D’Haen, T. and C. Barfoot (eds.) Language and Beyond.
TEXTET, Amsterdam, The Netherlands: Editions Rodopi B.V.
Mirowski, P. (1991) "The when, the how and the why of mathematical expression
in the history of economic analysis", Journal of Economic Perspectives,
Vol.5:1, Winter, 145-157.
Montague, J. (1993) Basic Perspective Drawing: A visual approach. 2nd Edition,
New York: Van Nostrand Reinhold.
Morgan, T. (1988) "Theory versus empiricism in academic economics: update and
comparisons", Journal of Economic Perspectives, Vol.2:4, Fall, 159-164.
Mparutsa, C. (1991) “Perceptions of the market: a discourse analysis of oral
reports by Zimbabwean economics students”, paper given at The University
of Birmingham International Seminar on The New Rhetoric and Discourse
Analysis: The case of economics. April 15-17.
Mparutsa, C., A. Love and A. Morrison (1991) “Bringing concord to the ESL
classroom”. ELR Journal, Vol.4, The University of Birmingham, 115-134.
Myers, G. (1989) “The pragmatics of politeness in scientific articles”, Applied
Linguistics, Vol.10:1-35.
Myers, G. (1990) Writing Biology: Texts in the social construction of scientific
knowledge. Madison: University of Wisconsin Press.
Myers, G. (1991) “Getting on the map: a story of an economics proposal”, paper
given at The University of Birmingham International Seminar on The New
Rhetoric and Discourse Analysis: The case of economics. April 15-17.

336
Nöth, W. (1995) Handbook of Semiotics. Bloomington: Indiana University Press.
Ong, W. (1982) Orality and Literacy: the technologizing of the word. (New
Accents Series) London: Methuen.
O'Toole, M. (1994) The Language of Displayed Art. London: Leicester University
Press.
O'Toole, M. (1995) “A Systemic-Functional Semiotics of Art”, in Peter Fries and
Michael Gregory, Discourse in Society: Systemic functional perspectives.
Vol. L in the series Advances in Discourse. Norwood, New Jersey: Ablex
Publishing Corporation, 159-179.
Panofsky, E. (1982) Meaning in the Visual Arts. Chicago: University of Chicago
Press.
Papps, I. and W. Henderson (1977) Models in Economic Theory. Philadelphia:
W.B. Saunders.
Pearce, D. (1991) “The rhetoric of economics ”, a review of McCloskey, D. If
You’re So Smart: The narrative of economics expertise, in The Times
Literary Supplement (TLS), April 12th, 25.
Pearce, D.W. (1992) Macmillan Dictionary of Modern Economics. 4th ed.,
London: Macmillan.
Pindi, M. (1988) Schematic Structure and the Modulation of Propositions in
Economics Forecasting Texts. Unpublished Ph.D. thesis. Language Studies
Unit, Aston University.
Pindi, M. and T. Bloor (1987) “Playing safe with predictions: hedging, attribution
and conditions in economic forecasting”. in T. Bloor, and J. Norrish (eds.)
Written Language. CILT, 55-69.
Porter, T. and B. Greenstreet (1980) Manual of Graphic Techniques 1: For
architects, graphic designers and artists. London: Butterworth Architecture.
Random House, (1992) Random House Webster's Electronic Dictionary and
Thesaurus. Version 1.0, Reference Software International.
Rotheim, R. (1988) "Keynes and the language of probability and uncertainty",
Journal of Post-Keynesian Economics, Vol.11:1, Fall, 82-99.
Royce, T. (1984) The Language of Economics: A language development program.
Unpublished MA in Applied Linguistics (TESOL) thesis. University of
Sydney, Australia.
Royce, T. (1992) “Economics discourse and economists: a working paper on
recent discussions”. ICU Language Research Bulletin, Vol.7:43-56.
Royce, T. (1993) “Economics language and economics content: an overview”. in
N. Bird, J. Harris, and M. Ingham (eds.) Language and Content. Institute of
Language in Education, Hong Kong Education Department, 288-303.
Royce, T. (1994) “Reading economics in English: aspects of course development
in a Japanese university”. RELC Journal, Vol.25:2, 26-55.
Royce, T. (1995) “The analysis of Economics discourse: a general review”.
Australian Review of Applied Linguistics. Vol.18. No. 2. 137-159.
Royce, T. (1998a) "A Metafunctional View of Intersemiosis in The Economist
Magazine: A framework for analysis", in D’Haen, T. and C. Barfoot (eds.)
Language and Beyond. TEXTET, Amsterdam, The Netherlands: Editions
Rodopi B.V.
Royce, T. (1998b) "Synergy on the Page: Exploring intersemiotic
complementarity in page-based multimodal text.", in JASFL Occasional

337
 Papers No.1. Tokyo: Japan Association of Systemic Functional Linguistics
(JASFL).
Ruskin, J. (1991) The Elements of Drawing. (Illus. Ed.) London: The Herbert
Press.
Saint-Martin, F. (1987) Semiotics of Visual Language. Bloomington: Indiana
University Press.
Saussure, Ferdinand de. (1916/1966) Course in General Linguistics. New York:
McGraw-Hill Paperbacks.
Schapiro, M. (1973) Words and Pictures: On the literal and the symbolic in the
interpretation of a text. The Hague: Mouton.
Schleppegrell, M. (1985) “Economic input: an ESP program”. The ESP Journal,
Vol.4:111-119.
Sebeok, T. A. (1972) Perspectives in Zoosemiotics. The Hague: Mouton.
Sebeok, T. A. (1989) The Sign and its Masters. Lanham: University Press of
America.
Sebeok, T. A. (1994) An Introduction to Semiotics. London: Pinter Publishers.
Sherman, C. R. (1995) Imaging Aristotle: Verbal and visual representation in
fourteenth century France. Los Angeles: University of California Press.
Sinclair, J. (1986) "Fictional worlds", in M. Coulthard (ed.), Talking About Text,
Discourse Monograph 13, English Language Research, University of
Birmingham, 43-60.
Smith, V.L. (1989) "Theory, experiment and economics", Journal of Economic
Perspectives, Vol.3:1, Winter, 151-169.
Spencer, A. (1975/6) “Semantic combinations in economics and law - a case for
special treatment”. ESPMENA Bulletin, No. 3, 30-33.
Stubbs, M. (1983) Discourse Analysis. Chicago: University of Chicago Press.
Swales, J. (1981) Aspects of Article Introductions. Aston ESP Monographs No.1,
Aston University.
Swales, J. (1990) Genre Analysis. Cambridge: Cambridge University Press.
Tadros, A. (1977) "Principles underlying materials preparation for students of
economics and social sciences", ESPMENA Bulletin, No.8, pp31-34.
Tadros, A. (1979) "Linguistic prediction in economics text", ESPMENA Bulletin,
No. 14, 11-16.
Tadros, A.(1981) Linguistic Prediction in Economics Text, Unpublished doctoral
thesis, University of Birmingham.
Tadros, A. (1985) Prediction in Text. Discourse Analysis Monograph No.10,
University of Birmingham: English Language Research.
Tadros, A. (1989) "Predictive categories in university textbooks", English for
Specific Purposes, Vol.8:17-31.
The Economist Magazine. (1990) The Economist Style Book. London: The
Economist Books Pty. Ltd.
The Economist Magazine. (1991) Numbers Guide: The essentials of business
numeracy. London: The Economist Books Pty. Ltd.
Thibault, P. (1997) Re-Reading Saussure: The dynamics of signs in social life.
London: Routledge.
Thomas, L. (1991) "Rhetoric/discourse analysis and economics education", paper
given at The University Of Birmingham International Seminar on The New
Rhetoric and Discourse Analysis: The case of economics, April 15-17.
Thompson, G. (1996) Introducing Functional Grammar. London: Arnold

338
Tobin, Y. (1990) Semiotics and Linguistics. London : Longman.
Tufte, E. R. (1983) The Visual Display of Quantitative Information. Cheshire,
Connecticut: Graphics Press.
Uspensky, B. (1973) A Poetics of Composition. Berkeley and Los Angeles:
University of California Press.
Ventola, E. (1995) “Generic and register qualities of texts and their realisation”, in
P.H. Fries and M. Gregory (eds.) Discourse in Society: Systemic Functional
Perspectives, Meaning and Choice in Language: Studies for Michael
Halliday, Vol.50 in the series Advances in Discourse Processes, Norwood
N.J.: Ablex, pp3-28.
Vestergaard, T. and K. Schrøder (1985) The Language of Advertising. Oxford:
Blackwell.
Vygotsky, L. S. (1986) Thought and Language. Massachusetts: The M.I. T. Press.
Watson, J. and A. Hill. (1997) A Dictionary of Communication and Media Studies.
4th ed., London: Arnold.
Weisman, D. L. (1968) Language and Visual Form: The personal record of a dual
creative process. Austin & London: University of Texas Press.
nd
White, Jan V. (1982a) Designing for Magazines. 2 ed., New York: R.R. Bowker
Company.
nd
White, Jan V. (1982b) Editing by Design. 2 ed., New York: R.R. Bowker
Company.
White, Jan V. (1983) Mastering Graphics. New York: R.R. Bowker Company.
White, Jan V. (1984) Using Charts and Graphs. New York: R.R. Bowker
Company.
White, Jan V. (1988) Graphic Design for the Electronic Age. New York: Watson-
Guptill Publications.
White, Jan V. (1990) Graphic Idea Notebook. Revised Ed. Rockport: Rockport
Publishers.
Winter, E.O. (1977) "A clause-relational approach to English texts: a study of
some predictive lexical items in written discourse", Instructional Science,
Vol.6:1-92.
Wright, A. (1989) Pictures for Language Learning. New York: Cambridge
University Press.
Zevi, B. (1993) Architecture as Space: How to look at architecture. New York:
Da Capo Press.

339