Appliable Linguistics (PDFDrive)

Appliable Linguistics
Related titles available from Continuum:

Meaning in Context
Edited by Jonathan J. Webster
New Discourse on Language
Edited by Monika Bednarek and J. R. Martin
Edited by
Ahmar Mahboob and Naomi K. Knight
Continuum International Publishing Group
The Tower Building 80 Maiden Lane
11 York Road Suite 704
London SE1 7NX New York, NY 10038
www.continuumbooks.com
© Ahmar Mahboob, Naomi K. Knight and contributors 2010
All rights reserved. No part of this publication may be reproduced or transmitted

in any form or by any means, electronic or mechanical, including photocopying,
recording, or any information storage or retrieval system, without prior permission
in writing from the publishers.
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
ISBN: 978-1-4411-6415-5 (hardcover)
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.
Typeset by Newgen Imaging Systems Pvt Ltd, Chennai, India

Printed and bound in Great Britain by the MPG Books Group
Contents
List of Figures vii

List of Tables ix
Notes on the Contributors xi
Acknowledgements xv
Chapter 1 Appliable Linguistics: An Introduction 1

Chapter 2 Pinpointing the Choice: Meaning and the Search 13
for Equivalents in a Translated Text
M. A. K. Halliday
Chapter 3 Appliable Linguistics and English Language Teaching: 25
The Scaffolding Literacy in Adult and Tertiary
Environments (SLATE) Project
Ahmar Mahboob, Shoshana Dreyfus, Sally Humphrey and
J. R. Martin
Chapter 4 Negotiating Evaluation: Story Structure and Appraisal 44
in Youth Justice Conferencing
J. R. Martin, Michele Zappavigna and Paul Dwyer
Chapter 5 Modelling Social Affiliation and Genre in the Civic Domain 76
Sally Humphrey
Chapter 6 ‘News’ and ‘Register’: A Preliminary Investigation 92
Annabelle Lukin
Chapter 7 Constructing Sports Stars: Appliable Linguistics and the 114
Language of the Media
Shoshana Dreyfus and Sandra C. Jones
Chapter 8 Naming Culture in Convivial Conversational Humour 130
Naomi K. Knight
Chapter 9 Visualizing Appraisal Prosody 150
Michele Zappavigna, Chris Cléirigh, Paul Dwyer and J. R. Martin
vi Contents
Chapter 10 Evaluative Stance in Humanities: 168

Expectations and Performances
Erika Matruglio
Chapter 11 The 3×3: Setting Up a Linguistic Toolkit for Teaching 185
Academic Writing
Sally Humphrey, J. R. Martin, Shooshi Dreyfus and
Ahmar Mahboob
Chapter 12 Why Are Logical Connectives Sometimes Detrimental 200
to Coherence?
Alan Jones
Chapter 13 Contestable Reality: A Multi-level View on Modality 221
in Multimodal Pedagogic Context
Yumin Chen
Chapter 14 Making Many Meanings in Popular Rap Music 234
David Caldwell
Chapter 15 Rank in Visual Grammar: Some Implications for 251
Multimodal Discourse Analysis
Sumin Zhao
Chapter 16 The Meaning of ‘Not’ Is Not in ‘Not’ 267
Ruqaiya Hasan
Index 307
List of Figures
Figure 3.1 Strata and metafunctions in SFL 28

Figure 3.2 Teaching–learning cycle (Rothery, 1996) 29
Figure 3.3 Feedback structure with required and optional stages 32
Figure 3.4 Consultation process for linguistics course 35
Figure 4.1 Appraisal systems (after Martin and White, 2005) 54
Figure 5.1 Cultural domains of adolescents’ lives 81
Figure 5.2 The position of elemental genres within Chilout genre 84
complexes
Figure 5.3 Orbital structure of political testimony 89
Figure 6.1 Hartley’s presentation modes mapped onto Cloran’s 97
continuum of the role of language in the social process
(Cloran, 1994, p. 132)
Figure 6.2 Hartley’s presentation modes (1982). Reading left to right 98
1. ‘voice over’; 2. ‘vox pop’; 3. ‘stake out’
Figure 6.3 Hasan’s 1999 network for field 98
Figure 8.1 Levels of affiliation with examples of their construal 137
Figure 9.1 Seating configuration of the ‘AffrayYJC’ (blurred to maintain 152
anonymity)
Figure 9.2 An example of tone group analysis performed in ELAN 153
Figure 9.3 Configuration for recording a YJC 154
Figure 9.4 Situating judgement within the appraisal system (based on 156
Martin and Rose, 2003)
Figure 9.5 Distribution of evaluation across the gesture space 160
Figure 9.6 Themeriver visualization of Associated Press news wire 161
stories (Havre et al., 2002, p. 12)
Figure 9.7 An example of a simple area graph 162
Figure 9.8 A simple example of a stacked area graph 162
Figure 9.9 Positive and negative judgement (invoked and inscribed) 163
in the ELO’s intervention stage
Figure 9.10 An example of coupling 164
Figure 11.1 The 3×3: a framework for describing linguistic resources of 188
student writing in the academic domain
Figure 11.2 Four domains of learning (adapted from Macken-Horarik, 189
1996)
viii List of Figures
Figure 11.3 Pathway of realization of ideational meanings in Text 11.1 194

Figure 11.4 Classifying taxonomy within phase of Text 11.1 194
Figure 11.5 Pathway of realization of interpersonal meanings in Text 11.1 195
Figure 11.6 Pathway of realization of textual meanings in Text 11.1 196
Figure 12.1 Mann and Thompson’s Discourse Relations (1988) 204
Figure 13.1 Modality scale for colour saturation (Kress and 224
Van Leeuwen, 2006, p. 160)
Figure 13.2 Sensory coding orientation (Excerpted from PEP Primary 226
English Students’ Book I for Year 3, 2003, p. 47). Reproduced
with permission
Figure 13.3 Abstract-sensory coding orientation (Excerpted from Go for it 228
Students’ Book II for Year 7, 2005, p. 15). Reproduced with
permission
Figure 13.4 Naturalistic coding orientation (Excerpted from New Senior 230
English for China Student’s Book 2, 2004, p. 25). Reproduced
with permission
Figure 14.1 System network for timing (after Van Leeuwen, 1999, p. 61) 238
Figure 14.2 System network for sound quality (Van Leeuwen, 1999, 238
p. 151)
Figure 14.3 Slow to fast rapping (as syllables per minute (spm)) 241
in the Kanye West Corpus
Figure 14.4 Slow to fast singing (as syllables per minute (spm)) 241
in the Kanye West Corpus
Figure 14.5 Low to high rapping pitch registers in the Kanye West Corpus 242
Figure 14.6 Low to high singing pitch registers in the Kanye West Corpus 243
Figure 14.7 Smooth to rough rapping in the Kanye West Corpus 244
Figure 14.8 Smooth to rough singing in the Kanye West Corpus 244
Figure 14.9 Some attributes (or meaning potentials) in the Kanye West 245
repertoire
Figure 14.10 The Kanye West repertoire from a topological perspective 245
Figure 15.1 Rank scale in English 252
Figure 15.2 A Queensland hero 253
Figure 15.3 Women cricketers fielding 255
Figure 15.4 Types of structures and types of meanings 258
Figure 15.5 The construal of membership in language and image 259
Figure 15.6 The unfolding path of the multimodal text 260
Figure 16.1 Moving and meaning (source: Halliday, 2004, p. 9) 271
Figure 16.2 Strata in a functional theory of language 274
Figure 16.3 Two systems and their relationships: realization and 279
instantiation
Figure 16.4 Simplified system network of options in demanding 282
goods/services
Figure 16.5 A simplified system network of options in MOOD 289
Figure 16.6 Simplified system network of options in demanding information 291
List of Tables
Table 4.1 Inscriptions of affect in the Interpretation stage of the 60

mobile phone recount
Table 4.2 Inscriptions of judgement in the Interpretation stage of the 61
mobile phone recount
Table 4.3 Inscriptions of honesty in the ELO’s Interpretation stage 64
of the affray recount
Table 4.4 Inscriptions of capacity in the ELO’s Interpretation stage 64
Table 4.5 Inscriptions of propriety in the ELO’s Interpretation stage 65
Table 4.6 Inscriptions of impropriety in relation to Mum 67
Table 4.7 Inscriptions of impropriety in relation to the Muslim community 67
Table 5.1 HNN: Interaction of evaluative and referential phases across 88
Involvement element
Table 7.1 Appraisal values of Cousins’ behaviour 126
Table 10.1 Comparison of Appraisal between syllabus rationales 175
Table 10.2 Comparison of attitude and graduation in student texts: 176
percentages
Table 10.3 Inscribed Attitude in student texts as percentage of 177
whole text
Table 10.4 Inscribed Attitude in student texts 177
Table 10.5 Total Attitude (inscribed and invoked) in student texts 178
Table 10.6 Number of instances of inscribed Attitude in student texts 179
Table 10.7 Number of instances of invoked Attitude in student texts 179
Table 10.8 Percentages of inscribed Attitude in student texts 179
Table 10.9 Percentages of total Attitude in student texts 180
Table 13.1 The distribution of visual styles in EFL textbooks for different 232
levels of education
Table 14.1 List of Kanye West’s awards (from Aceshowbiz: 236
Kanye West’s Awards, n.d.)
Table 14.2 The ‘moods’ of Kanye West and 50 Cent 247
(from Birchmeier, n.d.)
Table 16.1 Lexicogrammatical realization statements for some 285
semantic options
x List of Tables
Table 16.2 The meaning of polarity with some primary MOOD options 290
in English
Table 16.3 The meaning of ‘not’ in questions: semantic and 292
grammatical views
Table 16.4 The meaning of ‘not’ with demand information . . . 294
apprize: specify
Notes on the Contributors
David Caldwell is a PhD candidate in Linguistics at the University of Sydney

where he is applying social semiotics and discourse analysis to rap musician
Kanye West. His research interests include discourse analysis, multimodality,
and Systemic Functional Linguistics. As a discourse analyst, David has investi-
gated and published on a range of language contexts, including rap music,
post-match interviews between AFL footballers and ABC journalists, and medi-
cal consultations.
Yumin Chen is a lecturer and post-doctoral research fellow in the School of

Chinese as a Second Language, Sun Yat-sen University. Yumin received her PhD
degrees in Linguistics from the University of Sydney and Sun Yat-sen University.
Her research focus has been on the social semiotic analysis of pedagogic dis-
course. Major publications of hers include ‘Exploring dialogic engagement
with readers in multimodal EFL textbooks in China’ in Visual Communication,
and ‘The semiotic construal of attitudinal curriculum goals: Evidence from EFL
textbooks in China’ in Linguistics and Education (2010).
Chris Cleirigh is an unemployed linguist who received his PhD from the Uni-
versity of Sydney in 1999. He has worked on research projects since 1993 in
both Speech Technology and Linguistics, in both the university and private
sector, and has publications in both fields.
Paul Dwyer is currently Chair of the Department of Performance Studies at the

University of Sydney. In addition to researching restorative justice practices
in the NSW juvenile justice system (together with Jim Martin and Michele
Zappavigna), he has carried out fieldwork study of reconciliation ceremonies in
Bougainville (PNG) and has published widely on the use of drama in health,
education and welfare settings (particularly the methods associated with
Brazilian director Augusto Boal).
Shoshana Dreyfus is a postdoctoral fellow in the Department of Linguistics at

the University of Sydney where she is researching applications of Sydney School
genre pedagogy in the tertiary context and attempting to describe the knowl-
edge structure of the field of Linguistics. She completed her PhD, which stud-
ied the nonverbal multimodal communication of a child with an intellectual
disability at the University of Wollongong. Her research interests include the
xii Notes on the Contributors
intersections between critical discourse analysis and systemic functional linguis-

tics, critical literacy and atypical language use.
M. A. K. Halliday was born in the north of England in 1925, received his B.A. in
Modern Chinese Language and Literature from the University of London in
1948 and his Ph.D. at Cambridge in 1954; along the way he studied linguistics
at Peking University and Lingnan University in China. He taught at Cambridge
and at Edinburgh, and became Professor of General Linguistics at University
College London; he left to become a Fellow at the Center for Advanced Study
in the Behavioral Sciences and take up various short appointments, before
moving to Australia as Foundation Professor of Linguistics at the University of
Sydney, where he remained until his retirement at the end of 1987. Since
then he has held visiting appointments in various countries, the latest being at
The Hong Kong Polytechnic University in 2009. Professor Halliday has been
awarded honorary degrees by twelve universities in Europe, Asia, Australia and
Canada.
Emeritus Professor Ruqaiya Hasan taught Linguistics at Macquarie University,

Australia. She is an internationally recognised linguist who has held visiting
positions at various Universities in Europe and America. She has published
widely in such areas as Lexicogrammar, Semantics, Sociolinguistics (with spe-
cial reference to Semantic Variation), Discourse and Context Analysis, as well as
Stylistics. Seven Volumes of her Collected Works edited by Jonathan J Webster
are under publication by Equinox Publishing, London. The first two volumes
now available are, Volume 1: Language, Society and Consciousness (2005) and
Volume 2: Semantic Variation: Meaning in Society and in Sociolinguistics
(2009) respectively. The third Volume Language and Education: Learning and
Teaching in Society is expected this year.
Sally Humphrey is a post-doctoral fellow in the Department of Linguistics at

University of Sydney. She has worked for many years as a teacher and teacher
educator in the field of literacy and language education in school and higher
education contexts. Sally’s current research interests include: language develop-
ment in Biology across the undergraduate years; the role of Joint Construction
in literacy development and Appraisal in persuasive writing in academic and
civic discourse.
Alan Jones is a Senior Lecturer in the Department of Linguistics at Macquarie

University where he convenes postgraduate programs in professional and
organisational communication. He has taught and researched academic English
for more than twenty years, carrying out collaborative research and co-publishing
with subject area specialists in physics, accounting and law. His broader research
interests include the intersection of interpersonal meanings with ideational
content in writing and the implications of social theory for a social linguistics.
Notes on the Contributors xiii
He also carries out research in the cultural anthropology of Papua New Guinea
and the Solomon Islands.
Sandra Jones is the Director of the Centre for Health Initiatives, a Research
Strength at the University of Wollongong, and a Professor of Social Marketing/
Public Health. From 2004 to 2009 she was also Associate Dean (Research) in
the Faculty of Health & Behavioural Sciences. Sandra’s research focuses on the
relationship between media and health, including the impacts of advertising
in the print and electronic media on health behaviour, and the use of social
marketing to improve population health. Sandra has published more than
100 refereed papers and been awarded in excess of $3 million in research
funding.
Naomi K. Knight is a PhD candidate in the Department of Linguistics at the

University of Sydney. She is the co-author of Questioning Linguistics (2008) and
co-convener of the International Free Linguistics Conference with Dr. Ahmar
Mahboob. She has published in the areas of conversational humour, discourse
analysis, systemic functional linguistics, and sociolinguistics, and has also been
involved in projects in ape language studies involving data of language-competent
apes Kanzi and Panbanisha.
Annabelle Lukin is a Senior Lecturer and Macquarie University Research Fellow

in the Centre for Language in Social Life, at Macquarie University in Sydney.
Her research interests include: register and context studies from a systemic
functional perspective, media and political discourse, and stylistics, including
stylistics and translation.
Ahmar Mahboob is a Senior Lecturer in the Department of Linguistics at the

University of Sydney. Ahmar has published on a range of topics including:
language teaching, teacher education, language policy, educational linguistics,
and World Englishes. Ahmar is the co-editor of Questioning Linguistics with
Naomi K. Knight (2008), Studies in Applied Linguistics and Language Learning
with Caroline Lipovsky (2009), and The NNEST Lens: Nonnative English Speakers
in TESOL (2010). Ahmar is the Associate Editor of the journal Linguistics and the
Human Sciences.
Erika Matruglio is a PhD candidate in the Faculty of Arts and Social Sciences at
the University of Technology, Sydney. Her research interests include the con-
struction of interpersonal stance in student writing, disciplinary differences in
student literacy practices, the epistemological and axiological bases of school
subjects and educational linguistics generally. She is a former high school lan-
guages and ESL teacher, has lectured in ESL and TESOL methodology and is
currently involved in a project investigating disciplinary differences and the
building of cumulative knowledge in high school subjects.
xiv Notes on the Contributors
J. R. Martin is Professor of Linguistics (Personal Chair) at the University of

Sydney. His research interests include systemic theory, functional grammar,
discourse semantics, register, genre, multimodality and critical discourse
analysis, focussing on English and Tagalog - with special reference to the trans-
disciplinary fields of educational linguistics and social semiotics. Recent
publications include The Language of Evaluation (with Peter White) Palgrave
2005; Language, Knowledge and Pedagogy (Edited with Fran Christie) Con-
tinuum 2007; and with David Rose, a second edition of Working with Discourse
(Continuum 2007) and a book on genre (Genre relations: mapping culture,
Equinox 2008). He has also recently completed a 2nd edition of the 1997 func-
tional grammar workbook, with Clare Painter and Christian Matthiessen,
Deploying Functional Grammar (in press with Commercial Press, Beijing) and
an edited collection (with Monika Bednarek), New Discourse on Language (in
press with Continuum). Professor Martin was elected a fellow the Australian
Academy of the Humanities in 1998, and awarded a Centenary Medal for his
services to Linguistics and Philology in 2003.
Michele Zappavigna is a research fellow in the Department of Linguistics at

the University of Sydney. Her major research interest is electronic discourse
and social media and she is currently working on a corpus-based study of the
language of Microblogging. She alsohas an ongoing interest in text visualiza-
tion as a tool to aid discourse analysts. Michele works on a project investigating
NSW Youth Justice Conferencing, a form of restorative justice, using multi-
modal discourse analysis. She completed her PhD on language, tacit knowledge
and technology in the School of Information Technologies, University of
Sydney.
Sumin Zhao is currently completing her PhD. thesis on children’s educational

E-texts. As a part of an ARC discovery project, her research explores the dynamic
relations between language and other semiotic recourses in hypermedia texts
and the ways in which these resources are employed to construe the field of
knowledge, primary history and social science in particular. She also publishes
in school literacy, SFL and multimodal discourse analysis.
Acknowledgements
An edited volume is the joint effort of a number of people. It would be difficult

to list all of them here, but we would like them to know that we sincerely
appreciate their support and input. We would specially like to thank our
contributors – their commitment to this volume and to the field has been
exemplary. Special thanks are also due to the many reviewers who shall remain
anonymous, but who spent hours reading, reviewing, and providing valuable
feedback to the authors – without their work, this volume would not have
reached its final shape.
Many of the papers included in this volume were first presented at the
2nd Free Linguistics Conference (FLC), 2008. We would therefore like to rec-
ognize the significant role that this conference is playing in supporting the
work in our field. We would like to thank all the volunteers and organizers of
the FLC. We would also like to thank the sponsors who make it possible for FLC
to remain free of cost to the presenters and attendees. These main sponsors
include the Faculty of Arts, School of Letters Arts and Media (SLAM), and the
Department of Linguistics at the University of Sydney. We would also like to
thank Bridge Bookshop, Cafe Ottimo, the Coop Bookshop, Gleebooks, and
Starbucks Coffee for supporting FLC. We would like to thank Australian Sys-
temic Functional Linguistics Association (ASFLA) for providing support to
help us with developing the index for this volume.
We would also like to thank our families and friends, who supported us as we
worked on this project and on the Free Linguistics Conference.

This page intentionally left blank
Chapter 1
Appliable Linguistics: An Introduction

This chapter introduces the need to establish Appliable Linguistics as the

framework through which we study language-related issues – in theoretical lin-
guistics, Applied Linguistics and other language-related disciplines. Appliable
linguistics is an approach to language that takes everyday real-life language-
related problems – both theoretical and practical – in diverse social, profes-
sional and academic contexts as a starting point and then develops and
contributes to a theoretical model of language that can respond to and is appli-
able in the context. The concept of Appliable Linguistics used here is built on the
work of M. A. K. Halliday (1985, p. 7) who believes that ‘the value of a theory
lies in the use that can be made of it’; he continues: ‘I have always considered
a theory of language to be essentially consumer oriented’. Informed by this,
the book considers Systemic Functional Linguistics (SFL) Appliable Linguistics.
However, before further discussing Appliable Linguistics, we will outline why
such an approach is necessary by reviewing the work done under the more
familiar label of ‘Applied Linguistics’.
Since its inception in the mid-1940s, applied linguistics has grown to examine
how language relates to and works in different social contexts. In doing so,
applied linguistics has developed a number of sub-fields, each with its own body
of expert knowledge, such as: bilingualism, clinical linguistics, forensic linguis-
tics, language learning and teaching, language policy and planning, lexicogra-
phy, multilingualism, and translation and interpretation. Of these, the most
prolific area of research in applied linguistics is and has been second/foreign
language teaching, and more specifically, teaching English as a second/foreign
language. Over the last 60–70 years, while applied linguistics evolved and
extended its scope, it also developed a very ambivalent relationship with theo-
retical/descriptive linguistics. During its early days, applied linguists (especially
those working on language learning/teaching) drew on linguistic theory to
examine fields of study where language played a key role. In stating this, Halliday
et al. (1964) wrote that ‘the specialist [applied linguist] in this subject is a kind
of middleman who exists to bridge the gap between theory and newly acquired
knowledge on the one hand and the everyday problems of teaching a language
2 Appliable Linguistics
on the other’ (pp. xiv–xv). However, by mid-1970s, researchers in applied lin-

guistics weakened their links with linguistics. Pit Corder (1978, p. 5), in the first
issue of Studies in Second Language Acquisition, deplored this weakening link
between linguistics and applied linguistics (especially second language studies)
and stated that
It would not be too great an exaggeration to say that even as recently as

six years ago a sizeable proportion of people who called themselves applied
linguists would have considered that the ‘linguistics’ they were involved in
applying was ‘theoretical linguistics’ or ‘descriptive linguistics’, and that,
while they did not deny that there were psychological or sociological dimen-
sions to language learning, it was not part of ‘applied linguistics’ proper to
concern itself with the application of these theoretical sciences in the plan-
ning and execution of language teaching programs.
While Corder did not provide a detailed discussion of why applied linguists
moved away from linguistics, we posit that this shift (particularly in the United
States) was a response to the nature of the theories of language that were (and
continue to be) mainstream in the United States, for example, generative
linguistics. The formalist paradigm, which is dominant in the United States,
prioritizes ‘langue’ (system) over ‘parole’ (language use/text as an instance of
the system). As such, linguistics has moved its focus away from a discussion of
language as it is used and as it varies in social context. This limits its usefulness
for applied linguists who see their job as the ‘investigation of real-world prob-
lems in which language is a central issue’ (Brumfit, 1995, p. 27).
While one can see why linguistic theories such as generative grammar or
optimality theory have little relevance to applied linguistics, the lack of engage-
ment of sociolinguistics, which by definition studies language in social contexts,
with applied linguistics research is slightly more complex. Beatriz Lavandera
(1978) points out that sociolinguistic variation studies deal primarily with
morphological, syntactic and lexical variation and ‘suffer from the lack of an
articulated theory of meanings’ (p. 171). She problematizes this lack of atten-
tion to meaning in sociolinguistic research and argues that different forms
mean different things and therefore should be studied as such. Without such
consideration, she argues, a study of formal variables ‘can only be heuristic
devices, in no sense part of a theory of language’ (p. 179). This is a severe criti-
cism of studies in sociolinguistics that do not consider meaning to be an essen-
tial aspect of their study and also suggests why sociolinguistics does not play
a central role in studies in applied linguistics. Because it looks at language
use and context, applied linguistics needs to take meaning into consideration.
Sociolinguistic research that does not take account of meaning is therefore not
seen as useful or relevant to applied linguists.
Given the limited ability of mainstream theories of language to shed light on
the social and semiotic aspects of language, applied linguists have become quite
Introduction 3
selective in their use of work in linguistics and sociolinguistics. For example,

our review of some of the current literature in second language acquisition
studies shows that there has been a steady increase in research that draws on
pragmatics but not on other theories of linguistics. Pragmatics, with its focus on
language in use, has been gaining interest in applied linguistics (unlike genera-
tive linguistics or other branches of sociolinguistics) because it studies language
in context and therefore resonates with the needs of applied linguists. This
observation is central to the arguments and studies included in this volume:
to be of relevance and use, linguists need to take language use and meaning as
a starting point.
So far, we have criticized areas of mainstream linguistics for not developing
theories that serve the needs of applied linguists. However, we can also criticize
applied linguistics for not having contributed to or worked towards a compre-
hensive theory of language either. Given that applied linguistics focuses on
language (in) use, it has a role in developing and/or contributing to a theory
of language. One reason why this may not have happened is because in recent
years a growing body of applied linguistics research has focused on the politi-
cal, psychological and social aspects of language use without really engaging
with language (as lamented by Corder in the quote above). As such, this recent
body of work focuses on studies of speakers, users and uses of language rather
than studying language (in) use. Once again, what is needed is a kind of linguis-
tics that has the potential of being appliable in different contexts and for diverse
purposes. The studies included in this volume use and work towards such an
appliable linguistics.
As stated earlier, the concept of ‘Appliable Linguistics’ used in this volume
comes from M. A. K. Halliday. In keeping the users and the uses of his work
in focus, Halliday has attempted to develop an appliable linguistics. Halliday
(2006b, p. 19) states:
I have always tried to work with a functional orientation to language; not

eschewing theory, because without theory there can be no consistent and
effective practice, but treating a theory as a problem-solving enterprise and
trying to develop a theoretical approach, and a theoretical model of lan-
guage, which can be brought to bear on everyday activities and tasks. I call
this an ‘appliable’ linguistics: appliable rather than applicable, because the
word ‘applicable’ refers to one particular purpose, whereas ‘appliable’ means
having the general property that it can be put to use in different operational
contexts.
In further elaborating on the term, Halliday (2006a) stated that appliable lin-
guistics makes it possible for us to develop
a comprehensive and theoretically powerful model of language which, pre-

cisely because it was comprehensive and powerful, would be capable of being
applied to the problems, both research problems and practical problems,

that are being faced all the time by the many groups of people in our modern
society who are in some way or other having to engage with language.
The present volume takes this understanding of appliable linguistics as a

starting point. The chapters included in this volume focus mostly on issues of
education and multimodality and draw for the most part from (and contribute
to) a systemic-functional (SFL) understanding of language. However, while the
concept of an appliable linguistics is developed in this volume through SFL
theory, it is not the only appliable linguistics. SFL is compatible with the idea of
an appliable linguistics in that it was developed by M. A. K. Halliday to under-
stand language in ways that were appliable in educational contexts, but other
theories of linguistics may also be interpretable as ‘appliable’. One theory that
is appliable in ways similar to SFL is Tagmemics (Pike, 1981). Developed by
Kenneth Pike in the 1950s, the theory has been used extensively to describe
previously unknown languages and for translation purposes (especially Bible
translation as Pike was closely associated with SIL – Summer Institute of
Linguistics). Pike’s work has also been used in a number of other disciplines.
In particular, his discussion of ‘etic’ and ‘emic’ approaches to looking at any
data/phenomenon has been used by linguists as well as by people working in
other disciplines (e.g. anthropology, philosophy, psychology, etc.). As such,
Tagmemics can be seen as another theory that is compatible with an appliable
linguistics framework.
It is our hope that this volume will encourage additional work that is appliable
and bring together researchers and linguists who take an appliable approach to
language studies. There are 15 chapters in the volume, led by Michael Halli-
day’s own appliable linguistic study, and ending in Chapter 16 with a consider-
ation of linguistic meaning through the systemic functional model by Ruqaiya
Hasan. The chapters are organized based on the key concepts and aspects that
connect them so that not only does each chapter develop the meaning of appli-
able linguistics in its own unique way, but it also follows from and builds on the
previous chapter. We have consciously structured the book without dividing
the chapters into sections (such as ‘theoretical’ and ‘applied’) because in our
pursuit of an ‘appliable linguistics’ it is our aim to dissolve these distinctions.
Instead, this volume brings together what is traditionally separated, and the
chapters are associated by their conceptual linguistic connections.
Halliday begins the volume (in his chapter entitled ‘Pinpointing the choice:
meaning and the search for equivalents in a translated text’) by contextualizing
the chapters to follow within the theme of an ‘appliable linguistics’, and exem-
plifying this approach through a consideration of the translation of a tourist
text from Chinese to English. From a model of language as the exercise of
choice, Halliday exhibits how our applications of the linguistic framework can
always be located as choices in the linguistic systems of meaning in terms of
Introduction 5
a realizational relationship whereby ‘a theory is made manifest, or “realized”, in

the processes of being applied’. ‘Appliable linguistics’ is then proposed as a
unifying concept to bring together theoretical and applied linguistics, since the
application of a linguistic theory always operates with, or more specifically
makes choices within, systems of meaning. In this sense, linguistics – and in
particular, SFL as a theory of language as choice – is shown to be a valuable
resource for determining meaningful differences across contexts of application
as a fundamental matter of language. As one possible application of linguistic
theory, Halliday demonstrates how one may ‘pinpoint the choices’ made in
linguistic terms as translators shift across strata and ranks, and explains his
efforts as follows: ‘What I am trying to do, in “pinpointing the choice”, is to use
the analytic tools of linguistics, and particularly perhaps of grammatics, to
examine the significance of alternative renderings for a reader of English text.’
Through an informative table of findings, Halliday demonstrates how shifts
occur across linguistic dimensions in the translator’s efforts to achieve equiva-
lence in the target language. By providing the examination of this translated
text, Halliday makes it clear that linguistic theory is intrinsically tied to its appli-
cations particularly when language is perceived as ‘the exercise of choice’,
which leads fittingly into the following chapter exploring English language
teaching and learning.
In their chapter entitled ‘Appliable linguistics and English language teaching:
the Scaffolding Literacy in Adult and Tertiary Environments (SLATE) project’,
Mahboob, Dreyfus, Humphrey and Martin report on a project that depicts
appliable linguistics through an academic literacy programme called SLATE,
training students and tutors in tertiary education and online environments.
Referring to this work as the ‘fourth generation of Sydney School work on
genre pedagogy’, the authors exhibit language theory in practical use for the
improvement of education, both in teaching and in learning. In particular, by
providing tools to inform students before they perform a writing task (as the
authors show in a Linguistics paper example), the SLATE project arms them
with a ‘meta-discursive knowledge base’ that is achieved through a systemati-
cally principled cycle of learning, which also benefits tutors involved in their
learning. Essentially, this chapter shows how genre pedagogy continues to be an
effective model across levels and modes of education, and how this depends
upon its constant development and adaptation in regards to the specific needs
of its application. As the authors explain, the model is adapted in SLATE in
response to specific problems that arise for the students. The chapter ends with
promising future directions.
Generic structure is further employed in the context of Youth Justice Confer-
encing (YJC) in the following chapter, ‘Negotiation evaluation: story structure
and appraisal in youth justice conferencing’, by Martin, Zappavigna and Dwyer.
Focusing on the negotiation of evaluative meaning in young persons’ accounts
of their offending behaviour via appraisal analysis (Martin and White, 2005),
they pursue the question of ‘restoration’ and how successfully these types of
conferences achieve this end. The authors propose staging for what they refer
to as a ‘commissioned recount’ genre, showing how this genre supports the
construction of a particular identity for the young person, as evaluations are
‘co-created and guided by other participants’ through the ritualistic elements
of the process. While they note that criminological research most often takes a
quantitative approach, these authors provide a qualitative analysis with a back-
ground of theories across functional linguistics, social semiotics and perfor-
mance studies to uncover aspects of the actual dialogue that takes place. With
this approach, they conclude that the YJC offers at least a chance for the young
person and victim (along with others present) to affiliate and perhaps to reha-
bilitate through the ‘doing’ of the conference, differing in important ways from
the outcomes of the traditional criminal justice system.
Humphrey adapts the semiotic model of context in SFL in her chapter,
‘Modelling social affiliation and genre in the civic domain’, in order to apply
it to the speeches of social activists. Supporting the concept of a Positive
Discourse Analysis (PDA) (Martin, 2004), Humphrey adjusts her use of genre
theory through the notion of ‘civic domain’ in response to the fact that ‘the
emphasis on relatively stable academic genres has resulted in models of genre
classification which do not fully account for emerging, blended and less stable
genres created in other domains of adolescents’ lives’. She draws upon data of
speeches, interviews and newspaper commentaries from an Australian social
movement (called ‘Chilout’) to demonstrate how these activists achieve their
goal of persuasion in the text to promote their goals. By opening up the model
of context to account for cultural domains, Humphrey also proposes an inter-
mediate space of ‘social affiliations’ to describe how people come together
around a common endeavour, and she exhibits this through different elements
of persuasion in their texts. In this chapter, language is shown to significantly
affect, and be affected by, social context in a genre model.
From another point of view, Lukin offers an interesting application of
Halliday’s notion of ‘register’ to hard news in her chapter, ‘“News” and “regis-
ter”: a preliminary investigation’. As an initial investigation utilizing this con-
cept in the analysis of news discourse, and following from those studies
committed in the genre school, Lukin contends that the register model offers
a great deal of descriptive power to be found through the parameters of social
context, and that this ‘comes from its openness to the dynamism of social con-
text, an openness which derives from it being in a dialectic (i.e. realizational)
relation to social context’. Drawing on semantic tools including Cloran’s (1994)
Rhetorical Unit analysis and Hasan’s (1984) Cohesive Harmony analysis, this
chapter describes the registerial configuration of linguistic meanings associ-
ated with the particular situational field, tenor and mode in a news broadcast
concerning the invasion of Iraq. This preliminary investigation focuses on how
power relations are carried through those meanings that are the focus of regis-
ter analysis, reinforcing Bernsteinian notions that inform the social linguistic
model of SFL.
Introduction 7
A concern with the communication of media is also expressed in Dreyfus

and Jones’ chapter entitled ‘Constructing sports stars: appliable linguistics and
the language of the media’, and in particular the impact of the behaviour of
sports stars through their construction in the news. The authors put forward
the question of whether Ben Cousins, the well-known Australian footballer
and drug abuser, was constructed in media language as responsible, and use
the systemic functional tools of transitivity and ergativity analysis (displaying
agency) along with an analysis of attitude to determine whether this is so. Two
Australian print newspapers were considered from within the time frame of
Ben Cousins’ suspension. In their findings, the subject is not often depicted as
an agent in his own downfall, but is presented by the newspapers in attitudinal
terms as a sports hero on the field, while a disgraced hero off the field. Dreyfus
and Jones conclude that Cousins’ depiction through language as less than
blameworthy may have negative effects on the young adolescents who look up
to him. The way that we use language towards individuals who we identify as our
‘heroes’ is shown to be demonstrably skewed with the help of specific linguistic
tools for analysis, as employed by the authors of this chapter.
Attitudinal language and identity construction are equally salient in the
interpretation of what we laugh at, as shown in Knight’s ‘Naming culture in
convivial conversational humour’. Knight expounds a model of ‘affiliation’ that
accounts for how friends in conversation negotiate community and relational
identity together, both at the local and global level. Humour is presented as a
strategy for affiliation by which participants at talk manage their social com-
plexity as members of a culture, as they laugh off tensions between values that
are presented in the linguistic text as couplings (Martin, 2000) of attitude with
experience. Knight lays bare the implicit couplings that occur in her examples
through naming of individuals, groups and communities in humorous opposi-
tion, which works to expose hierarchical roles and power relations as laughable
in the service of solidarity. In applying her social model of affiliation that relates
the social semiotic relations of communities (and ideologies) to evidence in
text, Knight shows how ‘speakers not only share a laugh but also reaffirm their
roles in society as complex members of communities that are more or less
negotiable’.
Zappavigna, Cléirigh, Dwyer and Martin further the affiliation model in the
following chapter, ‘Visualizing appraisal prosody’, focusing on the prosodic
structure of evaluative meanings across an extended text of a Youth Justice
Conference (as detailed by Martin et al. in Chapter 4). Through a cross-stratal
and multimodal analysis, the authors address the problem that linguists face
in attempting to view long-range prosodic patterns, such as the ‘long-range
evaluative patterning’ by the Ethnic Liaison Officer (ELO) in their data, by
representing it visually through Text Visualization. Linguistic tools such as
appraisal and phonological analysis are combined with visual and semiotic tools
including gesture to further the evidence for the conclusions the authors make.
In particular, phonological and gestural evidence are given to support the pro-
sodic evaluative structure of opposites presented by the ELO. The data in this
case compels the use of multimodal tools, and a cross-stratal perspective, exhib-
iting the essential interaction of theory and application. The authors explain
that the ELO creates a contrast between prosodies by ‘oscillating’ between
judgements of the young person, as made clear through the visual representa-
tion, and they end the chapter by proposing that this contrast has the affiliative
function of integrating the young person back into the community.
Evaluative meaning and the linguistic framework of appraisal are brought
into the pedagogic context in ‘Evaluative stance in humanities: expectations
and performances’ by Matruglio. In particular, what kinds of implicit features
of language and writing are expected of students, and what they are expressing
in their writing, are shown to be retrievable through the application of a lin-
guistic framework such as appraisal analysis. Situating her study in four humani-
ties subjects within secondary education, Matruglio considers variation in how
attitudes are expressed in ‘high stakes examination writing’ (along with syllabus
documents) to make explicit the evaluative work that is necessary and ‘appro-
priate’ for students within each subject. By gathering what she calls ‘Appraisal
profiles’, differences in the use of explicit and implicit attitudes, and the inter-
play between them, are clarified according to these subjects. This chapter offers
an important contribution to the study of interpersonal meaning and expecta-
tions in the educational context, and it turns the focus towards the pedagogic
for the chapters to follow.
In Humphrey, Martin, Dreyfus and Mahboob’s chapter, ‘The 3x3: setting
up a linguistic toolkit for teaching academic writing’, the authors provide a
‘semiotic toolbox’ for training literacy teachers based on systemic functional
genre and register theory. Based on their work in the SLATE literacy project
(described in Chapter 3), the ‘3×3 framework’ combines three metafunctions
(ideational, interpersonal, textual) and three language strata (social activity,
discourse semantics, and grammar and expression) to distinguish areas of con-
cern in the examination of student writing. They describe how the 3×3 maps
out linguistic patterns that are relevant at each particular level in response to
what is expected of students in academic writing, and they focus on the genres
and hence linguistic resources that are privileged in the academic domain to
make these resources appliable across disciplines. The example of a student’s
science text given by the authors demonstrates the usefulness of the framework,
expounding the linguistic resources in use at different strata and showing how
they construe an academic register. By offering teachers a metalanguage for
assessment and feedback, the authors illustrate how these aspects can be made
more systematic and less ambiguous in education.
Jones examines the potentially ambiguous nature of logical connectives in
his chapter, ‘Why are logical connectives sometimes detrimental to coherence?’
Reviewing the categories given by linguists to logico-semantic relations, Jones
imparts ‘a preference for macro-categories over micro-categories’, while argu-
ing that all the categories proposed are polysemous and hence potentially pro-
blematic. He puts forward that implicit conjunction (or the absence of logical
Introduction 9
connectives), on the other hand, is current in ‘expert’ writing as evidenced in

scientific and professional texts, and functions to foreground the strategies and
knowledge the reader brings to the process of meaning-construction. Jones
offers a comprehensive investigation of logical connectives in the context of
second language pedagogy, focusing on their ‘inappropriate’ use by language
learners and their ‘underuse’ by mother tongue speakers to show the insuffi-
ciency of this feature in representing complex causal relations. Interestingly, he
also discusses how relationships such as correlation and emergence rather than
‘cause and effect’ are foregrounded in current writing such as that in the sci-
ences, furthering the backgrounding of connectives and underlining his argu-
ment that they are in essence problematic. He concludes by suggesting that
second language pedagogy can benefit from a holistic approach in terms of
representing cause/consequence relations.
Chen suggests a multimodal perspective in education in ‘Contestable reality:
a multi-level view on modality in multimodal pedagogic context’ to apply to the
use of visual semiotic resources in the classroom. In a comparative examination
of texts used across primary, junior and secondary school in China, Chen exhi-
bits support for Halliday’s pinpointing of choice (introduced in Chapter 2) by
considering the choices made between the texts in visual modality and what the
reasons behind these choices may be. Chen takes a social semiotic approach in
her analysis of the way that modality resources of visual displays are deployed
in these texts, and proposes that ‘what counts as real is socially defined and
specific to a given communicative context’, and involves a process of position-
ing the viewer in particular ways. She shows that the pictorial techniques used
across the texts follow different principles and are affected by the relation of a
constructed solidarity between the textbook editors and learners. The chapter
exposes the construal of particular values through visual communication as a
vital area of research in education, and Chen proposes that topics for the future
should include the consideration of access given to different choices in regards
to the particular field, or social activity, at hand.
Turning from multimodal relations in the pedagogic context to those in
popular culture, Caldwell pursues musical meaning in the rap music of
popular artist, Kanye West, in ‘Making many meanings in popular rap music’.
Van Leeuwen’s (1999) system networks for sound in language and music are
applied to map the meaning potential of West’s various ‘voices’ and to create
a semiotic profile of the artist for exploring why and how he appeals to such a
large and diverse consumer base. Caldwell also reinforces Halliday’s emphasis
on choice and the pursuit of contrasts in application of linguistic theory, as
he describes his interest in ‘the meaningful choices a voice can construe, as
well as the choices West has made in sampling those voices’. The chapter pres-
ents analysis of two voice types in West’s songs – the ‘rapping voice’ and the
‘singing voice’ – through the systems of rhythm, pitch register and roughness,
and the author concludes that West in fact engages with (or samples) a wide
range of meaning potentials to expand his repertoire, and thus ‘he makes
many meanings’. This informs us about his popularity and how West (in
comparison to rapper 50 Cent provided in the chapter) can affiliate with so
many through an inclusive and integrative style.
By concentrating on the analysis of multimodality, Zhao’s chapter on ‘Rank
in visual grammar: some implications for multimodal discourse analysis (MDA)’
follows suitably as an exploration of the issues that arise when a linguistic
framework (and specifically the SFL concept of ‘rank’) is used to analyse
modalities other than language (MOLs). In her review of the main MDA tradi-
tions in visual grammar, Zhao exhibits the issues, such as validity in theoretical
borrowing, that are at the forefront of MDA. Specifically, how far can we go in
extending the parameters of SFL to describe other semiotic resources? In this
sense, then, the question of how ‘appliable’ a linguistic framework is in multi-
modal relations is made relevant. In her chapter, Zhao extrapolates the ‘miss-
ing elements’ in the current debate on rank in MDA through a re-interpretation
of Halliday’s concept, and proposes that the model of constituency (parts) in
MDA does not provide for its dynamic or process-oriented nature. In her words,
a multimodal text is not made up of parts but ‘is a meaning making process,
in which choice from one semiotic system is constantly coupling with or decou-
pling from the choice made in another system’, and she outlines the impli-
cations on SFL and the challenges this presents, especially in finding a
metalanguage to account for this dynamism.
The final chapter ends the volume on the pursuit of meaning in Hasan’s
‘The meaning of “not” is not in “not”’. Hasan ‘opens afresh’ the debate on
meaning in linguistics, showing that ‘its practical participation in human social
life is both profound and extensive’ through a comprehensive discussion of
the history of meaning and its relationship to language as exemplified through
the semantics of ‘not’. Meaning is described as ‘the essence of language’, and is
shown to be central to all aspects of human life through a number of consider-
ations. Hasan brings us through a phylogenetic history of homo sapiens in
archaeological anthropology, and of the brain in this species in neuroscience;
then to an ontogenetic perspective on ‘the developmental trajectory of single
members of the community’ in Vygotsky’s concept of ‘semiotic mediation’, fur-
thered by Bernstein’s notions of ideology to separate orientations to meaning
that vary across social groups; and finally leading to Halliday’s conceptualiza-
tion of the ontogenesis of language and the systemic functional model. The
chapter is then moved into the exploration of linguistic meaning through SFL,
including its major concepts, and specifically the examination of the meaning
of negative polarity (‘not’) through Hasan’s semantic networks and a trinocular
perspective on surrounding linguistic strata (that is to say, a view above and
below the stratum of concern). While we can see the meaning of ‘not’ in its
default realizations (e.g. ‘denial’ in statements), Hasan argues that non-default
meanings of a category like this are just as important to consider, and we must
look up into the semantics from the lexicogrammar and across the range of its
paradigmatic and syntagmatic relations to gain a different view of the category’s
Introduction 11
semantic identity and value. Hasan concludes, ‘the meanings of a category of

wording are not in that category itself; they take particular shape in the specific
environment in which the wording occurs, and significantly, this environment
is most parsimoniously identified by reference to the semantic level’. Hasan
ends the chapter with a final orientation to the notion of ‘appliable linguistics’
as ‘a linguistic theory that is able to explain the efficacy of language in human
life’, and indicates that SFL’s commitment to ‘the idea of language as an ever
renewing system’ allows for this constant appliability.
The range of chapters in this volume provides a broad view of how SFL, as
one possible theory of language in linguistics as a whole, can strive towards the
integration of theory and application for a holistic – appliable – linguistics; in
other words, a linguistics that is not separated by a dichotomy of theory and use.
In Halliday’s description, these are instead connected in a token and value rela-
tionship. With language as the essential means by which we ‘do’ everyday life,
we require a model of language that applies across contexts and that can be
adapted according to the needs of our specific texts. Theorizing this model
thus goes hand in hand with how it is used, and conversely, to use a linguistic
theory we need it to be robust enough to handle that which is under our scope.
In this volume, systemic functional linguistic theory provides a model that
relates these concerns: one that foregrounds choice so that we may explore dif-
ference in the search for equivalence in a translated text; one that offers us a
semiotic toolbox by which we can improve teaching and learning in a system-
atic and principled way; one that is extendable to multimodal relations; one
that includes the intrinsic relation to social context that is open enough to
include various concepts such as genre, register and affiliation, and so on. At
the same time, SFL is a model that is constantly challenged, both in this volume
and otherwise, by its various applications. What these chapters show, however,
is that challenges compel change in the model, and this relates back to the
essence of an appliable linguistics – constant adaptation as we shunt between
theory and use, in alignment with the nature of language itself. And this is true
of linguistic theories across the discipline, which are not static but constantly
progressing. Hence, we offer this collection of research as an initial step in the
pursuit of Appliable Linguistics, which we hope will serve as a foundation for
future work across the discipline.
References
Brumfit, C. J. (1995). ‘Teacher professionalism and research’, in G. Cook and
B. Seidhofer (eds), Principle and Practice in Applied Linguistics. Oxford: Oxford
University Press, pp. 27–41.
Cloran, C. (1994). Rhetorical Units and Decontextualisation: An Enquiry into Some
Relations of Context, Meaning and Grammar. Nottingham: Department of English
Studies, University of Nottingham.
Corder, P. S. (1978). ‘“Simple codes” and the sources of second language learner’s
initial heuristic hypothesis’. Studies in Second Language Acquisition 1 (1), 1–10.
Halliday, M. A. K. (1985). ‘Systemic background’, in J. D. Benson and W. S. Greaves
(eds), Systemic Perspectives on Discourse, Volume 1. Selected theoretical papers
from the 9th international systemic workshop. Norwood, NJ: Ablex Publishing
Corporation, pp. 1–15.
Halliday, M. A. K. (2006a). ‘Working with meaning: Towards an appliable linguistics’.
Inaugural lecture to mark the launch of the Halliday Centre for intelligent appli-
cations of language studies at the City University of Hong Kong, Hong Kong.
Halliday, M. A. K. (2006b). ‘Some theoretical considerations underlying the teach-
ing of English in China’. The Journal of English Studies (Sichuan International Studies
University) 4, 7–20.
Halliday, M. A. K., McIntosh, A. and Strevens, P. (1964). Linguistic Sciences and
Language Teaching. Bloomington: Indiana University Press.
Hasan, R. (1984). ‘Coherence and cohesive harmony’, in J. Flood (ed.), Understand-
ing Reading Comprehension. Newark, DE: IRA, pp. 181–219.
Lavandera, B. R. (1978). ‘Where does the sociolinguistic variable stop?’ Language in
Society 7 (2), 171–82.
Martin, J. R. (2000). ‘Beyond exchange: appraisal systems in English’, in S. Hunston
and G. Thompson (eds), Evaluation in Text: Authorial Stance and the Construction of
Discourse. Oxford: Oxford University Press. pp. 142–75.
Martin, J. R. (2004). ‘Positive discourse analysis: solidarity and change’. Revista
Canaria de Estudios Ingleses 49 (Special issue on discourse analysis at work: Recent
perspectives in the study of language and social practice), 179–200.
Martin, J. R. and White, P. R. R. (2005). The Language of Evaluation: Appraisal in
English. London and New York: Palgrave.
Pike, K. L. (1981). Tagmemics, Discourse and Verbal Art. Ann Arbor: MI Studies in the
Humanities.
Van Leeuwen, T. (1999). Speech, Music, Sound. London: Macmillan.
Chapter 2
Pinpointing the Choice: Meaning and the

Search for Equivalents in a Translated Text
M. A. K. Halliday
2.1 Meaning as Choice
At a time when meaning was largely excluded from the discourse of what was
then becoming ‘mainstream’ linguistics, my teacher, J. R. Firth, perversely pro-
claimed that all linguistics was, in effect, the study of meaning. He used various
metaphors along the way; one was that of the spectrum of visible light. Meaning
was dispersed across the spectrum of language, which included the levels of
phonetics, phonology, lexis, grammar and context (‘context of situation’). The
linguist took account of all of these; and strategies for doing this constituted,
collectively, the ‘technique of semantics’.
Firth’s throwaway examples tended to obscure the message: when he said
that ‘part of the meaning of a Frenchman is to sound like one’ it made the
observations seem trivial. Firth’s point was that all the strata of language are
implicated in the making of meaning. This is the nature of the semiotic pro-
cess, and of the kind of system that is associated with such a process: a realiza-
tional system, one whose working parts are related not as cause and effect but
as signified and signifier, or (as I prefer to generalize it) as value and token.
In a sense what binds a theory to its applications is this relationship of value
and token: a theory is made manifest, or ‘realized’, in the processes of being
applied. If all aspects of the phenomenon that is being theorized – in this case,
language – are implicated in the making of meaning, then any application of
the theory will also involve some kind of operation with meaning. Of course,
‘applications’ of a theory will vary in where their energies are directed; some
are what we would probably think of as ‘practical’, such as teaching foreign
language skills to a class of technology students, while others will be of a research
nature, such as the linguistic analysis of literature, where we try to explain how
a text, or a canon of texts, comes to carry a particular value. But all are based on
the premise that the task is in some way or other one of working with systems of
meaning.
Given the demands we make on language, in almost everything we do, it is

not surprising to find a great range and variety of activities where a theory
of language can be of use: where it can contribute to formulating and under-
standing a problem, and may sometimes even point towards a solution. One
recent publication which gives an overview of more or less the whole of this
extensive territory is the two-volume ‘Continuing Discourse on Language’,
edited by Hasan, Matthiessen and Webster and published by Equinox (2005,
2007). This is a rich source of information about the scope of an ‘appliable’
theory of language. I use ‘appliable’ because ‘applied linguistics’ has come to
function in some kind of opposition to ‘linguistics’ (meaning ‘linguistics
proper’, or theoretical linguistics), whereas I want to reject that opposition;
I want a single concept which unifies the two. And I don’t say ‘applicable’
because that suggests ‘applicable to’ some particular sphere of activity or other,
whereas I want a general term which gives the sense of something that is capa-
ble of – having evolved in the context of – being applied.
My own way into linguistics was, in fact, as a search for something to think
with when faced with certain fairly specific tasks. One was my work as a foreign
language teacher, teaching what was to the learners – as it had been to myself
not long before – a seemingly rather exotic foreign language: teaching Chinese
to (largely monolingual) speakers of English. Another was my work in the
British Communist Party, as one of a group searching for a marxist, or at least
marxism-compatible, linguistics: one which would give value to what were at
that time undervalued languages and varieties: spoken as versus written lan-
guage, non-Indo-European languages especially those emerging as national
languages in former colonies, minority languages, so-called dialects as opposed
to ‘standard’ languages, and so on. Another task came from my love of English
literature: it seemed obvious to me that literature was made of language and so
must be appreciated as language, and I wanted to find out why some texts were
so effective, so compelling and so timeless. And then, fourth, I was involved in
a very early machine translation project, at Cambridge in the late 1950s; this
had a great effect of focusing the mind on how the whole traffic system of
language actually worked. So I really needed a bag of tools that would let me
come to grips with this varied set of language-grounded tasks.
If there was one motif that emerged as salient in all these different contexts,
it was that of choice: language as meaning, and meaning as choice. In teaching
a foreign language, one was always guiding the learners through networks of
choices, opening up – or helping them to open up – an expanding range of
meaning potential, increasing the delicacy of the choices they were making as
they went along. One of my own constant questions as a language learner was,
‘What’s the difference between . . .?’; as a teacher I had to field this question all
the time. In translation, whether human or mechanical, the basic problem for
the translator is the problem of choice – as is the decision of a writer whether
to prefer this form of expression over that one. But these are just the occasions
where the choice is, or can be brought, under focus of attention, as we can see
Pinpointing the Choice 15
in poets’ notebooks, or in think-aloud protocol records of translators – they are

choices that are made consciously. In fact, all use of language is the exercise of
choice; most of the time the choosing remains ‘unconscious’ – that is, below
the level of our conscious attention and awareness. It is nonetheless a process
of choice.
If we look into some of the other domains where linguistics is being ‘applied’,
the principle of choice is always likely to be in the foreground. In clinical lin-
guistic work, where one has to identify and to track disorders such as aphasia
arising from dementia or from some trauma such as a stroke, the sufferer may
partially lose access to their normal range of lexicogrammatical choice; lexi-
cally, for example, being unable to retrieve the word tennis, they will search out
a superordinate term like ball game, or a cohyponym such as netball; or they may
face a more complex dislocation of systems within the grammar, morphological
or syntactic (impairment of time systems, for example), which are not easy to
pinpoint with any certainty. Most sufferers will work hard to remain in the dis-
course – this may be unconscious, or it may be a process requiring considerable
conscious effort; but in either case it is the search for a meaningful alternative
choice. The therapist seeks to locate the disorder within the overall system of
the language. In the discourses of psychiatric treatment, such as conversational
therapy, the therapist will often attend not only to the choices made by the
patient but also to those he is making himself, and perhaps make some linguis-
tic analysis of the discourse to track the course of the therapeutic encounter.
In a forensic context, where for example a linguist is being asked as expert
witness whether a particular confession, or a purported suicide note, could have
been written by a particular individual, the question is one of determining
whether the choices made lay within that individual’s meaning potential. My
former colleague John Gibbons was consulted at some length on the selection
of the most appropriate choice of wording for the caution used by the police
on making an arrest. Matching specimens of handwriting is perhaps the
furthest removed from a concern with choice in meaning, although even here
it will be assumed that the writing is the realization of some meaningful text
(Gibbons, 2003).
All use of language is a process of meaningful choice; and many of the appli-
cations of a linguistic theory depend on bringing out the specific choices that
have been made, or that need to be made, in particular situational and textual
contexts – in other words, on locating them in their function in the overall
system of the language. This is what I meant by ‘pinpointing the choice’ in the
title of this chapter. One form of activity in which the process, or processes, of
choosing will be most clearly foregrounded is translating from one language to
another; since I have lately been engaging in a translation task, let me draw on
this as a repository of instances of different kinds of choice.
My text is one from what is now called the ‘culture industry’ – or should be;
in fact it often gets called ‘cultural industry’ instead, and that is itself an inter-
esting example to consider: ‘culture’ here needs to function as Classifier in the
nominal group, not as Epithet, and the adjectival form cultural is ambiguous
in this respect, and could be interpreted in the somewhat oxymoronic sense of
an industry which is itself of cultural value. The discourse of the culture indus-
try includes varieties like museum texts (Ravelli, 2006), tourist guides and
brochures, promotions and programmes of events like concerts or sporting
ventures, and so on. The text I am working on is a tourist guide to the Chinese
city of Guilin – or rather the city and its surrounding region, which is one of
China’s most outstanding beauty spots. There was already an English version of
the guide, and my original undertaking was just to polish up the English trans-
lation as a token to the friend who gave it to me, who happens to be advisor and
consultant to the Guilin Tourist Bureau.1 I had thought this would be a fairly
simple task, but it turned out that what was more appropriate was a new transla-
tion from the original Chinese text (which was both lavishly produced and very
beautifully illustrated).
I shall assume that we can discuss translation in terms of equivalence and
shift, taking these terms from Catford’s splendid little book on translation
(Catford, 1965) but more particularly as they are expounded and theorized by
Matthiessen in the book edited by Erich Steiner and Colin Yallop (Matthiessen,
2001). If x in the source language has been translated by y in the target lan-
guage, we ask whether the two, x and y, are equivalent, or there has been a shift.
Of course, this opposition will always be ‘with respect’: it is the job of a linguistic
theory of translation to specify and explain the ‘respects’ in which we may estab-
lish equivalence and shift, and this is what Matthiessen proceeds to do. These
are what he refers to as the environments of translation.
Matthiessen writes (2001, p. 78), ‘I shall assume that translation equivalence
and translation shift are two opposite poles on a cline of difference between
languages’, from ‘maximal congruence’ to ‘maximal incongruence’. ‘The gen-
eral principle’, he adds, is that ‘the wider the environment of translation, the
higher the degree of translation equivalence; and the narrower the environ-
ment, the higher the degree of translation shift.’ This is the principle of contex-
tualization: the ‘widest’ environment is that in which the text is ‘maximally
contextualized’ – and therefore, by the same token, is likely to be ‘maximally
effective’ (ibid., pp. 74–5).
Matthiessen’s ‘environments’ are defined by the dimensions along which every
human language is organized: stratification, rank, instantiation, metafunction,
delicacy and axis. I shall refer mainly to the first two of these, stratification and
rank, in looking at some examples from my bilingual tourist text (actually
quadrilingual, because it includes also versions in Korean and Japanese). What
I am seeking to do is to identify points of translation shift in explicit linguistic
terms, so as to be able to discuss the issues that are involved when we consider
possible alternatives. This means locating particular instances in their contexts
in the systems of English and of Chinese – primarily, here, in terms of their
location in stratum and rank.
This exercise is related to, although not identical with, the ‘error analysis’
that was popular in language teaching, or rather in linguistics when applied to
language teaching, in and around the 1970s. Error analysis went out of fashion
fairly quickly, because it turned out that there were nearly always a number
of different ways of repairing a linguistic error. This made it unpopular with
teachers; but it is precisely this feature that makes it interesting to linguists, and
it is a useful exercise to perform with students of linguistics since it helps them
to become conscious of the various dimensions along which languages are
organized.
A translation shift is not the same as an error. Indeed there will almost always
be shift on some dimension or other, because there is seldom total equivalence
between choices in two languages; the translator shifts here to gain equivalence
there, according to the value inhering in equivalence of different kinds in
the nature of the task in hand. Of course there are simple errors, mistakes
in the target language (we will start with one or two of these). But while these
also can be usefully pinpointed, they lie outside the range of (normal) transla-
tion equivalence.
In the second part of this chapter I will discuss some examples taken from
one paragraph of the English translation of my tourism text. Text A below is
the Chinese original; Text B is the English translation. The two were clearly
identified as being equivalent, as they appeared one beneath the other in a
single column (not all portions of the text were displayed as matching para-
graphs in this same way). I shall try to keep in focus the concept of choice, see-
ing each example as the output of choices made at some particular locations in
the target language. We cannot know, of course, what steps the translator may
have traversed in arriving at the published translation; that is not the issue here.
What I am trying to do, in ‘pinpointing the choice’, is to use the analytic tools
of linguistics, and particularly perhaps of grammatics, to examine the signifi-
cance of alternative renderings for a reader of the English text.
In the table of examples (Appendix 1), each of the four centre columns takes
a step towards pinpointing the problem in linguistic terms. The first column,
headed ‘point at issue’, gives an informal characterization of the instance that
is being highlighted: ‘English error’, ‘unmotivated shift’, ‘unnecessary word’,
‘unclear meaning’, and so on. The next three columns locate the point at issue
on each of three dimensions in the ‘architecture’ of language: the stratum
(phonetics, phonology, lexicogrammar (syntax, morphology, lexis), semantics,
context); the rank (in syntax: morpheme, word, group, phrase, clause or some
complex of any of these; in lexis: lexical item, collocation); the metafunction
(ideational (experiential, logical), interpersonal, textual). On all these dimen-
sions, more than one location may be involved (for example, rank: clause/ver-
bal group). The final column suggests how an alternative choice might be made
which would improve the effectiveness of the translation.
To make this discussion more accessible to those who do not read Chinese,
I have added a transcription of the paragraph in the roman alphabet, using the
Hanyu Pinyin (the authorized spelling of standard Chinese), including the
tone marks as an essential component. I have not thought it necessary, or even
desirable, to add an interlinear gloss on the Chinese words, because these will
be explained in the discussion of the examples in which they occur. There

too I have used the Pinyin transcription, as well as the Chinese charactery.
For the Chinese characters, rather than the simplified I have chosen the full
(elaborated) variants, since these are used in Hong Kong and Taiwan (and also
Japan), as well as being easily accessed electronically on the mainland.
2.2 Some Examples of Choice in Translation

In its cultural context, as a tourist guide, the text describes tourist attractions
and other amenities of Guilin and its surrounding country, while at the same
time encouraging visitors to come and helping them to spend their money.
Much of the Chinese original is descriptive and informative; but it is generously
flavoured with clichés, ritually praising the environment in lyrical and romantic
terms. The English reader is not familiar with them and does not expect them
in a tourist guide. The Chinese translator (author of the English version) in fact
left many of them out; but then he left out many other things as well.
One could try to look out some suitable quotations from English poets,
such as perhaps Tennyson or Wordsworth; but that would be just another kind
of translation shift, and in general I have also left them out. This is a shift in
semantics (and also of course in lexicogrammar) designed to achieve equiva-
lence of the text as a whole, in the sense of operating with the same function in
the same context as the original. Catford regarded this as one of the possible
types of translation. To try to achieve equivalence at a lower stratum, even that
of semantics, in a case like this would compromise the effectiveness of the text,
introducing a component of exoticism that is entirely absent from the original.
The problem of the unwanted exotic is a familiar one in translation, which also
needs to be pinpointed with some care. It is related to, though not identical
with, the problem of the unintended humour: passages in a translated text that
a reader is likely to find funny. Here are two examples from another tourism
text, this time a small brochure devoted to just one particular location:
As a matter of fact, this temple is a rock cave. Its ceiling is made of rock and,
therefore, it is also called ‘one piece of tile’.
(The) Concert Hall was built in 1989. The design of the hall is so good that
in every seat music can be heard.
Here the humorous effect comes from the semantic incongruity: the anomaly
between the stated cause (therefore; so good that) and the lack of any perceived
causal relation: why would you call something a ‘piece of tile’ if it’s made of
rock? And music being audible from every seat is a minimal requirement of a
concert hall, not something to be picked out as a claim for special merit.
When we talk of ‘pinpointing the choice’, in the theory and practice of trans-
lation, this means locating, within the systems of the two languages concerned,
the moments of equivalence and shift that come to our attention. These may, of
course, be almost any moments in any pair of texts that are related as source
and target texts in translation, since equivalence on all dimensions is rather
improbable. At the same time, the concept of translation, as process and as
product, depends on the search for equivalence and the assumption that equi-
valence can be achieved in at least certain respects. There will always be a trade
off, such that we are able to say that, in the given context, the greatest value
is carried by equivalence of this or that particular kind: usually in some combi-
nation of stratum and metafunction. (My first published translation was an
English rendering of a Chinese song; since this was done for performance at
a recital, it had to fit the rhythm and spirit of the music.) The translator will
give these forms of equivalence priority, in making choices, and accept the
resulting shift in other locations. All work of translating is the exercise of
choice, conscious or unconscious. But then, so is every other performance of
language.
When we talk of an ‘appliable linguistics’, this means a way of engaging with
language that is theoretically robust and at the same time serviceable – capable
of being put to use in addressing a range of problems and tasks. These two
requirements support each other, because the robustness of the theory derives
from its long association with activities where language plays a central part –
including some educational (e.g. language across the curriculum), some
computational (natural language processing) and others; while its appliability
comes from its consistent grounding in theory – in this case, a theory offering
functional explanations both for the system of language and for each instance
of language use, and locating any feature of language in its paradigmatic context,
as selection from a multidimensional meaning-making resource. Translation is
one area where it is valuable to be able to reflect on what we are doing, and to
explore the meaning potential of a language – of two or more languages as they
are brought into contact – in an explicit and recoverable way. Translators often
protest that they find little or no use for linguistics; so it is perhaps a challenge
to what seeks to be an ‘appliable’ kind of linguistics to put it to work in this
domain.
2.3 Texts
Text A Chinese original, (i) in characters, (ii) in Pinyin romanization
(i) Ёᖗऔඳ᱃咲ᰃҹⱒ䞠ⓧ∳⚎Ё䓌㍿ˈ䔏ᇘܽኌ㨫ৡⱘ㞾✊乼‫ܝ‬ǃҎ᭛
᱃㾔ˈᾟ៤Ḗᵫቅ∈ⱘ㊒㧃Ǆ݊Ё᳝ҹⓧ∳乼‫⚎ܝ‬ҷ㸼ⱘ5㰩೟ᆊ㋮4A
᱃औ˗᳝Ϫ⬠᳔໻ⱘ䲩ลࡉ԰➳ഄᛮ㞾ῖ೦˗᳝Ϫ⬠᳔໻ⱘ㰢❞仞Ⅺ෎
ഄ䲘Ể❞㰢ቅ㥞˗Ꮦऔⱘܽ∳ಯ␪᳈ᰃᾟ៤њ‘ϔ∈ᢅජ⌕’ⱘජᏖ໻݀
೦Ǆ݊䭧गዄⶫゟˈ乼ᚙ㨀。ˈҸҎ⼲‫ޱ‬ᔶ䞟ˈ⌕䗷ᖬ䖨Ǆ
(ii) Zhōngxīn qūyù jǐngdiǎn shì yǐ bǎi lǐ Líjiāng wéi zhōng zhóuxiàn, fúshè
liǎng àn zhùmíng-de zìrán fēngguāngǃrénwén jǐngguān, gòuchéng
Guìlín shānshuǐ-de jīnghuá. Qízhōng yǒu yǐ Líjiāng fēngguāng wéi
dàibiǎo-de 5 chù guójiājí 4A jǐngqū; yǒu shìjiè zuìdà-de diāosù chuàng-
zuò yíngdì Yúzì Lèyuán; yǒu shìjiè zuìdà-de hǔ xióng yǎngzhí jīdì
Xióngsēn Xióng Hǔ Shānzhuāng; shìqū-de liǎng jiāng sì hú gèng shì
gòuchéng-le ‘yī shuǐ bào chéng liú’-de chéngshì dà gōngyuán. Qíjiān
qiān fēng chùlì, fēngqíng wànzhǒng, lìng rén shén níng xíng shì, liú lián
wàng fǎn.
Text B English translation given in the tourist guide
The scenic sports of central area is taking the 100 miles of Li River as central
axes, covering the famous natural landscapes and human-culture scenes on
both banks which form the soul of Guilin’s landscape, including five national
level 4A spots in which the scene of Li River is representative; the biggest
sculpture creation campsite in the world – Yuzi Paradise; the biggest tigers &
bears raising base – Xiongseng Tiger & Bear Villa; the Two-River and Four-lake
scenery area form a big city-park of ‘a river flowing around the city’. In this
area, there are thousands of upstanding hills and a lot of charming attrac-
tions which make your pleasure and forget to return.[sic, passim]
Notes
1
My warmest thanks to my friend Bao Jigang, Professor of Geography and Tourism at Sun
Yat-sen University, for bringing this book to my attention – and for presenting my wife and
myself with a handsomely bound copy on the occasion of our visit to Guilin.
The Chinese text and the examples were set up for me by Joe Chen (Chen Jiansheng).
I am most grateful for his assistance.
(Continued )
Appendix 1
AMahboob_02_Final.indd 23 5/25/2010 11:24:52 AM
References
Catford, J. C. (1965). A Linguistic Theory of Translation. London: Oxford University
Press (Language and learning series).
Gibbons, J. (2003). Forensic Linguistics. Oxford: Blackwell (Language in society
series).
Hasan, R., Matthiessen, C. M. I. M. and Webster, J. (eds) (2005/2007). Continuing
Discourse on Language, Vols 1 & 2. London and Oakville: Equinox.
Matthiessen, C. M. I. M. (2001). ‘The environments of translation’, in E. Steiner
and C. Yallop (eds), Exploring Translation and Multilingual Text Production: Beyond
Content. Berlin: Mouton de Gruyter, pp. 41–124.
Ravelli, L. J. (2006). Museum Texts: Communication Frameworks. London: Routledge.
Chapter 3
Appliable Linguistics and English Language

Teaching: The Scaffolding Literacy in Adult
and Tertiary Environments (SLATE) Project
Ahmar Mahboob, Shoshana Dreyfus,
Sally Humphrey and J. R. Martin
3.1 Introduction
This chapter introduces the Scaffolding Literacy in Adult and Tertiary Environ-
ments (SLATE) project1, which is an ongoing action research project that aims
to scaffold the academic literacy skills of students from a non-English-speaking
background (NESB) studying at an English medium university, via the use of
online learning environments. The SLATE project uses an appliable linguistics
framework in that it builds on and extends a theory of language in response to
a real-world issue (Halliday, 2006). The SLATE project draws on Sydney School
theories of genre (Martin, 1993; Kress, 1993; Martin and Rose, 2008), register
(Halliday 1991; Martin, 1992) and other dimensions of language and semiosis
within the Systemic Functional Linguistics (SFL) tradition. In this chapter, we
outline the broad principles and knowledge that this project is based upon,
share some of the project’s achievements (with examples from one of the
courses supported by the project) and indicate some future directions for the
project.
3.2 Background to the SLATE Project
The SLATE project evolved from a need identified by staff at the City University
of Hong Kong (henceforth CityU) – namely that students at this institution are
not always successful in producing appropriate academic texts in English (this
issue, of course, is not only pertinent to CityU but of much broader concern,
both nationally and internationally). The staff at CityU were concerned that the
students’ lack of ability to write effective texts in their discipline areas had nega-
tive consequences in terms of their future employment and career opportunities.
They therefore started considering ways of providing students with language
support while enrolled in their subject classes. In the initial phase of the proj-
ect, called the Language Companion Course (LCC)2, students were provided
with a language coach to whom they submitted drafts of their written coursework
online; the coaches provided feedback on these drafts using an online comment
bank. While helpful for the students, this model is limited in that it gives the
students no instruction on how to go about their writing tasks before they set out
to write – the help only came ‘after the horse had bolted’, so to speak.
Given that within the Sydney School pedagogy much of the support for
writing is provided prior to students having to write independently, the nature
of the project started to change once the University of Sydney (henceforth
USYD) became involved in early 2008. The changes introduced included sup-
port for both CityU students and their language coaches. For the students, the
USYD team built in some ‘frontloading’ support for students prior to writing,
that included both reading support (where applicable) and a modelling stage,
where students were provided with a model text that explicates the features of
the texts the students were required to write. For the coaches, the USYD team
developed an extensive training programme, which expanded the coaches’
own metalinguistic resources, using teaching materials such as the 3×3 matrix
(see Humphrey, Martin, Dreyfus and Mahboob, this volume). The aim of the
training was to extend the language coaches’ gaze beyond word-level structural
units to considerations of context and unfolding of meanings across texts.
Finally, the USYD team developed an expanded feedback/consultation process
and the development of assessment criteria and protocols. Some of these pro-
cesses and materials will be presented in the section below in relation to one
linguistics course supported by USYD SLATE.
3.3 Theoretical Underpinnings to the SLATE Project
The SLATE project is the fourth generation of Sydney School work on genre
pedagogy based at the University of Sydney. The first generation, the Language
and Social Power project, and the second generation, the Write it Right project,
focused on the literacy demands of primary and secondary schools respectively
(Martin, 2000). These projects identified the literacy practices crucial for suc-
cess in school and made explicit the language resources needed to enact these
practices, both for teachers and students. This work was supported by a number
of international literacy educators, particularly those concerned with develop-
ing critical literacy practices for students from low socio-economic status (SES)
and NESB (e.g. Schleppegrell and Colombi, 2002; Gee, 2000; Kalnin, 1998;
Macken-Horarik, 1996; Schleppegrell, 2004). The third generation of work on
genre pedagogy was developed by David Rose3 and focussed on the integration
of reading and writing across sectors. The SLATE project adopts the principles
developed and tested in the earlier projects and applies them to the tertiary
education context in online learning environments.
English Language Teaching 27
In continuing to develop a pedagogy of empowerment, the SLATE project

incorporates aspects of genre theory (Martin and Rose, 2008), sociology of
education (Bernstein, 2000), and socio-cultural theory (Vygotsky, 1978). Within
genre theory, genres are defined as ‘staged goal-oriented social processes’
which function in society as institutionalized discourse (Martin and Rose, 2007;
Martin and White, 2005). A central idea of using this understanding of genre
in education, especially in teaching literacy and writing, is that learners of all
socio-economic and cultural backgrounds must be taught these genres explic-
itly in order to succeed in society. Genre pedagogues argue that if standard
genres are not taught effectively to students, they will be unable to produce
texts that are valued in academic disciplines and therefore not be able to fully
benefit from educational experiences.
Genre pedagogies have drawn on SFL theory, which views language as a social
semiotic system (Halliday, 1978), that is, a resource for making meaning in
social context from which notions of ideology and power are inseparable
(Eggins, 2004; Martin and Rose, 2007). In SFL theory, language as a social semi-
otic system is realized on five different levels of abstraction, which have been
termed strata: phonology/graphology, lexicogrammar, discourse semantics,
register and genre. The most basic resources for meaning-making are phono-
logical or graphological units. At the stratum of lexicogrammar, the units of
phonology and graphology are realized as words and structures (Martin and
White, 2005, p. 9). At the discourse semantic-level meanings are created across
phases of text as a whole, rather than just within clauses. Context, modelled as
register and genre, stands at the higher levels of abstraction. In educational
linguistics, inspired by the Sydney School, register has been interpreted as
referring to context of situation and genre to context of culture. Register real-
izes genre, and is organized metafunctionally as field, tenor and mode. Field is
concerned with the nature of social action, viewed as institutional practice;
tenor refers to the relationship among participants, their status and affinity;
and mode refers to the role language plays in organizing ideational and inter-
personal meanings according to the textures different channels of communi-
cation afford. In a model of this kind it is the responsibility of genre to describe
how field, mode and tenor variables combine and are phased into purposeful
social semiotic processes (Martin and Rose, 2008).
In academic context/texts, ideational meanings (realizing field) construe
technical and specialized disciplinary knowledge. Interpersonal meanings
(realizing tenor) are constructed in relatively distanced and objectified ways
to position writers authoritatively in their field of knowledge. Textual meanings
(construing mode) are used to package information into comprehensible
pulses of news in meaning-heavy genres (Eggins, 2004; Martin and Rose, 2007;
2008). The strata and metafunctions can be mapped on to each other in dia-
grams such as that in Figure 3.1.
Also informing Sydney School literacy practices are Bernsteinian concerns
that individuals from a lower SES may not have the orientations to meaning
Figure 3.1 Strata and metafunctions in SFL
to create texts in ways that are valued in educational contexts (Martin, 1999).
Genre theory suggests that a pedagogy of empowerment must make visible
these genres of power and explicitly teach people who come from disadvan-
taged backgrounds ways of meaning that match the social expectations of
their readership. The pedagogical implications of genre theory encompass
Vygotskyian notions of scaffolding4 (Bruner, 1966; Applebee and Langer, 1983).
Vygotsky’s work, in addition to studies on child language development (such
as Halliday, 1975; Painter, 1984), indicate that successful learning happens
when teachers/carers first model the practices that need to be learnt, then
jointly construct these with the learners, before allowing learners to attempt to
carry out a task independently. For learning within institutional contexts, the
Sydney School has formalized this set of scaffolding practices as ‘The teaching–
learning cycle’ (Rothery, 1996), illustrated in Figure 3.2.
Rothery’s (1996) model implements the idea that knowledge is constructed
in a social context and that in order to successfully gain control of writing and
reading, learners need to be led through cycles of deconstruction, joint con-
struction and independent construction, while simultaneously building their
understanding of the field. In doing so, they move towards a critical orientation
to, and control of the skills, knowledge and language that is required within
specific genres and valued in particular social contexts. The teaching–learning
cycle requires that in the deconstruction stage, the teacher first models the
text and, in thus deconstructing the text, enables students to understand its
purpose, structure and important language features. Following this is the joint
construction stage, where together with the teacher, who provides the leader-
ship and guidance, students write a text of the same type on another topic.
Finally, and after successfully scaffolding this writing process, learners are given
the opportunity to write independently. Throughout this sequence the field
Figure 3.2 Teaching–learning cycle (Rothery, 1996)
(i.e. disciplinary knowledge in academic contexts) needs to be built up. In the

academic context, this occurs primarily through lectures and reading, as well as
discussions. Reading strategies such as those advocated by Rose (2005a; 2005b)
have adopted and expanded the teaching–learning cycle in ways that are rele-
vant not only to junior and secondary school contexts but tertiary contexts
as well.
3.4 Developments and Achievements of the SLATE Project
The contributions of the SLATE project can loosely be divided into support for
language coaches and support for students. As mentioned above, the support
for the coaches included an extensive training programme, the development of
the 3×3 matrix and scaffolding for an expanded feedback/consultation process
involving additional assessment criteria and protocols. The support for the stu-
dents consists of the implementation of one stage of the teaching–learning
model that supports students prior to writing – including both a modelling
stage and an expanded feedback stage that build in supplementary modelling,
and in some cases reading support.
3.4.1 SLATE support for language coaches

As one initiative the SLATE project ran an 18-hour training programme for
language coaches to build shared understandings about language, and to
encourage a particular approach when working with the CityU student texts.
This approach, as was discussed in a previous section, was based on genre and
register theory. Using an understanding of SFL, the SLATE team developed
a 3×3 matrix that mapped a simplified model of strata (genre, discourse, and
grammar & expression) in relation to the three metafunctions (ideational, inter-
personal and textual); for a detailed introduction to the 3×3, see Humphrey,
Martin, Dreyfus and Mahboob, this volume. In training the language coaches,
the 3×3 matrix was used to ensure consistency in the coaches’ approach to
reviewing student texts and providing feedback. By using the 3×3 matrix, the
coaches were able to distinguish meanings according to strata and metafunc-
tion and this enabled them to move beyond a preoccupation with the low-level
grammar and expression mistakes that are common in texts written by NESB
students and engage with the texts from a range of perspectives: from the whole
text down through the paragraph level to the clause and on, in this contextual-
ized manner, to problems with spelling, word choice, use of articles, subject
verb agreement, punctuation, and so on. At the whole text level, for example,
coaches were trained to examine whether the student’s text followed the
required structure which had been modelled: did it move through the appro-
priate stages, and within those stages, through the appropriate phases? The
training aimed to systematize the coaches’ gaze as much as possible so that all
coaches were using the same understandings about language.
The 3×3 matrix was also used to develop an assessment rubric. The assess-
ment criteria used in this rubric was developed to reflect both a stratified and
metafunctional view of language. The assessment rubric consisted of three
main criteria:
z Criteria A: Purpose and structure of text (focus on the whole text level);
z Criteria B: Development of meaning across paragraphs (focus on discourse
semantics and phases); and,
z Criteria C: Grammar and expression (focus on clauses, sentences and other
structural and graphological features).
Each of these criteria addressed a number of language features that were

needed to construe specific meanings (across the three metafunctions). Stu-
dent texts were given one of the following ratings: ‘shows excellent control’,
‘shows good control’, ‘shows fair control’, ‘shows poor control’ and ‘not appli-
cable’. The language coaches filled in a rating sheet as they read through the
student’s text and then used these ratings to help identify the relative strengths
and weaknesses of the text and to determine what linguistic issues would be
focused on in providing feedback to the students on that particular draft of the
assignment. For each draft, coaches focused on only 2–3 linguistic issues, and
provided in-depth feedback on these. The coaches were encouraged to start

their feedback addressing Criteria A: Purpose and structure of text, as it is rec-
ognized that the first step in having control over a genre is getting the purpose
and structure right.
In addition to introducing the language coaches to the linguistic theories
adopted in this project, they were also introduced to the feedback structure
that the SLATE team has been developing since the project began. This feed-
back structure was developed based on an analysis of successful feedback
(i.e. feedback that was taken up by the students) in the previous semesters.
The feedback structure includes 6 stages (3 required and 3 optional) and each
of the stages includes 1 or more phases. The current feedback structure used
in this project is presented in Figure 3.3 with examples from two sets of feed-
back, one containing the required stages only, and one with both the required
and the optional stages.
Figure 3.3 outlines the structure of the feedback given to the students by the
language coaches. It includes three required stages: Orientation, Feedback and
Encouragement. The Orientation stage includes two required phases: purring
(greeting and positive comment) and previewing (outlining the focus of the
feedback provided). The Feedback included three required phases and one
optional phase. The required phases of the Feedback stage were: stating the
problem, explaining why it is a problem and suggesting how to fix it. In addi-
tion, language coaches could also suggest alternatives as part of the Feedback.
In the Encouragement stage, the final stage of providing comments on stu-
dents’ drafts, the language coaches ended their consultation on a positive
note. In addition to the three required stages, language coaches sometimes also
included three optional stages: Recap, Recommendations and Expanded
Explanation. The purpose of these optional stages was to enhance the feedback
provided in different ways: by summarizing and identifying priorities for the
student to address in their next draft/assignment; by suggesting activities and
resources for students to work on independently; and by providing additional
explanations and examples.
Having a consistent and structured approach to providing feedback on
student drafts contributed to the success of the project. Among other things, it
assisted students to identify key issues in their writing and provided ample con-
textualized feedback to help them understand and learn from their mistakes.
The feedback structure also helped standardize and bring some consistency to
the feedback provided by different language coaches and, given the large num-
ber of coaches/students in the programme, ensured quality maintenance.
3.4.2 SLATE support for students – modelling

Another contribution of the SLATE project to the LCC programme has been
the introduction of a modelling stage to support students prior to writing. The
LCC was initially set up to provide feedback to students on the texts that they
STAGE PHASE EXAMPLE 1 EXAMPLE 2
32
Orientation Purr: warm greeting Dear TM, Well done on submitting your Dear GY, Well done on considering
Statement of strengths first draft for your second assignment. my feedback from draft 1.
You have a very good sense of You have certainly made a good
Preview: states what structure in terms of staging with effort in strengthening your work
feedback will focus on an appropriate beginning, middle as a comparative text. You have
and end. Well done. There are still made a considerable improvement,
some things you can do to improve especially in creating a proper
your work on a paragraph level. introduction and conclusion and
Therefore my comments this draft refraining from using personal
will deal with paragraph structure pronouns. My comments this time
and topic sentences. I will also make will address ways you can help the
some remarks about appropriate reader to understand your points. I
linking words, level of formality and will focus on ways you can improve
referencing. your structure on a paragraph level,
as well as touching on effective use
of signposting.
Feedback Statement of problem You have made a good attempt here The two paragraphs above seem
Explanation of why it’s a at a linking phrase to introduce a to describe some words which are
problem comparison. However, the expression difficult to classify. Try to group them
‘On the contrary’ is not quite what you together with a clear topic sentence.
Suggestion for how to need here because you are making This will alert the reader that there
remediate (including a specific contrast, so why not use is a new idea about to be discussed.
directions to Blackboard something like ‘Contrastingly…’ or ‘In (NB This is one of a number of points
resources and model contrast…’ raised by the tutor)
text, and alternatives) (NB This is one of a number of points
raised by the tutor)
(Recap) Summary of what student This is a good comparative essay,
needs to work on in next Y. Keep my comments in mind when
draft working on your final submission,
particularly the way you can arrange
your sentences in a logical sequence
to move the reader through your
ideas in a sensible way.
(Recommendations) Suggestions for activities Your essay could also be
or general ideas to strengthened by using appropriately
improve student’s writing formal language suited to an
academic essay. It can be tempting
to write how we speak. This is one of
the challenges of academic writing,
but you must always keep this sense
of formality in mind and you will
improve. Your use of the personal
pronoun ‘I’ is too informal for an
academic essay. Personal pronouns
like ‘we’ or ‘I’ are discouraged in
English Language Teaching

academic writing. This is partly
because the focus should not be
on the author, but should remain on
the subject matter. The exercise on
passive voice should help you here.
(Expanded Provision of more Like I said in my feedback last
Explanation) detailed explanations time, you need to effectively
with examples integrate these definitions into your
paragraphs. As it is, they are not
functioning as headings nor as part
of the body. Try to integrate them
with linking words and introductory
phrases.
Encouragement Praise & well-wishing for This is a good linguistic commentary, Good luck and we look forward to
their next draft TM Please take my comments into receiving your work!
account when working on your next
draft and your work will improve.
Good luck, L
Figure 3.3 Feedback structure with required and optional stages. Note: Stages/phases in brackets are optional.
33
wrote for their various courses using an online comment bank. The comment
bank included a list of common problems faced by English as second language
(ESL) students and could be inserted in the students texts as a hyperlinked
numerical item. When students clicked on the hyperlink, the link took them to
a web page that provided a description of the problem, examples of correct and
incorrect use of the feature and sometimes an external link to additional
resources/information about the problem. This use of the comment bank as
the primary form of feedback did not require any support material for students
to use prior to their writing. The support given to students took place once the
students had completed an independent construction of a text. In moving from
LCC to SLATE, the aim was to shift our support forward to inform students
about the kind of text they were being asked to write before they were required
to write it independently. This meant that students would come to the writing
process more informed about what it is they are required to write, instead of
writing blind in an uninformed manner and thus producing texts that needed
wholesale ‘repair’. The aim of providing model texts is not only to guide stu-
dents in their writing, but to also begin to build a shared metalanguage that
would develop students’ knowledge and understandings about language. This,
in turn, would mean that the support students could be given further on in the
programme could assume a modicum of a shared understanding about lan-
guage. The ultimate aim is to provide a spiral curriculum for writing (Bruner,
1960) whereby subsequent teaching can contribute to, build on and then
assume an ever-increasing metadiscursive knowledge base.
For the introduction of the modelling stage, the SLATE team developed
annotated and detailed model texts of the kinds the students need to write. In
recognition of the specificities of knowledge construction in each genre and
field, different models are provided for different assignments and different
subjects. The models show the students the kind of staging and language
requirements they need to master in order to successfully complete their writ-
ten assignments. These models are also accompanied by notes that explain the
model text, including its purpose, structure and language features. The context
of the online environment at present means that while model texts are pro-
vided, there is no teacher-led deconstruction of the text, and the students have
to navigate through the modelling/deconstruction stage independently5.
The chapter will now describe one course in some detail to illustrate how the
SLATE project operated. It will also provide some evidence of the effectiveness
of the approach used in this project.
3.5 Supporting First-year Linguistics Students
Over the three semesters of its involvement in the project, the USYD SLATE
team has supported approximately 1100 students in 11 courses. One of these
included a first-year introductory course in linguistics. Before the SLATE inter-
vention, students taking this course were primarily assigned problem sets that
required them to carry out linguistic analysis, but not to write up or describe
their analysis. According to staff, one key reason for this was that students’
language proficiency was not considered sufficient to write such discussions.
The students were, however, required to submit a final-year research project
that required such writing. The SLATE team, in discussions with the relevant
staff, developed two new assignments for this course that required students to
first carry out an analysis, based on their lectures, course readings and tutorials,
and then write-up their results. The genre of such texts has tentatively been
called an interpretation. Furthermore, since the assignments required students
to compare two linguistic phenomena, the genre can be subcategorized as a
comparative interpretation. The purpose of this kind of text is to compare and
contrast two aspects of language or interpret the language differences in two
(or more) different texts using a particular linguistic frame. Students were
given annotated notes and models for working on these assignments.
In addition to working with the lecturer to set up the assignments, the
SLATE team also shared the rationale for the assignments with the students.
The Information Sheet sent to the students informed them of the SLATE (LCC)
programme and included a listing of the assignments that were supported by
the SLATE team, the important dates for these assignments, the rationale for
selecting these assignments for language support as well as students’ rights and
responsibilities.
Students submitted three drafts of each of the two LCC/SLATE assignments:
Draft 1, which was based on the notes provided to the students; Draft 2, in
which students integrated the feedback given to them on Draft 1; and the Final
draft – submitted both to the language coaches and the subject teacher, in
which students responded to and integrated the feedback given to them on
their earlier writing. The turnaround time for submitting feedback or a revised
draft was 48 hours for both the language coaches and the students. This dead-
line was strictly observed and late submissions by students were not accepted.
This created a cycle of drafting and feedback for each assignment that lasted
Figure 3.4 Consultation process for linguistics course

just under two weeks. The support for students was supplemented with anno-
tated models and notes at various points during the cycle. The drafting and the
consultation process adopted for this linguistics course is shown in Figure 3.4.
Figure 3.4 represents the consultative process used for the linguistics course.
The consultative process includes a number of initiatives. In the first place,
students were provided annotated models and notes to help them understand
the linguistic requirements of their assignment. The students used these notes
to write their first draft and submitted these to their language coaches. The
language coaches provided feedback on these drafts using the SLATE feedback
practices. In addition to individual feedback, additional notes and activities
were designed and given to the students that address common problems.
Students used the individual and group feedback to revise their drafts and then
resubmitted these to their coaches. The language coaches, once again, pro-
vided individual and group feedback. The students then revised their work
one more time and submitted it to the language coaches, who wrote a final
report on the student language, as well as to their subject teachers, who assessed
the final assignment. This cycle was then repeated (SLATE supported two
assignments in linguistics, but the number of cycles can be adjusted based on
the number of assignments supported).
As was stated earlier, the SLATE team believes that it is very important that
students be explicitly informed about what it is that they need to do in order to
succeed in their work before actually being asked to write their assignments. In
the current stage of the SLATE project, this need was addressed by providing
models of writing to students. These models were annotated and supported
with notes that first aimed to focus the students’ attention on mastering the
requirements of the text at the level of genre. The purpose of a comparative
interpretation was introduced and the model text was annotated for stages, and
also phases within stages. The functions of each stage were explained with
examples from the model text showing where and how these functions were
realized. Following this, the notes demonstrated and discussed the logical orga-
nization of the paragraphs within the body of the text and highlighted the com-
parative language used to help build the model text’s arguments. The notes
then touched on some less genre-specific language issues such as providing
evidence in appropriately academic ways and backgrounding personal feelings
and experiences. The notes finished with a brief step-by-step procedural section
on how to go about writing the interpretation. These notes were used by stu-
dents as they drafted their work. Further, the language coaches also referred
students to specific sections of these notes during the revision/redrafting
phases. The first set of notes were supplemented by activities that the team
felt were needed to help students work on specific problems. For example, it
became evident in Linguistics that while students were able to produce texts
that roughly followed the generic structure of the model interpretation, the
texts were often lacking at the phase, paragraph and sentence levels, parti-
cularly in terms of thematic development. As a result, we introduced notes
focusing on choice of Theme and further annotated an additional model text

to assist students in building thematic development.
The outcome of this intervention has shown that students who use the pre-
pared notes and therefore engage with the modelling phase of the teaching–
learning cycle are able to produce well-structured texts that followed the generic
conventions. As there was no way of compelling the students to use the model
texts, it was evident that the tutor comments which pointed the student to the
model texts were powerful in supporting students to engage with the modelling
stage. This meant, however, that some students engaged with the model text
and notes only after writing their first draft. This can be seen in the following
two examples from the work of one student. The examples show the student’s
first attempt at writing a comparative interpretation and her final submission
after two rounds of feedback. The topic of the interpretation was derivational
and inflectional morphology, and the students were required to first read a
short passage and analyse the words within the passage that displayed inflec-
tional or derivational morphology, and subsequently discuss the differences
between the two, using the examples they had analysed.
Example 1 shows the student’s nascent attempt at the genre of comparative
interpretation. However, her text is fairly unstructured generically, and, as she
does not produce the required stages of the text, she also does not produce the
appropriate phases within those stages. The text is presented below without
the comments inserted by the tutor so that it can be read as the language
coach received it, that is, in toto and in an uninterrupted manner. However,
as it is important in understanding the process of support for the students,
and how the tutor uses the model to guide the students in their writing, the
tutor comments are listed below, each one contained between curly brackets {},
corresponding to numbers inserted in the text.
Example 1: Sample of first draft of comparative interpretation.

In our text book, there are two morphological processes: inflectional morphology and
derivational morphology. Inflectional morphology refers to the usage of words accord-
ing to the syntactic rules by adding inflectional morphemes to existing roots or base
forms. Derivational morphology refers to the creation of new nouns, verbs, adjectives,
and adverbs by adding derivational morphemes to existing roots or base forms (1).
New words cannot be formed though inflectional morphology, as inflectional mor-
phemes must be use according to grammar that grammatical morphemes like –ed is for
past tense and must be used when the writer want to write something that happened.
For example, from the analysis above, -ed is for past tense so the word deliver-ed,
remark-ed, inflict-ed from the passage that we are given all have the morpheme –ed.
Moreover, -s is added on pluralnouns, nation-s, term-s, vowel-s are examples and –s is
added on verbs for Singular present tense like represent-s.On the other hand, new
words can be created and can enter a language though derivational morphology by
(Continued)
Example 1 (Cont’d)
adding different morphemes to existing roots. For example, In the article, flat(adjective)
is the root, and –ness is the morpheme added, and a new word flat-ness which is a
noun is formed. Certain adjectives can also form superlative adjective, adverb and
noun by adding -est, -ly, and -ity respectively. New words can be formed from a root
that is a noun or a verb too. For example, in the article, tradition which is a noun,
can form an adjective tradition-nal by adding –nal. And arrive which is a verb, can
form a nounarrive-al by adding -al. (2)
However, there are some words that may or may not be broken down into morphemes,
most of them contain roots or stems that are from Latin. For example, eliminate may be
broken down as e-limit(limin)-ate because e- means out of, limin is the root of limit and
–ate is a loanword from Latin. Also, prominent may be can be pro-eminent, prevent
may be broken down as pre-vent(venire) as venire means come in Latin, salutary may
be broken down as salut-ary as salut means health in Latin. (3)
Tutor comment (1):

{Your introduction begins well but needs to end here, except that you need
to add one more point. If you look back at the model you’ll see the introduc-
tion has 3 components and one is to introduce the language sample you are
using. Your language sample is the article ‘The Australian accent’. Can you
put a sentence here that introduces it?
If you look at the model, you’ll see that you need one paragraph for each
of the topics you are to cover: inflectional and derivational morphology, in
this case. This means you need to have a topic sentence for each paragraph
that introduces and defines or elaborates the brief definition of the type of
morphology you’ve introduced in the introduction. Then after the topic sen-
tence, you need to discuss the types of morphology with examples. You need
to do this with all the information below, because it is not well organised.}
Tutor comment (2):

{Some good points, but they need reorganising as I’ve said above.}
Tutor comment (3):

{For this paragraph, you need to introduce it in your introduction and say
whether it relates to inflectional or derivational or both.
Finally, you need a conclusion. Have a look at the model you were given to
get some ideas of what the conclusion is meant to contain.
Good luck with working on your second draft, which I look forward to
receiving.}
As the language coach points out in the first part of comment (1), the student
has begun the text well by introducing and defining the topic, however the
student omits the next phase of the introduction, which should identify the
language sample that is used for the analysis. The coach provides a fairly
explicit comment to support the student to include this phase of the introduc-
tion and points the student back to the model text as a way of guiding her to
engage with the support materials. It is important to remember here that stu-
dents are not used to frontloading (specifically the deconstruction stages of
the teaching–learning cycle). Additionally, frontloading is typically mediated
by a teacher in classroom interaction, whereas in the SLATE project, thus
far, students must engage with the frontloading materials independently online.
As can be seen from these coach’s comments, students’ engagement with the
models ended up being mediated to some degree by the coaches, who pointed
them to the models, and to specific points in the model that were applicable.
As the coach points out in the second part of comment (1) and in comment
(2), the points in the student’s text are not well organized. The coach explicitly
states what the student needs to do here to shape the text into its required
stages and phases. Tutor comment (3) also focuses on guiding the student
towards constructing the required stages of the text as the student has both
begun an explanation of something not mentioned in the introduction and has
omitted the conclusion stage. The student used this feedback in revising her
work. The final draft of this student’s assignment is provided below. The draft
has been annotated for the stages and phases within stages to show how the
student is developing control of the genre of comparative interpretation.
Example 2 Sample of final draft of comparative interpretation

Stages Text
Introduction to Fromkin discusses two morphological processes: inflectional
topic. morphology and derivational morphology. (Fromkin 2006).
Definition of topic. Inflectional morphology refers to the usage of words while derivational
morphology refers to the creation of words. This essay contrasts these
Preview of points two morphological processes that appear in the language sample, the
to be covered and article “The Australian accent”. In the article, there are some words
introduction to that may or may not be broken down into derivational morphemes.
language sample.
First feature to Inflectional morphology refers to the usage of words according to the
be compared syntactic rules by adding inflectional morphemes to existing roots or
with definition. base forms. For example, nation-s is a word from the article where
Examples from nation is the existing root (or call base form) and -s is the inflectional
the text as morphemes that added on the root to form the plural noun nation-s,
evidence. and term-s, vowel-s are other examples. Also, inflectional morphemes
must be used according to grammar that grammatical morpheme like
–ed is for past tense. It must be used to write something that happened
in the past, so new words cannot be formed though inflectional
morphology as only the tense or plural is indicated. -ed is for past
tense so the word deliver-ed, remark-ed, inflict-ed from the passage
that happened in the past all have the morpheme –ed. Moreover, –s is
added on verbs for Singular present tense like represent-s.
Second feature On the other hand, derivational morphology refers to the

to be compared creation of new nouns, verbs, adjectives, and adverbs by
with definition. adding derivational morphemes to existing roots or base forms.
Examples from Therefore, new words can be created and can enter English
the text as though derivational morphology by adding different morphemes
evidence. to existing roots. For example, in the article, flat (adjective) is
the root, and – ness is the morpheme added, and a new noun
flat-ness is formed. Certain morpheme can also form superlative
adjectives, adverbs and nouns by adding -est, -ly, and -ity
respectively. New words can be formed from a root that is a noun
or a verb too. For example, in the article, the noun tradition can
form an adjective tradition-nal by adding –nal. Moreover, the
verb arrive can form a noun arrive-al by adding -al.
Third feature to However, there are some words that may or may not be broken
be discussed with down into roots and derivational morphemes, because most of
explanation and them contain roots or stems that are from Latin (Random House
examples from Dictionary 2009). For example, eliminate may be broken down
text as e-limit(limin)-ate because e-means out of, limin is the root of
limit and –ate is a loanword from Latin. Also, prominent may be
broken down as pro-eminent. Prevent may be broken down as pre-
vent(venire) as venire means come in Latin ,while salutary may
be broken down as salut-ary as salut means health in Latin.
Summary of In conclusion, apart from the words that are not sure if they
points from are analysiable or not, all the words formed by inflectional
comparison morphology and derivational morphology are made of roots and
morphemes. But they are different in the ways that inflectional
morphology need to be used according to grammar, while
derivational morphology can create new words by adding different
morphemes to existing roots.
References Reference:
Victoria A. Fromkin (editor) 2006 An introduction to Linguistic
Theory . Blackwell Publishing Ltd
http://dictionary.reference.com. Dictionary.com Unabridged Based
on the Random House Dictionary, © Random House, Inc. 2009.
As can be seen from Example 2, the student took the coach’s comments seri-
ously and attended to the recommendations. As a result, Example 2 functions
as a more organized text that contains both the required stages and the phases
within those stages. The student’s text is now organized into:
z an introduction, which contains an orientation to the topic, a brief defini-

tion, a preview and a listing of the article used for the analysis;
z the body of the interpretation, which contains a paragraph each on inflec-

tional and derivational morphology, as well as a paragraph on the words that
don’t fit neatly into one category or the other; and
z a conclusion which sums up the points.
While there is still some room for improvement, particularly with the internal
logic of the points within the paragraphs in the body of the interpretation6,
Example 2 shows a much-improved text in comparison to the first draft. This
improvement in student writing demonstrates the possible effectiveness of the
current consultative process.
In addition, in order to measure the impact of this process on student
language development over time, we also tracked students’ writing across
assignments. One question that we focused on was whether the notes and the
consultation process that students engaged with during Assignment 1 influ-
enced their writing as they worked on future assignments. There is some evi-
dence that students do transfer their developing knowledge about language to
future tasks; however, the majority of students in this group showed a similar
range of problems in the first draft of their second assignment as in the first
draft of their first assignment. This is not unexpected since language develop-
ment requires continuous support over time to have an impact. The SLATE
team is currently tracking students over a longer period of time to evaluate
long-term effects of their intervention.
3.6 Future Directions
The SLATE project has made significant contributions to the original LCC
vision and preliminary results and analyses of student writing provide evidence
that the project is successful in helping students gain control over their aca-
demic writing needs. However, there are still a number of aspects of the project
that need to be developed. In particular, the SLATE project needs to expand
the support provided to students before they write independently. One way of
doing this would be to further expand the modelling stage and add online joint
construction activities. With such expansion of the support provided to students
before they write, it is envisaged that the repair work currently required will be
minimized. It is here that the project raises a number of questions about the
nature of online pedagogy in relation to the Sydney School pedagogical cycle
(see Figure 3.2). We are yet to explore how the online deconstruction and joint
construction exercises will be taken up by students and how these will impact
their language development and writing. The SLATE project is currently exper-
imenting with these new technologies and ways of supporting students’ needs.
In doing so, we hope to contribute to an Appliable Linguistics that both draws
from and contributes to a theory of language that relates to a real-world issue.
Notes
1
Funding for this project comes from the City University of Hong Kong and the
University of Sydney.
2
In the examples attached to this chapter, SLATE and LCC are sometimes used
interchangeably.
3
To access David Rose’s work, please visit: http://www.readingtolearn.com.au/#/
home/.
4
Historically speaking, Rothery’s pedagogic initiatives drew directly on Halliday’s
and Painter’s work on language development, at about the same time that
Applebee and Langer popularized Bruner’s notion of scaffolding in relation to
Halliday’s and Vygotsky’s work.
5
The SLATE team is currently preparing material to pilot an online modelling and
joint construction session.
6
Notes on thematic development, discussed earlier, were developed in response to
the work of students like this one, who needed support in producing thematically
well-developed paragraphs within the stages of the text.
References
Applebee, A. M. and Langer, J. (1983). ‘Instructional scaffolding: reading and
writing as natural language activities’. Language Arts 60 (2), 168–75.
Bernstein, B. (2000). Pedagogy, Symbolic Control and Identity: Theory, Research, Critique
(revised edition). London and Bristol, PA: Taylor & Francis.
Bruner, J. S. (1966). Toward a Theory of Instruction. Cambridge: Harvard University
Press.
Bruner, J. S. (1960). The Process of Education. Cambridge: Harvard University Press.
Eggins, S. (2004). An Introduction to Systemic Functional Linguistics (2nd edn). London:
Continuum.
Gee, J. (2000). ‘New People in New Worlds: Networks, the New Capitalism and
Schools’, in B. Cope and M. Kalantzis (eds), The Powers of Literacy: A Genre Approach
to Teaching Writing. Pittsburgh, PA: University of Pittsburgh Press, pp. 43–69.
Halliday, M. A. K. (2006). ‘Working with meaning: Working towards an Appliable
Linguistics’. Inaugural lecture to mark the Halliday Centre for Intelligent Appli-
cations of Language Studies at the City University of Hong Kong.
Halliday, M. A. K. (1975). Learning How to Mean. London: Edward Arnold.
Halliday, M. A. K. (1978). Language as a Social Semiotic. London: Arnold.
Halliday M. A. K. (1991). ‘The Notion of “context’’ ’, in T. Le and M. McCausland
(eds), Language Education in Language Education: Interaction and Development
(Proceedings of the International Conference held in Ho Chi Minh City,
Vietnam, 1991), pp. 1–26.
Humphrey, S., Martin, J. R., Dreyfus, S. and Mahboob, A. (this volume). ‘The 3×3:
setting up a linguistic toolkit for teaching and assessing academic writing’.
Kalnin, J. (1998). ‘Walking on the commons: genre as a tool in supporting adoles-
cent literacy’, in Donna E. Alvermann, Kathleen A. Hinchman, David W. Moore,
Stephen F. Phelps and Diane R. Waff (eds), Reconceptualizing the Literacies in

Adolescents’ Lives. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 267–81.
Macken-Horarik, M. (1996). ‘Literacy and learning across the curriculum: towards
a model of register from secondary school teachers’, in R. Hasan and G. Williams
(eds), Literacy in Society. London: Longman, pp. 232–78.
Martin, J. R. (1992). ‘Macro-proposals: meaning by degree’, in W. C. Mann and
S. A. Thompson (eds), Discourse Descriptions: Diverse Linguistic Analyses of a Fund-
raising Text. Amsterdam/Philadelphia: John Benjamins, pp. 359–95.
Martin, J. R. (1993). ‘Technicality and abstraction: language for the creation of
specialized texts’, in M. A. K. Halliday and J. R. Martin (eds), Writing Science:
Literacy and Discursive Power. London: Falmer Press, 203–20.
Martin, J. R. (1999), ‘Mentoring semogenesis: “genre-based” pedagogy’, in F. Christie
(ed.), Pedagogy and the Shaping of Consciousness: Linguistic and Social Processes.
London: Cassell (Open Linguistic Series), pp. 123–55.
Martin, J. R. (2000: Reprinted in full 2006). ‘Grammar meets genre – Reflections on
the “Sydney School”’, in Educational Research on Foreign Languages and Arts 2.
Guangzhou, Sun Yat Sen University, pp. 28–54.
Martin, J. R. and Rose, D. (2007). Working with Discourse: Meaning beyond the Clause
(2nd edn). London, New York: Continuum.
Martin, J. R. and Rose, D. (2008). Genre Relations: Mapping Culture. London:
Equinox.
English. London and New York: Palgrave Macmillan.
Painter, C. (1984). Into the Mother Tongue: A Case Study in Early Language Development.
London: Frances Pinter.
Rose, D. (2005a). ‘Democratising the classroom: a literacy pedagogy for the new
generation’. Journal of Education 37, 131–67.
Rose, D. (2005b). Learning to Read; Reading to Learn: Teacher Resource Book. Sydney.
Website: http://www.readingtolearn.com.au/#/resources.
Rothery, J. (1996). ‘Making Changes: Developing an Educational Linguistics’, in
R. Hasan and G. Williams (eds), Literacy in Society. London: Longman, pp. 86–123.
Schleppegrell, M. J. (2004). The Language of Schooling : A Functional Linguistics
Perspective. Mahwah, NJ: Lawrence Erlbaum.
Schleppegrell, M., and Colombi, M. C. (eds.) (2002). Developing Advanced Literacy in
First and Second Languages: Meaning with Power. Mahwah, NJ: Lawrence Erlbaum
Associates.
Vygotsky, L. S. (1978). Mind in Society: The Development of Higher Psychological Processes.
Cambridge, MA: Harvard University Press.
Chapter 4
Negotiating Evaluation: Story Structure and

Appraisal in Youth Justice Conferencing
J. R. Martin, Michele Zappavigna and Paul Dwyer1
4.1 Introduction
Youth justice conferences operate as a diversionary form of sentencing in the

juvenile justice system of New South Wales, Australia. Typically, they involve a
young person who has committed an offence coming face to face with the
victim of the person’s crime, in the presence of family members, community
workers, police and a conference ‘convenor’. Conference participants are
encouraged to discuss the full range of material, emotional and psychological
impacts of the young person’s crime and to cooperate in developing an ‘out-
come plan’ according to which the young person undertakes certain tasks (e.g.
community service) as a way of ‘making amends’. In this chapter, we report on
some of our preliminary work on genre structure, particularly in relation to
the accounts young people provide of their offending behaviour. We examine
the role of evaluative language in and around these accounts to show how the
young person’s behaviour is being interpreted and re-framed in line with a
‘genre identity’ which the conference structure makes available to the young
person. We conclude with some observations about the way this connects to the
ritual-like characteristics of the genre as a whole.
4.2 Youth Justice Conferencing
Youth justice conferencing, a diversionary programme operating in the juve-

nile justice system of New South Wales, Australia, is one of several innovative
genres of legal practice that have been recently introduced in jurisdictions
around the world under the banner of a reform movement generally known as
‘restorative justice’.2 In the most basic terms, these conferences involve people
who have been caught up in an incident of youth crime coming together, sitting
in a circle, discussing the impact of the criminal behaviour and (within certain
legislative guidelines) deciding for themselves how the young person who
Negotiating Evaluation 45
committed the offence can best make amends. The willingness of governments
to experiment with such reforms might be seen as counter-intuitive given
that many politicians owe their electoral successes to populist ‘get tough and
lock ’em up’ policies and regularly engage in the rhetoric of ‘zero tolerance’
against even fairly minor criminal activity. However, as the failure of these poli-
cies to reduce crime rates becomes more apparent than ever and as prison
populations continue to grow, it is not so surprising that alternatives are being
considered, albeit quietly and often with minimal resources. At any rate, initia-
tives such as youth justice conferencing demonstrate a quite remarkable invest-
ment of hope in the power of talk to effect profound change in attitudes and
behaviour.
Criminological research into conferencing has predominantly taken a quan-
titative approach to investigating such matters as whether or not the process
can help reduce recidivism rates, how the time and cost factors involved com-
pared to those in conventional court proceedings, and the levels of satisfaction
reported by conference participants (Maxwell and Morris, 2001; Palk et al.,
1998; Strang et al., 1999; Trimboli, 2000). While the question of recidivism rates
is perhaps not fully settled, results from these studies have generally been
encouraging in this regard and have also repeatedly demonstrated that both
offenders and victims overwhelmingly report a very high degree of satisfaction.
There has been little research, however, of a more qualitative nature directed
towards describing and explaining what it is that participants in a conference
are actually doing, namely interacting through spoken discourse and other
communicative modes. Most of the theoretical work carried out by crimino-
logists and social psychologists who have sought to account for the efficacy of
conferencing has instead revolved around concepts such as ‘re-integrative
shaming’ (Braithwaite, 1989; Ahmed et al., 2001), ‘social bond threats’ (Retzinger
and Scheff, 1996) or, via the neo-Darwinian affect theory ‘emotional contagion’
and ‘collective vulnerability’ of the psychologist Silvan Tomkins (Moore and
McDonald, 2001). The ideal/typical conference, as portrayed in the restorative
justice literature, is often a kind of passion play with a core sequence of remorse,
apology and forgiveness (Hayes, 2006).
As specialists in functional linguistics, social semiotics and performance
studies, our aim is to look more closely at the verbal and non-verbal interactions
in conferencing in order to describe the way in which meaning is being negoti-
ated among participants. In our fieldwork to date (observations of 10 youth
justice conferences, of which 5 were videotaped and have been transcribed),
conference participants have, from time to time, made manifest their height-
ened emotions; however, this has not occurred as often, nor necessarily in the
same ways, as suggested by the ‘passion play scenario’ described above. In what
follows, we report on some of our preliminary work on genre structure, particu-
larly in relation to the accounts young people provide of their offending behav-
iour. We examine the role of evaluative language in and around these accounts,
using this appraisal analysis (Martin and Rose, 2007) to show how the young
person’s behaviour is being interpreted and re-framed in line with a ‘genre

identity’ which the conference structure makes available to the young person.
We then conclude with some observations about the way this connects to some
ritual-like characteristics of the genre as a whole.
4.3 Generic Structure
Youth justice conferences take place as an alternative to sentencing in the chil-

dren’s court (at the recommendation of either a magistrate or the police) in
cases where a young person has admitted his/her guilt and agrees to have the
matter settled by conference.3 Typically, the conference will bring the young
person into a face-to-face meeting with the victim of the crime (or a victim’s
representative), in the presence of each party’s family members, friends, possi-
bly a school teacher or social worker who knows the young person, and a police
youth liaison officer (also, on occasion, the arresting officer and, where rele-
vant, an ethnic or indigenous community liaison officer). The conference is
convened by a private citizen who, while trained and accredited by the state to
do this work, is not acting in a direct judicial or law enforcement capacity.
Rather, the convenor’s role is to facilitate a ‘structured conversation’ (Moore
and McDonald, 2000, p. 14) in which the participants are encouraged to discuss
not only how the crime occurred but also how they have been personally
affected by any material damages, emotional/psychological distress or harm to
relationships. The participants must then negotiate agreement on an ‘outcome
plan’ according to which the young person commits to certain tasks (a formal
apology, payment of some monetary compensation, volunteer work in a com-
munity organization, etc.) as a way of ‘putting things right’.
As this description suggests, youth justice conferencing is very much a
designed genre and convenors are trained to implement this design.4 As part of
this training, convenors are provided with a synopsis of what are considered the
major steps in a conference, along with suggested questions/prompts to use
(although there is some debate in the field as to how closely these prompts
should be followed; see Moore and McDonald, 2000; Hoyle et al., 2002). For us,
technically speaking, conferencing is a macro-genre (Martin and Rose, 2008),
involving a series of elemental genres, each of which appears to have its own
recurrent configurations of meaning and staging according to the task involved.
We offer the following as a provisional characterization of these elements
(parentheses indicate optional elements):
gathering
legal framing
commissioned recount of the offence
exploring consequences for various parties
(apologies and acknowledgements)

tabling possible remedies
(break and private negotiations)
brokering a collective agreement
ratification of outcome plan
(apologies and acknowledgements)
formal closing
(shared refreshments)
dispersal
Even as participants arrive and gather for a conference, roles and expectations
are already being negotiated (convenors, when interviewed, have commented
on the impact which a participant’s choice of clothing can have on other par-
ticipants and have mentioned the value of sometimes ‘stage-managing’ the
order in which participants arrive, either to heighten or lessen possible feelings
of tension and confrontation). Once gathered and seated in a circle, the conve-
nor will make a brief speech during which the conference is constituted as a
legal process, with confirmation of, among other things, the offender’s admis-
sion of guilt. This is succeeded by a series of recounts and rejoinders by various
parties, establishing from different perspectives the events leading up to and
constituting the offence, and exploring their repercussions. Often, the conve-
nor will return at this point to the offender and ask if there is ‘anything else’
he/she want to say to any of the other conference participants (an implicit
invitation to make an apology which is not always taken up). Participants are
encouraged to make some initial suggestions for an outcome plan. They are
offered a break, during which the different parties (associated with the offender
and victim, respectively) can consider more concrete proposals. When the par-
ticipants are brought back into the circle by the convenor, these proposals are
collectively discussed and, following verbal agreement, the details are written
up and read back to the group. Where the outcome plan explicitly requires an
apology, this can sometimes be made on the spot. This is followed by a formal
closing, the option of shared refreshments and dispersal.5
We are currently working towards a more precise functional labelling of the
elemental genres described above and suggest the term ‘Commissioned
Recount’ for the account of offending behaviour which is given by the young
person just after the start of the conference. This term is intended to capture
the way in which, typically, the account has to be ‘extracted’ by the convenor
from a less than forthcoming adolescent. It has the following generic structure
(^ signals sequence in this notation and parentheses are used to indicate
optional elements of structure):
Orientation ^ Record of Events ^ (Re-Orientation) ^ (Extension) ^ Interpre-

tation ^ Ramifications
This genre begins with an Orientation which sets the recount in time and space
and introduces the main participants; it continues with a Record of Events,
which presents the sequence of events leading up to and constituting the
offence from the young person’s perspective; the next stage, Re-Orientation is
optional and wraps up the recount and returns participants from reconstructed
past events to the spatio-temporal setting of the conference itself; this is fol-
lowed by an optional Extension stage which a convenor may deploy to elicit a
fuller account from the young person6; in the following Interpretation stage the
recount is evaluated, as emotions and social values in relation to the offence are
explored; finally in the Ramifications stage, some of the pertinent consequences
of the offence are canvassed. In the next section of the chapter we’ll consider
two examples of this genre, one concerning a mobile phone which has been
stolen and passed on to a young man who is subsequently charged with posses-
sion of the stolen goods, the other concerning an affray.7 Elsewhere (Martin
et al., 2009), we have used these examples to place the commissioned recount
genre in relation to the larger family of story genres (Martin and Rose, 2008)
and to show how (in contrast to, say, personal recounts) evaluation tends to get
bundled separately from the Record of Events. We will cover some of this
ground again here but extend the argument by means of a more detailed
appraisal analysis, showing not only where but also more precisely how, and by
whom, the offending behaviour is evaluated.
4.4 Two Stories
Our first example of the commissioned recount genre is developed below.

Given the floor by the convenor, the young person begins as follows:
I was walking to a mate’s house

And [inaudible] and I was walking with him.
This guy just came up to me
and he goes ‘Do you want to buy this phone?’
At this point the convenor interrupts him, in order to ask him to speak louder8
(for the benefit of the Police Liason Officer who, she says, has a cold, and
perhaps also for the benefit of our recording process). He then resumes, in
a slightly louder voice:
Yeah, I was, I was walking to a mate’s house.

and goes ‘Do you want to buy a phone?’
and I go ‘No’
and I go ‘Do you want to swap?’
[inaudible] want to swap with my phone
and he looked at my phone

and he goes ‘Yeah’
and we swap
and I went and stayed at my mate’s house
and when it came to night time I was going back home,
and we was walking, was walking up the road
and the police just came and got us.
The convenor at this point nods encouragingly, in case the young person wishes
to expand on his statement – an invitation the young person declines:
. . . [convenor nodding]
That’s it.
In terms of genre staging, this is a typical recount (Martin and Rose, 2008),
beginning with an Orientation, establishing the Record of Events and closing
with a Re-orientation, as follows:
Orientation
Yeah, I was, I was walking to a mate’s house.
Record of Events
and goes ‘Do you want to buy a phone?’
and I go ‘No’
and I go ‘Do you want to swap?’
[inaudible] want to swap with my phone
and he looked at my phone
and he goes ‘Yeah’
and we swap
and I went and stayed at my mate’s house
and when it came to night time I was going back home,
and we was walking, was walking up the road
and the police just came and got us.
Re-Orientation
That’s it.
The convenor then scaffolds an extension of what happened, covering events at

the police station:
[And then what happened? They came and got you. They found the phone.
What did they say to you?]
– They go that this phone was stolen
[OK. What did you say?]

– I go, you know, I swapped it. Yeah, they just took took me.
[Took you where?]

– Police station
[And who did they ring when they brought you to the police station?]
– My dad.
Our second example of a commissioned recount has a longer Record of

Events, as the young person recounts what happened when he went to back up
a mate who had arranged to fight someone one-on-one; the ‘duel’ gets out of
hand, someone is stabbed; the young person responds by chasing the rival gang
member who did the stabbing, threatening him, throwing bricks across the
road at him, causing damage to property and apparently terrifying a number of
bystanders. He has pleaded guilty to a charge of affray (though at the confer-
ence he initially introduces himself as ‘the victim’ and has to be corrected by
the convenor). His account begins as follows:
It was up there –
It was - [inaudible] on a Tuesday yeah. [inaudible]
and I got a call from my friend, Vxxxx, [inaudible] to come down to [name
of suburb].
Something going on;
but when I went down with them to [name of different suburb] station
and jumped on the train [inaudible]
and the train we got it straight to [name of suburb],
we got off at [name of suburb]
[inaudible] Mxxxx and other - his other two friends.
And they had a one on one
and I jumped in
and I turn around.
I was having a go with his friends
and the [one] next to me got stabbed
and he [inaudible] goes ‘Chase him!’
and I went and chased him,
started chucking stuff at him,
hitting him.
I couldn’t stop him.
He still had the knife in his hand
and after that I walked back to the station to see Vxxxx
and I see the policeman coming as I walked away
and [inaudible] happened to him
and I said to the two officers ‘Assault me, search me.’
They took out all my stuff
and they found out I was involved

and they took me back to the police station.
And my [inaudible]
and I had an interview
and they took my pants, my hat, my jacket
and I was released.
This recount once again begins with an Orientation and continues with a
Record of Events; but there is no formal closing (i.e. Coda):
Orientation
It was up there –
It was- [inaudible] on a Tuesday yeah. [inaudible]
Record of Events
and I got a call from my friend, Vxxxx, [inaudible] to come down to [name
of suburb].
Something going on;
but when I went down with them to [name of different suburb] station
and they found out I was involved
and they took me back to the police station.
...
and I had an interview
and they took my pants, my hat, my jacket
and I was released.
The Convenor then scaffolds a long Extension stage, exploring why the young
offender went to back up his mate, and establishing that he was charged with
possession of a knife:
[Convenor And when you- when you decided you would go down and help
your friend, what did he actually say to you?]
– He was going to see Mxxxx.
[And why was he going to see Mxxxx?]
– Because Mxxxx offered him out.9
[Offered what?]
– Nah [inaudible] two of us have a go like one on one.
[To fight?]
– Yeah.
[And so why did he need you there if he was going to fight him one on one?]
– Because I’m closer [inaudible]. He trusts me like in case anything
happens because I’ve known him since primary.
[And why did he think something would happen?]

– [pause] In case.
[And when you went to this location you had something on your person.
You had something with you. What was that?]
– The knife.
[So you want to tell us about the knife?]
– [inaudible] didn’t know I even had it on me. Forgot I had it on me the
whole time and then I couldn’t remember I had it until they [inaudible]
and pulled it out and showed them. It doesn’t even work. You can’t even
use it.
[But it’s not actually a knife.]
– No.
[A knife is included in it isn’t it?]
– Yes.
[What else is it?]
– A can opener, screwdriver
[And so when the police, um, searched you and found the um, what are
they, a Leatherman? Is that what they’re called?]
[Arresting Officer A Leatherman sort of tool.]
[Convenor When they found that what did you say to them?]
– Said it’s for work.
[Tell everybody what you did for work back then.]
– Panel beater.
[So, as I explained to you the other day, why do you think the police
charged you with having that weapon in your possession?]
– [inaudible] They thought I got ready for a fight . . . to use it
[And what else did they say to you? Do you remember?]
– Nup.
[Did they say anything to you about the fact that it, you know, as a panel
beater you probably don’t need a knife?]
– Yes.
[Is that – do you think that’s reasonable?]
– Yes.
[So you accept the fact that you got charged with the possession of the
knife.]
– Yes.
[Alright and then when you chased after Nxxxx what were you going to do
when you caught him?]
– Wasn’t going to fuck with him. Was going to bring him back.
[Bring him back to where?]
– The station . . . Well I couldn’t, he had the knife in his hand still.
[What did you think when you saw Nxxxx with the knife?]
– I didn’t want to get – I didn’t want to get [next to him] in case I got
stabbed. If I only had the knife I would have pulled it out on him too but
I didn’t have it. I forgot I had it on me.
[So what did he say to you when you – when you caught up with him?]
– He just goes ‘What did I do? I didn’t do nothing’.
[And then what did you do to him?]
– I chased him until he went in – inside a shop.
[Yeah. And then what did you do?]
– I went back to the station and I went back down and that’s when I got
stopped.
Unlike personal recounts, which have an ongoing prosody of evaluation,

these commissioned recounts are ideationally focused. Their ‘point’ is to estab-
lish the young person’s account of the sequence of events leading to his arrest
(comparable to the police Record of Events). Evaluation of their significance
for the youth justice conference is negotiated by the convenor after their dia-
logic extension. In this respect they contrast markedly with personal recounts
deployed to sustain solidarity with partners, friends and family. The following
excerpt from a long written recount demonstrates how ongoing evaluation of
this kind can be deployed. We’ve chosen a text featuring the kind of language
young offenders might well use with their mates, to bond with them and alien-
ate outsiders; it’s from a Year 8 (so about 14 years old) female student, from an
indigenous Australian background.
Fucken Hell man, who the hell told you I liked doing this kind of shit. On
Saturday I saw Brian and Brendon and his Girlfriend at Waterloo, I was
waiting to catch the bloody bus, anyway they started talking to me so that
killed a lot of time. Anyway I had to go to the Laundromat Yesterday and I saw
my ex-boyfriend man he looks fucken ugly god knows what I went out with
him, he looks like a fucken dickhead
ANY WAYS HE WAS
so ugly only a blind woman would go out with him. I ran into this elderly man
that lived down one of my old streets and because I had a bag of clothes the
stupid cunt said to us are you running away from home which is bull-shit
because the sooner that I got home the happier I would have been. Then my
ex-boyfriend comes up which makes it even worse and he starts calling this
old cunt a cradle snatching little ass-hole. I mean as if its any of his business,
and like this is totally humiliating cause I mean everybody and I mean every-
body tried to see who the hell was making all the fucken noise and yes there
I was trying to hide my face as soon as possible . . .
Young offenders studiously avoid anti-language of this kind on youth justice

conferences. And the ideational focus of their ‘testimony’ is so strong that
they regularly avoid evaluative language altogether in their Record of Events
stage.
4.5 Interpreting the Stories
The flat ideational focus of young offenders’ testimony in the Record of Events
means that convenors and liaison officers often take responsibility for interpret-
ing what has happened. To analyse these evaluations we draw on appraisal the-
ory (Martin and White, 2005). As outlined in Figure 4.1, appraisal comprises
three main systems, attitude, graduation and engagement. Attitude is
monogloss
ENGAGEMENT
heterogloss
AFFECT
APPRAISAL ATTITUDE JUDGEMENT...
APPRECIATION...
FORCE...
GRADUATION
FOCUS...
Figure 4.1 Appraisal systems (after Martin and White, 2005)

concerned with types of feeling, graduation with their strength and bounded-
ness, and engagement with their sourcing.
Our main concern here will be with attitude, which embraces affect (our
emotional reactions), judgement (how we evaluate people’s character and
behaviour) and appreciation (the value we place on things). In the examples
below, for example, an Ethnic Liaison Officer queries the young person in the
affray conference about his reactions (upset, pissed off – affect), challenges him
to re-make his character (how much of a man, wrong – judgement), and com-
ments on how easy it is to break the law (so easy – appreciation).
ELO: Hey. You think I’m talking to you now I’m – I’m upset with ya? Do you
think I’m pissed off with ya? I don’t know you but do you think I’m upset with
ya? [affect highlighted]
ELO: You find out how much of a man it takes, hey?, to stay out of trouble.
You tell me which one takes more of a man to do. How – does it take more of
a man to listen to your friends or to say to your friends, ‘No, what you’re
doing is wrong’. [judgement highlighted]
ELO: Do you know how easy it is to break the law? It’s so easy. [clicks fingers]
It’s so easy. [clicks fingers] It is so easy. [appreciation highlighted]
In a similar vein, as part of the guided interpretation of the Record of Events

in the mobile phone conference, the Convenor explores the emotional reac-
tions of the young person and his parents, beginning with Dad’s anger:
Convenor
And who did they ring when they brought you to the police station?
Young Person
My dad.
Convenor
OK. And what did your dad say? Was he angry, happy?
Young Person
[nods] Angry.
Convenor
Angry. [laughs] Yeah. OK.
And then Mum’s:
Convenor
What did mum say when you got home?
Young Person
She was sleeping. [nodding]
Convenor
She was sleeping. Was Mum angry?
Young Person
[nods] Yeah.
Convenor
Did she get upset?
Young Person
[half nod/raises eyes]
And later that of the young person himself, were he to find himself in a compa-
rable situation as victim:
Convenor
So if that was your phone, how would you feel if somebody stole your phone
and then, you know, I decided I wanted to swap it with mine because mine
was older. How would you feel about that?
Young Person
Pretty angry.
Convenor
Yeah.
As far as judgement is concerned, the Convenor also focuses on the young

offender’s misbehaviour, including his lack of concern about the phone belong-
ing to someone else:
Convenor
[bobs head] You need to tell us why you took the phone.
Young Person
Because it was new.
Convenor
Because it was new. Newer than yours.
Young Person
[nods] Yeah.
Convenor
So you didn’t care that it was somebody else’s phone? Is that right?
Young Person
Yeah.
Convenor
OK
She checks to make sure the young person can now acknowledge that he did
the wrong thing:10
Convenor
Did he [dad] say anything to you? [pause]
Young Person
Don’t go anywhere.
Convenor
Don’t go anywhere. As in when you get home you’ve got to stay at home?
Young Person
[raises eyes in a half-nod]
Convenor
Do you think your father was disappointed in you?
Young Person
[nods] Yep.
Convenor
So you know you did the wrong thing.
And asks him to reaffirm this acknowledgement later on:
Convenor
Tell me what happened when mum found out what you did. [tilts head]
[pause] Did she cry?
Young Person
St-() Lecture.
Convenor
You got a lecture. [nodding]
Young Person
[nods]
Convenor
Do you think you deserved a lecture?
Young Person
[looks down]
Convenor
Why did you deserve the lecture?
Young Person
Because I did something wrong.
Convenor
Yep. [nods]
She gets the young person to confirm that his parents have acted reasonably in
grounding him:
Convenor
So it’s quite limiting in terms of what you can do. And that’s – Do you
understand that that was the consequence of what you did?
Young Person
[nods] Yeah.
Convenor
Yeah? Do you think that was reasonable what mum and dad did, grounding
you?
Young Person
[nods]
Convenor
Yeah?
Young Person
[nods]
She also checks twice that the young person has apologized11 to his victim, who
is not in fact present at this conference:
Convenor
Right and what did you – Did you say anything to him when you found out
that it was his phone?
Young Person
[shakes head] Nah.
Convenor
So you didn’t say sorry to him? [shaking head]
Young Person
(Nah), I said sorry and he goes ‘you don’t have to say sorry , it wasn’t you
that did it.’
Convenor
Right. [half nodding] OK.
...
Convenor
Hey? So you’ve apologized to Jxxx already.
Young Person
Yeah.
Convenor
OK.
The Convenor also guides the young offender towards expressions of remorse:
Convenor
Do you think that mum and dad were disappointed in you?
Young Person
[nods]
Convenor
[nods] Were you disappointed in yourself? Or Not? Or you don’t care?
Young Person
[nods] Yeah.
Convenor
Yeah or you don’t care? [nodding]
Young Person
Disappointed in myself.
Convenor
[tilts head] You are.
Young Person
[nods]
The Convenor’s exploration of affectual meaning in the Interpretation stage

of the mobile phone Commissioned Recount is outlined in Table 4.1. [As foot-
noted, disappointed, sorry, and apologize have been double coded for affect
and judgement and so entered in both Tables 4.1 and 4.2.] Including the two
inscriptions of happiness, which are tendered ironically by the Convenor, the
family experiences only negative emotions.
Judgement is also largely negative, focusing mainly on the young person’s
behaviour.
Looked at in terms of turn-taking (Martin et al., 2009), what is significant
here is that the Convenor initiates virtually all exchanges and introduces
virtually all explicit evaluation, to which the young person responds compli-
antly a word or phrase at a time. The result is an interpretation of the recount
Table 4.1 Inscriptions of affect in the Interpretation stage of the mobile phone
recount
speaker appraising ATTITUDE POLARITY emoter trigger
item
convenor angry dis/satisfaction negative father of YP YP at police station
happy un/happiness positive father of YP YP at police station
(ironic)
YP angry dis/satisfaction negative father of YP YP at police station
convenor angry dis/satisfaction negative father of YP YP at police station
happy un/happiness positive father of YP YP at police station
(ironic)
disappointed dis/satisfaction negative father of YP YP
sorry un/happiness negative young person receiving phone
YP sorry un/happiness negative young person receiving phone
convenor angry dis/satisfaction negative mother of YP YP receiving phone
upset un/happiness negative mother of YP YP receiving phone
cry un/happiness negative mother of YP YP receiving phone
YP angry dis/satisfaction negative young person hypothetically
having own
phone stolen
convenor apologized un/happiness negative young person receiving phone
disappointed dis/satisfaction negative YP’s parents young person
disappointed dis/satisfaction negative young person young person
YP disappointed dis/satisfaction negative young person young person
shaded = affect; shaded + bold = double coding of affect and judgement

Table 4.2 Inscriptions of judgement in the Interpretation stage of the mobile

phone recount
speaker appraising item attitude polarity appraiser appraised
young person stolen propriety negative police person stealing phone

convenor stolen propriety negative young person person stealing phone
convenor stolen propriety negative young person person stealing phone
convenor disappointed propriety negative father of young person
young person
convenor wrong propriety negative convenor YP receiving phone
convenor sorry propriety negative young person YP receiving phone
young person sorry propriety negative young person YP receiving phone
young person wrong propriety negative convenor YP receiving phone
convenor stole propriety negative convenor hypothetical thief
convenor stolen propriety negative convenor hypothetical thief
young person stolen propriety negative convenor hypothetical thief
convenor good propriety positive convenor young person
agreeing
reasonable propriety positive convenor Mum/Dad grounding
YP
apologized propriety negative young person YP receiving phone
disappointed propriety negative young person’s young person
parents
disappointed propriety negative young person young person
young person disappointed propriety negative young person young person
shaded = affect; shaded + bold = double coding of affect and judgement
determined by the Convenor – it’s the Convenor, not the young person, who
controls what the recount means.
Interpretation of the impact of the Commissioned Recount is of course
crucial to the re-alignment of the young person, away from their circle of mis-
guiding mates, and towards their family and wider community. Returning to
the example of the affray conference, it is the Ethnic Liaison Officer who inter-
venes to extend the convenor’s interpretation, apparently out of frustration
with its effectiveness to that point in the conference. He makes explicit refer-
ence to icons of the Muslim cultural background which he shares with the young
person, beginning with the headscarf worn by the young person’s mother:
Ethnic Liaison Officer

Listen, [looking to the convenor] I want to take, with your permission,
I wanna take a different angle. OK? Mate, what’s your mum wearing on
her head?
Young Person
Scarf.

Yeah. OK.
He then focuses attention on the fact that because of what the young person
has done, his mother, through no fault of her own, is now sitting in the pres-
ence of uniformed police.

What a – where is she now? In the presence of who?
Young Person
= Me.

= Who- who’s- No. Who’s sitting here? Who’s sitting here right now? Have
a look across.
Young Person
Men.

Have a – but have a look across. What uniform are they wearing?
Young Person
Police uniform.

OK.
This sets up the ELO’s concern abut the bad impression the young person is
creating for himself, and the Muslim community:

[pointing to the university researchers] Where are these guys from? They’re
from a certain place. OK. What’s the perception going to be?
Young Person
Think bad of me.

What are they gonna – when they see your mum wearing a scarf, I’m
Muslim background myself. What are they going to think?
Young Person
Bad.

OK.
This was a very sensitive issue at the time since former Prime Minister John
Howard’s neo-conservative government had fanned the flames of anti-Muslim
ethnic prejudice by playing up the issue of ‘border protection’, incarcerating
mostly Iraqi and Afghani asylum seekers and manipulating nationalist senti-
ment in support of Australian troops joining the American invasion of Iraq.
In Sydney, racial tension in fact erupted in violent conflict between ‘white
Australian’ and Lebanese youth in what became known as the Cronulla riots.
The ELO then questions the idea that there is something tough about the
young person’s behaviour:

I’m asking you YP. OK. Because I’m listening to you man and I don’t see you
as a leader at the moment. I see you following your friends. I see your
friends say jump, you say how high. That’s how I see you. OK.
Young Person
Yeah.

You wanta be tough. But you just – you’re not, number one.
He makes clear his revulsion at the fact that the young person is embarrassing
his mother (and the Muslim community she symbolizes) in this way:

Number two, man, when I see someone of my own background bringing
their mum in wearing a Hejab, OK, honestly man inside I feel sick. You
understand?
Young Person
Yes.
As far as judgement is concerned, the ELO’s Interpretation centres on honesty,

mainly in relation to the young person’s sincerity (see Table 4.3).
The young person’s capacities are also explored, including his lack of leader-
ship, manhood and ability to control his temper (see Table 4.4).
But the majority of the explicit judgements deal with the impropriety of the
young person’s behaviour (see Table 4.5).
As flagged above, the ELO is especially concerned about the lack of respect
being shown by the young person for his family (see Table 4.6).
And, paralleling this, the lack of respect being shown to the Muslim commu-
nity (see Table 4.7):
As with the Convenor’s Interpretation of the mobile phone recount, the
ELO initiates all of the exchanges; and, once again, the interpretation of the
commissioned recount is jointly constructed with the ELO assuming control,
Table 4.3 Inscriptions of honesty in the ELO’s Interpretation stage of the affray
recount
speaker appraising item social esteem social sanction explicit implicit
ethnic liaison I honestly don’t veracity inscribe

officer think you do
ethnic liaison I’m being honest veracity inscribe
officer with ya
ethnic liaison honest veracity inscribe
officer
young person honest veracity inscribe
ethnic liaison swear veracity inscribe
officer
young person swear veracity inscribe
officer
young person Yeah [honest] veracity inscribe
officer
Table 4.4 Inscriptions of capacity in the ELO’s Interpretation stage of the affray
recount
ethnic liaison don’t see you as capacity inscribe

officer a leader
tough… you’re capacity inscribe
not
(you think...) capacity inscribe
hard criminal
ethnic liaison I’ve got half a capacity provoke
officer brain
how much of a capacity
man
more of a man capacity
more of a man capacity
ethnic liaison you can control capacity inscribe
officer yourself
young person I can control capacity inscribe
myself
I can’t [control capacity inscribe
myself]
Just can’t capacity inscribe
[control
myself]
ethnic liaison temper capacity
officer
young person Yeah [temper] capacity
ethnic liaison (if...) smart capacity inscribe
officer
Table 4.5 Inscriptions of propriety in the ELO’s Interpretation stage of the affray recount
young person bad propriety inscribe
bad propriety inscribe
ethnic liaison officer put down our community propriety inscribe
put down our religion propriety inscribe
put down the Hejab propriety inscribe
contradicting our religion propriety inscribe
respect your mum propriety inscribe
young person Yes [respect] propriety inscribe
Negotiating Evaluation
ethnic liaison officer No you don’t [respect] propriety inscribe
you don’t respect your mum propriety inscribe
You have no respect for your mum propriety inscribe
You have no respect for your mum whatsoever propriety inscribe
You have no respect for what your mum’s got propriety inscribe
on her head
You have no respect for our community propriety inscribe
You have no respect propriety inscribe
trying to impress propriety
(the right) to go and hurt other people propriety inscribe
young person No right [hurt] propriety inscribe
he hurt my mate propriety inscribe
That’s it [hurt] propriety
ethnic liaison officer if you get a job you’ll be out of trouble propriety inscribe
if you don’t get a job you’ll get into trouble propriety inscribe
responsibility propriety inscribe
break the law propriety inscribe
65
(Continued)
66
Table 4.5 Cont’d
break the law propriety inscribe

breaking the law propriety inscribe
try staying out of trouble propriety inscribe
stay out of trouble propriety inscribe
wrong propriety inscribe
disrespecting propriety inscribe
disrespect propriety inscribe
no respect propriety inscribe
laugh at your face propriety inscribe
pretend that they care propriety inscribe
wrong propriety inscribe
trouble propriety inscribe
getting your mum into this crap propriety inscribe
getting your family into this crap propriety inscribe
not right propriety inscribe
no good propriety inscribe
good person propriety inscribe
not good propriety inscribe
trouble propriety inscribe
hurting propriety inscribe
don’t care propriety inscribe
alright propriety inscribe
alright propriety inscribe
Table 4.6 Inscriptions of impropriety in relation to Mum

ELO bad propriety inscribe
respect your mum propriety inscribe
young Yes [respect] propriety inscribe
person
ELO No you don’t propriety inscribe
[respect]
you don’t respect propriety inscribe
your mum
You have no propriety inscribe
respect for your
mum
You have no propriety inscribe
respect for your
mum whatsoever
disrespecting propriety inscribe
disrespect propriety inscribe
no respect propriety inscribe
getting your mum propriety inscribe
into this crap
getting your family propriety inscribe
into this crap
Table 4.7 Inscriptions of impropriety in relation to the Muslim community

bad propriety inscribe
ELO put down our propriety inscribe
community
put down our propriety inscribe
religion
put down the Hejab propriety inscribe
contradicting our propriety inscribe
religion
for what your
mum’s got on her
head
for our community
proffering virtually all evaluations, and with the young person responding a
word or phrase at a time.
As far as attitude is concerned, the ELO’s interpretation of the recount has
some affect but concentrates mainly on judgement. Interestingly, the ELO does
not condemn the young person per se but focuses largely on his behaviour.

Why don’t you look at why we’re here today? Are we here because of me?
Young Person
No.

Are we – who are we here for?
Young Person
Because of me.

Not because of you as a person because of something you’re doing that’s not
right. You’re probably a good person. What you’re doing is not good. You
understand the difference?
Young Person
Yes.

We’re not saying, ‘Mxxx, you’re a this and you’re a that’. What we’re saying is
your behaviour is getting you into trouble, man. You’re hurting your family,
brother. You’re hurting your [Arabic], brother. You understand?
Young Person
Yes.

If I didn’t care about ya, man, I didn’t care about your mum, I didn’t care
about, you know, the Den and everything, I wouldn’t even be here. I mean,
I’ve finished my work. But if we – if everyone here today could help you just
to sorta think to yourself, ‘What am I doing to my family? What am I doing to
myself?’, man, Shalam. That’s what it’s all about. That’s is what today is all
about. It’s about you sitting down and having a look – . . .
The ELO seems to exemplify here what John Braithwaite, one of the leading
theorists of restorative justice, sees as central to the efficacy of conferencing,
namely a process of ‘reintegrative’ (rather than ‘stigmatizing’) shaming: ‘com-

municating disapproval of an act with respect, with special efforts to avert
outcast identities and to terminate disapproval with rituals of forgiveness or
reconciliation’ (Braithwaite in Ahmed et al., 2001, p. 39). In the concluding
section of this chapter, we will place our analyses above in relation to some
of the issues that other restorative justice theorists have raised in regard to
Braithwaite’s conceptualization of shame. At the same time, however, we want
to recuperate the other part of his argument, namely that conferencing as a
new genre of legal process has taken on some ritual, or at least ritual-like,
characteristics.
4.6 Conferencing as Ritualized Redress
The notion of re-integrative shaming has not been uncontested in the restor-
ative justice literature. Indeed, Braithwaite and his associates have readily
acknowledged the need for ‘repair work to the theory’, including the need to
make distinctions between shame as an emotion and shaming as practice,
between ‘unresolved shame’ and ‘embarrassment—exposure’, between shame/
shaming and pride/praise etc. (Ahmed et al., 2001, p. 41). Other theorists have
downplayed altogether the centrality of shame (Morris, 2002) or else inter-
preted it more as a collective experience, a ‘visceral reminder’ to participants
that they ‘can experience positive emotions in each other’s company’ which
thereby prompts a transition towards cooperation and resolution of the confer-
ence (Moore and McDonald, 2001, p. 138). Nevertheless, one way or another,
the emotional dynamics of conferencing are seen as critical to its restorative/
re-integrative functions, as the following commentary by psychoanalyst Donald
Nathanson (1997, p. 141) suggests:
Not surprisingly, the initial response of the perpetrator was both indifferent
and unconcerned [. . .]. Yet as the conference ran on and both family groups
began to speak about their estrangement from the perpetrator, that individ-
ual came swiftly to learn that the love of the community was a deeply missed
and quite important part of his or her world. With such recognition came an
avalanche of shame, after which the individual was likely to cry, accept the
forgiveness of all concerned, and sign a document pledging to work in some
way to repair or undo the damage produced by the antisocial act.
It has not been our intention here to add yet another layer to these interpre-
tations of shame and other affects but rather to highlight the fact that, at least
for the conferences we have observed and documented, the evaluative language
of affect, appreciation and judgement is rarely used by young people and to
show how, where such evaluations do occur, they are very much co-created
and guided by other participants. Of course, the analyses reported above are
dealing mainly with what we have called the Commissioned Recount genre and
more research is needed before we can confidently say that these patterns
established at the start of conferences are maintained throughout. Notwith-
standing this caveat, our early findings offer a strong contrast to the observa-
tions of Braithwaite and Mugford in a seminal paper outlining the ‘conditions
of successful reintegration ceremonies’ (which is what they take conferencing
to exemplify). In line with the principle that ‘Non-authoritative actors (victims,
offenders, offenders’ families) must be empowered with process control’,
Braithwaite and Mugford interpreted ‘[the] practice of temporally privileging
the accounts of the young people as a desirable way of seeking to empower
them in the dialogue. For all parties, success is predicated upon a significant
degree of agency’ (Braithwaite and Mugford, 1994, pp. 148–50).
Against this view, our observations are more in line with critique made by
criminologist Hennessey Hayes in a recent re-examination of some major evalu-
ations of conferencing. His most relevant comments can be summarized as
follows:
Results of these major [research] projects, as well as results from the various
evaluations studies, are remarkably consistent and show that offenders and
victims view conferencing processes as fair and are generally satisfied with the
outcomes. However, there is less evidence that shows conferences are also
‘restorative’ [. . .] there may be marked limits on how far conferences can go
in repairing harm, inducing remorse, and helping victims and offenders
move on [. . .] conferences sometimes do not induce remorse, young offend-
ers sometimes do not feel sorry and offer an apology, and victims sometimes
do not forgive [. . .] in New Zealand, Maxwell & Morris observed [. . .] that
[. . .] 25% of victims felt worse for having attended a conference, mainly
because ‘the victim did not feel that the young person and his or her family
was truly sorry’. (Hayes, 2006, pp. 370, 377–8)
In his paper Hayes reviews two sources of difficulty with the restorative
apology-forgiveness ideal. One is that there can be a drift from apologetic dis-
course to mitigating accounts, with the young person offering excuses for his/
her behaviour. The following exchange from our second conference illustrates
this point:

So if you don’t like anybody dictating to you what to do right, what gives you
the right to go and hurt other people?
Young Person
No right, but he hurt my mate . . .
Hayes’ second point has to do with the presence of third parties (e.g. convenor,
liaison officers), so that ‘The youth justice conference process . . . transforms
the private act of apologizing and offering forgiveness into a public drama of
restorative justice’ (2006, p. 379). To these misgivings we can now add our
observations concerning power relations inherent in the genre, and the regula-
tory role taken up by convenors, police and liaison officers, which seem to
leave little room for the young person to do other than respond compliantly
to evaluations introduced by others. What kind of ‘sincere’ sounding apology is
possible when the young person is so heavily scaffolded?
That said, we do not mean to suggest that scaffolding is necessarily a bad
thing. On the contrary, given that youth justice conferencing is still an emerg-
ing genre and a novel, typically one-off experience for most participants (not
to mention the low levels of literacy and fluency in English of some of these
participants), some form of scaffolding is clearly essential. Furthermore, while
the regulative discourse in conferencing does inscribe hierarchical power
relations between participants, this is often paired with close and certainly
negotiable relations of solidarity. Indeed, this readily observable tension
between the vertical and horizontal dimensions of tenor aligns with what we,
like Braithwaite, are beginning to think of as the ritual-like, performative ‘force’
of conferencing.
However dismissive we may feel at times about the social value of ceremonies
and ritual (or, conversely, however much we might wish to leave their ineffable
mysteries intact), in common sense terms we can all appreciate the difference
between winning a race and the medal/trophy presentation, between passing
exams and the graduation ceremony, between getting together with someone
and marriage, between a casual prayer and a religious service, between dying
and the funeral or, in the case of youth crime, between getting caught by the
police and being sentenced. In social semiotic theory, questions about the
power of such genres are only beginning to be explored (cf. Bednarek and
Martin, forthcoming) but, as a starting point, we can look to how scholars in
anthropology and performance studies have thought about these ‘special events
(a category that includes ritual)’ as ‘intensifications of some of the tendencies
inherent in ordinary activity, but often latent or subliminally present’ (Lewis,
2008, p. 43).
Following Victor Turner (1982), for instance, we can consider the problem of
youth crime (and the attendant failure of institutions like the children’s court
and juvenile detention centres to deter and/or rehabilitate young offenders) as
a form of ‘social drama’, a breach in the social order which has reached a point
of crisis and now requires redressive action if a profound schism in the com-
munity is to be avoided. Turner identifies legal–judicial processes and ritual
performances as the two most important mechanisms of redress though clearly
these are not mutually exclusive options. Some elements of ritual have always
existed in courtroom proceedings but these elements have arguably become
less efficacious in recent decades. At the same time, an emergent genre such
as conferencing may be understood as the ritualization of alternative social pro-
cesses (police cautions, family counselling, parent/teacher interviews, etc.) as
adjuncts to conventional legal–judicial remedies. We use the term ‘ritualization’
here to highlight some ‘logical entailments’ (Rappaport, 1999, p. 26) of the

genre:
z Conferencing practice frequently involves an appeal to tradition and various

kinds of authority.12 In New Zealand, conferences involving Maori young
people typically take place on the marae and begin with a communal prayer;
in other jurisdictions, they may begin with a Bible reading. In New South
Wales, though it is a legally constituted process, youth justice conferencing
certainly allows for, and often seems to encourage, participants to re-frame
the matter they are dealing with as an infringement of family, religious or
cultural values, not simply a legal violation: for many citizens, these are no
doubt greater sources of moral authority than the state.
z Commensurate with such a sense of moral purpose, conferencing involves a
high degree of repetition and formality. While restorative justice advocates
are probably right when they describe the protocols of conferencing as less
alienating than those of courtroom practice, the genre nevertheless involves
its own kind of ‘restrictive code’ (Bernstein, 1975; Douglas, 1973): hence,
the care taken to train convenors in the appropriate use of script/prompts,
the attention given to briefing participants before the conference, the highly
deliberate turn-taking structures within the conference, and so on.
z To participate in a conference is also, in a ritual-like manner, to respond to
a demand for performance. Participants will frequently talk of ‘facing up to
the challenge’ of meeting the other participants, of ‘getting through’ the
conference process in order to be able to ‘draw a line in the sand’ and ‘move
on’ with their lives and so on. As Catherine Bell writes, a ‘fundamental
dimension of ritualisation’ is ‘the simple imperative to do something in such
a way that the doing itself gives the acts a special or privileged status’ (1997,
p. 166). Indeed, much of what is commonly thought to be ‘symbolic com-
munication’ in ritual, or ritual-like processes, might be better understood
in terms of the cultivation of embodied dispositions: ‘the act of kneeling
does not so much communicate a message about subordination as it gener-
ates a body identified with subordination’ (Bell, 1992, pp. 99–100). While
the physical spaces in which youth justice conferences are held, in our expe-
rience, rarely feature such obvious rallying symbols or ‘bond-icons’ (Martin,
2008; Martin and Stenglin, 2007; Stenglin, 2004) as those of a courtroom
(the national flag, the coat of arms, etc.), the enactment of the conference
itself seems to become a process in relation to which participants bond.
The public nature of conferencing, as Hayes has argued, may render the
putative core sequence of remorse–apology–forgiveness problematic, yet many
participants are still able to say they are satisfied with the process. It is, after all,
an orderly convocation of persons who present themselves to one another as
citizens (and as family members, members of a religious/cultural group, etc.)
with some interest in promoting orderly behaviour (and in promoting the
authority of family, religion, culture, etc. in addition to, or perhaps over and
above, the authority of the state). At the very least, a victim’s words are attended
to by an appropriately constituted ‘community of care’. To do the conference
at all requires enacting, on the spot, some of the (re)integrative effects it is
hoped the conference might be able to foster in the longer term: a young
person is made to talk and to listen but is also given, then and there, an oppor-
tunity to affiliate with his family, his ethnic group, the wider community (with
the possible support of police liaison officers) alongside, or in place of, his
hitherto dominating affiliation with mates. Who knows which affiliations will
stick? The conference is still likely to have emerged as a suitably formal legal
process, though accessible to lay persons, in and through which various pre-
ferred affiliations may be ‘re-presented’ or ‘modelled’ (Handelman, 1990).
The opportunity for young people, victims, their families, friends and police to
affiliate simply in relation to the doing of a conference is already a paradigm
shift for criminal justice systems which have for so long manifestly failed to
rehabilitate offenders and to meet the needs of victims.
Notes
1
We are indebted to Chris Cléirigh for his help with the appraisal analyses reported
on here. We also gratefully acknowledge the support for this research provided by
the Australian Research Council.
2
See Van Ness et al. (2001) and Strang and Braithwaite (2001) for overviews of this
movement and descriptions of related genres such as ‘victim-offender mediation’,
‘family group conferencing’ and ‘circle sentencing’.
3
Under the New South Wales Young Offenders Act (1997), some offences (for
instance, drug-related crime and offences involving serious assault or sexual
violence) are excluded for the conferencing programme. However, similar pro-
grammes in other jurisdictions (e.g. South Australia) do include some of these
more serious matters.
4
See Moore and McDonald (2001) for an account of how the genre developed in
early trials. Our comments here are also based on participant-observation of four
training workshops conducted by the NSW Department of Juvenile Justice.
5
We have documented some variation in the sequence of participants’ recounts
and rejoinders; and some convenors have already canvassed in some detail the
major elements of an outcome plan in pre-conference meetings with the different
parties so that the break becomes a time for filling out paperwork (documents to
go back to the children’s magistrates court, participant survey forms etc.).
6
Convenors have read the police record of events, and have met with the police,
young person and victim/s before the conference, and so are fairly familiar with
the offence.
7
‘A person has committed the offence of affray when they use or threaten violence
towards another and whose conduct would cause a person (of reasonable firm-
ness) to fear their personal safety’ (Criminal Law Consolidation Act 1935, s83C).
8
In general our male adolescent young offenders speak very softly, making tran-
scription difficult, and causing hearing difficulties even for those present at the
conference.
9
Colloquial expression referring to the offer of a one-on-one fight (duel).
10
Disappointed is double coded as affect and judgement here, since Dad is con-
strued as feeling sad about something his son has done wrong.
11
Sorry and apologized are also double coded for affect and judgement, since it
entails the speaker feeling sad about something he/she has done wrong.
12
Many restorative justice advocates invoke as precedents for these innovations in
Western legal systems the example of dispute resolution customs in Melanesian,
indigenous Australian or other ‘traditional’ cultures. While, as Cuneen (2002)
has pointed out, these comparisons can be very misleading, conferencing, as a
‘fragmented’ form of justice has at least proved ‘flexible and accommodating
toward cultural differences’ (Daly, 2001, p. 65) and its early development in
New Zealand was certainly seen as part of a wider political response to the over-
representation of young Maori and Pacific Islander people in the criminal justice
system.
References
Ahmed, E., Harris, N., Braithwaite J. and Braithwaite, V. (2001). Shame Management
through Reintegration. Cambridge: Cambridge University Press.
Bednarek, M. and Martin, J. R. (eds) forthcoming. New Discourse on Language: Func-
tional Perspectives on Multimodality, Identity, and Affiliation. London: Continuum.
Bell, C. (1992). Ritual Theory, Ritual Practice. New York and Oxford: Oxford Univer-
sity Press.
Bell, C. (1997). Ritual: Perspectives and Dimensions. New York and London: Oxford
University Press.
Bernstein, B. (1975). Class, Codes and Control, Volume 3: Towards a Theory of Educa-
tional Transmissions. London: Routledge and Kegan Paul.
Braithwaite, J. (1989). Crime, Shame and Reintegration. Cambridge: Cambridge Uni-
versity Press.
Braithwaite, J. and Mugford, S. (1994). ‘Conditions of successful reintegration
ceremonies: Dealing with juvenile offenders’. British Journal of Criminology 34 (2),
139–71.
Douglas, M. (1973). Natural Symbols. New York: Random House.
Handelman, D. (1990). Models and Mirrors: Towards an Anthropology of Public Events.
Cambridge: Cambridge University Press.
Hayes, H. (2006). ‘Apologies and accounts in youth justice conferencing: Reinter-
preting research outcomes’. Contemporary Justice Review 9 (4), 369–85.
Hoyle, C., Young, R. and Hill, R. (2002). Proceed with Caution: An Evaluation of the
Thames Valley Police Initiative in Restorative Cautioning. York, UK: York Publishing
Services.
Lewis, J. L. (2008). ‘Toward a unified theory of cultural performance: A reconstruc-
tive introduction to Victor Turner’, in G. St John (ed.), Victor Turner and
contemporary cultural performance. New York: Berghahn, pp. 41–58.
Martin, J. R. (2008). ‘Intermodal reconciliation: Mates in arms’, in L. Unsworth
(ed.), New Literacies and the English Curriculum: Multimodal Perspectives. London:
Continuum, pp. 112–48.
Martin, J. R., Zappavigna, M. and Dwyer, P. (2009). ‘Negotiating shame: Exchange and
genre structure in youth justice conferencing’, in A. Mahboob and C. Lipovskpy
(eds), Studies in Applicable Linguistics and Language Learning. Newcastle: Cambridge
Scholars Press, pp. 41–84.
(revised edn). London: Continuum.
Equinox.
Martin, J. R. and Stenglin, M. (2007). ‘Materialising reconciliation: negotiating
difference in a post-colonial exhibition’, in T. Royce and W. Bowcher (eds), New
Directions in the Analysis of Multimodal Discourse. Mahwah, NJ: Lawrence Erlbaum
Associates, pp. 215–38.
English. London: Palgrave.
Maxwell, G. and Morris, A. (2001). ‘Family group conferences and reoffending’, in
A. Morris and G. Maxwell (eds), Restorative Justice for Juveniles: Conferencing, Media-
tion and Circles. Oxford, UK: Hart Publishing, pp. 243–66.
Moore, D. and McDonald, J. (2000). Transforming Conflict in Workplaces and Other
Communities. Sydney: Transformative Justice Australia.
Moore, D. and McDonald, J. (2001). ‘Community conferencing as a special case of
conflict transformation’, in H. Strang and J. Braithwaite (eds), Restorative Justice
and Civil Society. Cambridge: Cambridge University Press, pp. 130–48.
Morris, A. (2002). ‘Critiquing the critics: a brief response to critics of restorative
justice’. British Journal of Criminology 42, 596–615.
Nathanson, D. L. (1997). ‘From empathy to community’. The Annual of Psycho-
analysis 25, 125–43.
Palk, G., Hayes, H. and Prenzler, T. (1998). ‘Restorative justice and community
conferencing: summary of findings from a pilot study’. Current Issues in Criminal
Justice 10 (2), 125–37.
Rappaport, R. (1999). Ritual and Religion in the Making of Humanity. Cambridge, UK:
Cambridge University Press.
Retzinger, S. and Scheff, T. (1996). ‘Strategy for community conferences: emotions
and social bonds’, in B. Galaway and J. Hudson (eds), Restorative Justice: Interna-
tional Perspectives. Monsey, NJ: Criminal Justice Press, pp. 315–36.
Stenglin, M. (2004). ‘Packaging curiosities: towards a grammar of three-dimensional
space’. Unpublished PhD thesis, Department of Linguistics, University of
Sydney.
Strang, H., Barnes, G., Braithwaite, J. and Sherman, L. (1999). Experiments in Restor-
ative Policing: A Progress Report on the Canberra Reintegrative Shaming Experiments
(RISE). Canberra: Research School of Social Sciences, Australian National
University.
Trimboli, L. (2000). An Evaluation of the NSW Youth Justice Conferencing Scheme.
Sydney: NSW Bureau of Crime Statistics and Research.
Turner, V. W. (1982). From Ritual to Theatre: The Human Seriousness of Play. New York:
Performing Arts Journal Publications.
Van Ness, D., Morris, A. and Maxwell, G. (2001). ‘Introducing restorative justice’, in
A. Morris and G. Maxwell (eds), Restorative Justice for Juveniles: Conferencing, Media-
tion and Circles. Oxford, UK: Hart Publishing, pp. 3–16.
Chapter 5
Modelling Social Affiliation and Genre

in the Civic Domain
Sally Humphrey
5.1 Introduction
Developing models of context which can inform literacy education has been
a central concern of systemic functional linguists. Halliday’s (e.g. 1991) model
of social activity recognizes two aspects of context. Context of culture refers to
the broader background against which the text has to be interpreted (Halliday,
1985, p. 46) and context of situation can be glossed as the immediate environ-
ment in which the text is actually functioning (Halliday, 1991, p. 8). These
dimensions offer a powerful foundation for exploring the literacy practices
of young people within institutions such as schools and universities as well
as within more loosely bounded affiliations beyond schooling. This chapter
reports on one aspect of a study (Humphrey, 2008), which has built on Systemic
Functional Linguistics (SFL) models of context to account for the powerful
forms of persuasion enacted by adolescents in what I have termed the ‘civic’
domain of their literacy lives. In so doing, it demonstrates the application of
these models for making visible the persuasive goals of texts.
Models of context developed within SFL educational research have fore-
grounded the discursive reading of context of culture in proposing genre as
‘staged, goal oriented purposeful activity in which speakers engage as members
of a culture’ (Martin, 1984, pp. 24–5). Martin’s model allows both for a catego-
rizing of genres according to global characteristics and for the linguistic pat-
terning of valued curriculum genres to be made visible. Sydney School genre
theorists have also used the model to map pathways of language development
across a range of secondary school subjects (e.g. Coffin, 2006; Macken-Horarik,
1996; Veel, 1997).
Typically, pathways of literacy development see genres as developing from
more familiar spoken and everyday genres such as personal and autobiographi-
cal recount and observation to institutionalized, written and specialized genres
which privilege reasoned argument, such as analytical exposition. In terms of
tenor, valued persuasive genres are those which eschew emotion, intuition and
Modelling Social Affiliation 77
personal experience and privilege the logical, objective and ‘factual’. Although
recent studies using the analytical tools of appraisal have made visible the
importance of evaluative stance in persuasive writing (Coffin, 2006; Hood,
2006), valued evaluation tends to be realized covertly – in institutional and
impersonal terms.
While descriptions of genres and hierarchies have been very effective in
enabling teachers to build the necessary metalinguistic knowledge for students
to access the valued language of the school curricula, the emphasis on relatively
stable academic genres has resulted in models of genre classification which do
not fully account for emerging, blended and less stable genres created in other
domains of adolescents’ lives. Kress (2003), for example, warns of the tensions
between regularity and convention and of the dynamic for constant flux and
change, resulting partly from ‘the constantly transformative action of people
acting in ever-changing circumstances’ (p. 102).
One area where this tension has been particularly obvious is in descriptions
of effective persuasion in environments beyond schooling and particularly
in sites of public debate (McCormack, 2003). Martin (1995), for example, has
found that although the expository texts he examined in these sites did involve
recontextualizations from the specialized domain of academic language use,
they also drew on linguistic resources which he describes as ‘emotive, alive
and oriented to change’. Social and political theorists (e.g. Melucci, 1996;
Maddison and Scalmer, 2006) also recognize that specialized knowledge is
diverse and may be built through interactions within social and political
affiliations as well as through institutionalized learning in schools and
workplaces.
In recent years genre theorists have responded to challenges such as these by
extending their models to account for genres created in response to social
changes within and beyond schooling. Significantly, Martin (2002, p. 20) argues
that the diversity of social factors at play in such discourse change necessitates
richer models of genre including layerings of genre families within macrogene-
ric configurations he glosses as universes and galaxies of meaning.
Consideration of complex configurations of genre has been complemented
by research within and beyond SFL which has sought to relate textual practices
to the communicative goals and roles within broad cultural spheres (Maddison
and Scalmer, 2006; McCormack, 2003; Macken-Horarik, 1996), institutions (e.g.
Iedema, 1997) and more delicately within social affiliations (Gee, 2000; 2005)
and discipline areas (e.g. Coffin, 2006). This work has enabled the literacy prac-
tices of adults and adolescents in educational institutions to be characterized
in relation to those of other domains in their literate and multiliterate lives.
In the following sections of this chapter I will first explicate the model of
cultural context developed from both social and semiotic perspectives to make
visible the persuasive goals of texts produced by a group of adolescent activists.
Following that, I examine the unfolding of these appropriated texts from the
perspective of genre, illustrating how their persuasive purposes can be made
visible through semiotic analysis which includes an interpersonal as well as

experiential dimension.
5.2 Situating the Study
The study reported in this chapter examined the literacy practices of adoles-
cents who are engaged critically with issues of concern beyond the school
curriculum in order to make visible to educators the semiotic resources realiz-
ing persuasion at the level of genre and at the discourse semantic level.
Methodologically, the study is situated within the tradition of Positive Discourse
Analysis (PDA), which places emphasis on the interpretation and celebration
of texts which ‘make the world a better place’ (Martin, 2004, p. 177).
The source of data discussed here was a grassroots social movement called
Chilout, which was set up in 2001 to lobby the Australian government to
free children and their families from Immigration Detention Centres (IMCs).
Central to the success of this campaign were texts produced as speeches, inter-
views and newspaper commentaries by refugees who had themselves been
detained in IMCs. The stories of these ‘insider’ activists have been attributed to
changing public attitudes towards detainees by bringing ‘anonymous, faceless
men, women and children to light and life’ (Ozdowski, 2004). An example of
such a text, produced by an 18-year-old Afghani ‘Chilout Ambassador’ at a
speech at a major metropolitan rally is shown below.
Text 1: HSS1 – Speech delivered at World Refugee Day Rally

In the name of God the most merciful and compassionate. Good morning
ladies and gentlemen. My name is Sayed Reza. I am honored to be given the
opportunity to speak here and I am thankful to the organizers. At the start
I would like to say that I am a refugee from Afghanistan and I left my country
in 2000. I was only 14 years old. Since then I have not had contact with my
family and I do not know really what is going on in Afghanistan where are
they live in Afghanistan are they alive or dead. I hear this bad news all the
time about my homeland. I am worried about my family that I left behind.
It was a very bitter moment when I left all my family behind but sometimes
there are things that one has to do for his survival. When I was in Afghanistan
my life turned dark. I lost my older brother to the war by extremist groups.
The extremist groups persecuted everybody; the War-lords changed the
country into a blood battle. There was no peace at all and my life was in
danger. My father sent me out with the hope to get me to safety, security and
a future. It seems like a nightmare for ordinary parents, in normal condi-
tions to hand over their child to a smuggler who had links with the extremist
people to take me out of the country. But this is the real story most of us have
experienced in our lives.
How should I explain my separation with my family and how I was put this
way and how I took a risky and dangerous trip, this is another long, long and
sad story. But what I can say is that the trip was full of danger unsafely and
insecurity. I could see moments that I felt I would die. Finally our boat
arrived in Australian waters. Where I was hoping to be safe and welcomed.
We knocked on Australian door, hoping to find safety and security. I was
hoping to be among people who have a very good reputation in generosity
and humanity. It is the first time in my life I am coming to understand what
peace and security are here in Australia. I have realized the value of a human
when I see all these people who looked after me it is the first time that I can
study and I can learn. However for three years I held a temporary protection
visa that put me in limbo. Luckily I have now been given permanent resi-
dence. But I feel for those who are on temporary visas, because their lives are
tortured by uncertainty and fear. I beg the government to end the system of
temporary visas. I feel a great deal of responsibility to contribute to Australia
in the future if I am given a chance. I owe a lot to all of you for your thoughts
and sympathy. This is not something I would be able to do in my own coun-
try of origin where democratic rights are not upheld.
Unfortunately, Afghanistan is still not a country where people can speak
freely and openly and without fear of persecution. I, like many, of my fellow
country men and women have suffered from many years of war and political
unrest for many years Afghanistan was ruled by a fanatical and extremist
group. They ruled in the name of Islam but their cruel and bloody ways had
nothing to do with the teachings of our holy prophet, Mohammad. During
this terrible time many people in Afghanistan lost all hope to have a better
life and future for themselves and their children. People lost every thing –
their livelihood, their homes, and even their loved ones. Many had to flee
and go into hiding. This happened to my family. We lived from day to day
not knowing what would become of us.
Fortunately for me the door of Australia opened. Australians heard my cry
for refuge and with open hands delivered me from the terror of my oppres-
sors. As you are listening to me, I extend my gratitude to you for recognizing
my plight, and the plight of my countrymen and women, and for giving us
shelter. Unlike the country I left behind Australia is a country which upholds
and values human right – the rights to peace, compassion and freedom.
During my time in Australia I have enjoyed the opportunity of an educa-
tion. Because of the war I was never able to go to school. Now I am a student
at Holroyd High School. All my teachers have shown much dedication
towards me and I thank them. Now that I am a permanent resident I can
look forward to the future without fear. I hope others on temporary visas will
be shown the same mercy.
5.3 Preliminary Analysis of Genre
In accordance with the principle that differences in purpose will be reflected in

the patternings of language across texts, genre theorists argue that the starting
point for assigning texts to genre is an analysis of their recurrent global pat-
terns (Martin and Rose, 2008, p. 8) or generic structure (Eggins and Martin,
1997, p. 290). In educational descriptions of genre, the point of departure for
such categorization has tended to be field, which relates to the nature of the
activity going on. For example, Martin and Rose (2008) distinguish between
recount genres used in school history according to criteria such as how time is
realized and whether participants are individual or generalized.
From the perspective of field, Text 1 can be broadly characterized as narra-
tive because of its temporal organization and focus on specific things, people
and places. Like autobiographical recounts, which are typically used in the
academic domain to apprentice students into the curriculum area of history
(Coffin, 2006), the Ambassadors’ texts are built around a sequence of events in
their lives. From this perspective, the typical stages of the texts and the func-
tions of these stages can be represented as follows:
Orientation: introducing the narrator and setting (and greeting)

Background: describing the lives of narrators and their families prior to
beginning their journey to Australia
Record of events: retelling and evaluating significant events of
the journey to Australia
time spent in IMCs
Reorientation: describing the present (happy) situation and future
aspirations.
Some of the young refugees’ texts also include an Appeal stage which func-
tions to directly request action from the audience. However, as will be further
discussed below, this stage is not always realized explicitly.
The recognition of features associated with autobiographical recount is sig-
nificant to educators concerned to make links between the genres deployed by
adolescents in their academic and civic domain practices. However in order to
fully account for the rhetorical function of genres such as Text 1, it is necessary
to consider the influence of the multilayered social spaces and affiliations in
cultural environments beyond schooling as well as the macrogeneric configura-
tions in which the texts are created and read.
5.4 Modelling the Cultural Environment of

Adolescent Literacy Practices
While recognizing that the boundaries between contexts of culture and situa-
tion are not easily defined (Halliday, 1991, p. 9), the model described here is
situated within context of culture in order to account for the multiple layers of
genre identified in the analysis of texts. This reserves context of situation as the
immediate environment of the texts, construing and construed by linguistic
features at the level of discourse semantics (see Humphrey, 2008 for a discus-
sion of the realization of context of situation through the resources of
appraisal).
5.4.1 Cultural domains

Drawing on frameworks proposed by Macken-Horarik (1996) and McCormack
(2003), the literacy lives of adolescents can be understood in terms of four
loosely bounded domains. Each of these can be viewed from an external (mate-
rial) and internal (semiotic) perspectives. Figure 5.1 presents these domains
and outlines their broad communicative goals.
The conceptualization of the civic domain, with its focus on the practices of
non-voting citizens, can be seen as complementing the formal political sphere
within the broader space of the public sphere posited by Habermas (1979).
5.4.1.1 Communicative goals and predicted genre configurations

of the civic domain
The communicative goals of adolescents within the civic domain are complex
and multifaceted. In addition to the debate of public views and actions, core
business of the civic domain includes building consensus around shared values,
both to nourish solidarity within social movements and to ignite support from
broader audiences. Such goals are achieved through deploying interpersonal
(Specialized)
specialized
Figure 5.1 Cultural domains of adolescents’ lives

resources which are sensitive to the positioning of the audience and the rela-
tionship between the rhetor and the audience. A repertoire for enacting these
goals needs to include resources for building specialized and technical knowl-
edge in academic and workplace domains as well as those for building relation-
ships in the personal/social domain. The highly stylized and evocative rhetorical
resources deployed by politicians to align audiences around common goals and
values in the broader public sphere are also vital in construing persuasion in
the civic domain.
In light of these communicative purposes, it is predicted that the genres likely
to be ‘at stake’ in the civic domain are those which persuade the audience,
either to carry out some social action, that some social action (by somebody
else) is needed or that the position and evaluative stance of the rhetor on an
issue is valid. Significantly, however, genres within the civic domain need to be
understood in Bakhtinian terms (Bakhtin, 1935 [1981]), as being uttered in
relation to other texts – as arguments within a debate.
As is shown on Figure 5.2, the civic domain itself can be seen from a semiotic
perspective as a discussion genre complex (supra-genre), consisting of multiple
layers of genre configurations.
5.4.2 Social affiliations within the civic domain

While framing the literacy practices of adolescent social activists within the civic
domain provides a useful way of situating their broad persuasive purposes,
more delicate descriptions of the context are needed to account for the influ-
ence of social processes associated with particular social movements, groups
and spaces within this domain. Gee’s (2000; 2005) concept of social affiliation,
which can be seen in terms of Affinity space and Affinity group, is conceived
as an intermediate space along a continuum of context. Social affiliations
are characterized by a focus on common endeavour, by the prominence of
solidarity and shared bonds and by their potential to become ‘morally heated’
(Beck, 1994, p. 105 cited in Gee, 2000). Affinity group foregrounds the shared
purpose of the text producers while Affinity space (Gee, 2005) acknowledges
the distinct influence of the sites or forums on textual practices. Gee (2005)
implies the notion of space as an alternative to that of community or group;
however, both dimensions of social affiliation have a complementary role in the
analysis of the complex configurations in which adolescent activists participate
and in explaining the genres and registers of the texts produced.
5.4.2.1 Communicative goals and genre configurations

of the Chilout social affiliation
The lobby group Chilout can be conceived as a loosely bounded Affinity group,
whose diverse participants were united around a defined goal of persuading
the federal government to change its immigration policy. Although an analysis

of the rhetorical realization of the Chilout campaign is beyond the scope of
this chapter, this goal suggests that the campaign itself functioned generically
as a hortatory exposition, persuading the audience to carry out some social
action. As with the discussion genre which construes the civic domain, this
genre needs to be seen as a genre complex (mega-genre), instantiated in
turn by complex genre configurations (macro-genres), which construe the dif-
ferent Affinity spaces of the campaign (e.g. rallies, meetings with politicians,
web pages).
The macro-genre which construes the Affinity space at which Text 1 was pro-
duced serves a hortatory function, with the central Appeal stage realized promi-
nently on banners, slogans and posters (e.g. ‘Free the Refugees’, ‘Children out
of Detention!’). The speeches of the young refugees as well as those delivered
by adult refugees and ‘expert’ witnesses instantiate a range of elemental persua-
sive genres, which are recontextualized as the Argument stage of the macro-
genre (see Figure 5.2).
5.4.3 The role of social identity in determining genre

The structure of elemental genres within civic domain social affiliations depends
to a large extent on the roles of the rhetors – what Gee (2000) refers to as Social
Identity and Halliday (1985, p. 46) recognizes as institutional role. The social
identity of the young refugees within the Chilout campaign can be broadly
termed ‘activist’ because of their interest in changing the status quo. However,
the ‘insider’ ideological status of these young activists as Ambassadors enables
them to enact two specific social roles – as witness and advocate. Both roles
involve building credibility, rapport and empathy with potentially hostile audi-
ences; however, broader contextual variables such as the type of Affinity space
influence which role of their activist identity is foregrounded.
While all dimensions of social affiliation impact upon the choice of genre, it
is the dimension of social identity and the more delicate roles involved which
largely determine the structuring of the individual texts of the Ambassadors
as recount. According to Rothery and Stenglin (1997, p. 239), recount is reas-
suring to an audience and is often used to entertain and create solidarity
among groups of people. However, unlike recount in the personal/social and
academic domains, the intertextual relationship of the elemental genres of
the Chilout Ambassadors within complex expository genre configurations
can be seen as positioning the audience to ‘read’ the Ambassadors’ texts as
persuasive, even when the Appeal is not foregrounded. In this way, the social
identity of the rhetors made available by their social affiliations and cultural
domain significantly extends the meaning potential available to them as social
identities within other domains of their lives. The positioning of the Ambassa-
dors’ texts within the layered genre configurations and social affiliations is
shown in Figure 5.2.
Figure 5.2 The position of elemental genres within Chilout genre complexes
5.5 Recontextualizing Recount as Political Testimony
The perspectives of text and context described above allow for Text 1, in
common with other texts produced by the young refugee activists within the
Chilout campaign, to be categorized in broad generic terms as testimony – a
genre which appropriates structural and linguistic features of autobiographical
recount for rhetorical use, but which, unlike autobiographical recount, focuses
on communal socio-political conditions and actions. Like legal testimony, the
texts of the Ambassadors function to ‘witness justice or truth’ (Martin and Rose,
2008, p. 55) and allow the voice of victims of injustice to be heard in the
civic domain. However, the overt persuasive function of the Ambassadors’
texts distinguishes them as political testimony, which, like the ‘testimonio’ of
Latin-American activists, calls on the reader to join the struggle to overcome
oppression (Jehenson, 1995, p. 141). As political testimony, events are recounted
from the perspective of both marginalized victim and witness of injustice and
advocate for change.
From an ontogenetic perspective (Martin, 2001; Coffin, 2006), it can be
argued that the appropriation of recount is determined largely by the reper-
toire of linguistic resources available to the Ambassadors, most of whom had
been learning English for fewer than three years. However, the variation can
also be seen as a response to the opportunities offered to the rhetors as ‘insider’
social identities within the Chilout social affiliation. The choice of testimony
genre allowed the activists to use their powerful speaking position to break
down perceptions of asylum seekers as queue jumpers and illegal immigrants

who lacked gratitude and to show ‘the human face of refugees’ (Ozdowski,
2004).
5.5.1 Unfolding of persuasion in political testimony

As is evident in the discussion so far, the relationship of the young refugees’
texts to other texts within the cultural context allows for their characterization
as broadly persuasive, despite their linguistic realization as recount. These
persuasive purposes are also reflected clearly in the unfolding structure of the
text, particularly if the analyst recognizes the contribution of interpersonal
meanings on text structure (Iedema, 1997). From this perspective, political
testimony can be seen as moving through an Appeal element with a number of
supporting Motivating elements, which can be seen as appropriated stages of
recount (see Figure 5.3).
5.5.1.1 Appeal element

The Appeal is an interpersonally oriented element, which realizes a request for
action (i.e. to free children and their parents from IMCs). In the testimonies of
some of the young Ambassadors, the Appeal is realized as a discrete stage, while
in others, such as Text 1, it is realized prosodically through ‘petition’ phases,
which reinforce the prominent Macro-Appeals on the banners and posters of
the rally. For example:
I beg the government to end the system of temporary visas. [HSSI]

I hope others on temporary visas will be shown the same mercy. (HSS1)
The prominence of the Appeal within individual texts can be attributed to

whether the rhetor has foregrounded his or her social identity as advocate or
witness. Typically, texts produced in more public Affinity spaces where media
prominence was high do include direct Appeal elements.
5.5.1.2 Motivating elements

In addition to the Appeal, Motivating elements which foreground experiential
meanings (e.g. reasons) as well as interpersonal meanings (e.g. evaluations,
authority) play an important role in persuading the audience. Recognition
of these multiple persuasive roles supports rhetorical theories of political dis-
course which recognize the importance of appealing to the audience from a
number of perspectives – i.e. logos, pathos and ethos (Halmari and Virtanen,
2005). From a logogenetic perspective, it is possible to discern a movement
across the Motivating elements from those which foreground authority and
rapport (Identification), to those which foreground reasoned justification
(Legitimation) and emotion (Involvement).
5.5.1.3 Identification
From the perspective of genre theory, Motivating elements are conceived as
appropriations of constituent stages of recount. The Identification element,
which functions experientially to orient the audience to events, can be seen
from an interpersonal perspective as establishing the credibility of the rhetor
in terms of ‘what matters’ to the largely middle-class Western audience. In
this sense it is the appeal to ethos which is foregrounded, primarily through
making explicit to the audience the relevant dimensions of status. For the
rhetor of Text 1, the social identity of refugee is considered the major source
of credibility. For example:
At the start I would like to say that I am a refugee from Afghanistan and I left
my country in 2000. [HSS1]
However, for a number of other ambassadors, the identity of student is also

foregrounded. For example:
I am an 18-year-old female refugee from Bamiyan, Afghanistan. I am in year

12 at Holroyd High School and I am studying for my HSC. [HNN]
This dual identification can be seen as a complex bid for credibility on the basis
of the rhetor’s authority as an insider witness to events and as an insider of the
educational institutions which matter to the Australian audience, i.e. a success-
ful student. Relating to the broad Australian audience in this way, as well as
greeting and acknowledging that audience, is also evidence of the role of the
Identification element in building rapport or solidarity with the audience.
5.5.1.4 Legitimation
The Legitimation element, which experientially provides Background informa-
tion about the lives of the young people prior to their journey, can also be seen
from an interpersonal perspective as a response to public perceptions of asylum
seekers as economic refugees or queue jumpers. Through this element, events
and conditions within Afghanistan are recontextualized as evidence to support
the decision to leave the country and thus justify the status of refugee. Appeals
to logos, in the form of reasons for leaving, are foregrounded.
My father sent me out with the hope to get me to safety, security and a future.
It seems like a nightmare for ordinary parents, in normal conditions to hand
over their child to a smuggler who had links with the extremist people to take
me out of the country. But this is the real story most of us have experienced
in our lives. [HSS1]
As is evident in this excerpt, the reasoning provided by the rhetors typically

rests on grounds of threats to security. In the texts of the female rhetors, how-
ever, threats to educational opportunity and freedom for women are also prom-
inent. The choice of these factors shows an astute awareness of the values held
by the Western target audience. This focus is shown in the following excerpt.
We left Afghanistan because of civil war, persecution, ethnic cleansing of my

people, the Hazara, the dangerous environment and the unfair treatment of
girls and women. We children had no educational opportunities at all. [HNN
Legitimation element]
From a lexico-grammatical perspective, the appeal to logos in the Legitimation

is realized through logico-semantic relations of enhancement (e.g. because
of . . . ). Although events throughout the text are recounted in the first person,
‘I’ and ‘we’ frequently represent entire social categories (i.e. asylum seekers,
children, parents) and not just the individual rhetor and their families.
Despite the move towards recontextualizing personal events as generalized
reasons, however, the Legitimation element unfolds temporally through phases
typical of the narrative and recount genres valued in the personal/social
domain. The interaction of ‘problems’ phases, dealing with the untenable con-
ditions in Afghanistan and ‘solutions’, dealing with the decision to leave and
come to Australia engages the audience in an emotional journey, arousing and
releasing tension (Rose, 2007). Through these unfolding phases, the audience
is positioned rhetorically to expect and thus approve of the decision to seek
refuge in Australia.
5.5.1.5 Involvement
The Involvement element, which is by far the most extended stage of these
texts, deals experientially with the significant episodes of the refugees’ journey
to Australia and the time spent in IMCs. Interpersonally, this stage is significant
in aligning the audience rhetorically through appeals to pathos.
Emotional appeals are realized through a number of interacting semiotic
resources across the Involvement element. As in the Legitimation element,
experientially oriented ‘problems’ and ‘solutions’ phases move the action for-
ward by swings of expectancy from phase to phase (Rose, 2007, p. 5). However,
also vitally important in positioning the audience emotionally are evaluatively
oriented ‘reaction’ phases which suspend the action in order to intrude the feel-
ings or comments of the participants (Macken-Horarik, 2003, p. 300). The inter-
action of these interpersonally loaded ‘reaction’ phases with the ‘problems’ and
‘solutions’ is illustrated in the following excerpt from a female rhetor [HNN].
Table 5.1 HNN: Interaction of evaluative and referential phases across Involve-
ment element
‘EPISODE 1’ A smuggler hid us in the back of a truck for our escape from Afghanistan to
‘problems’ Pakistan. Then we were smuggled to Indonesia where we had to stay in
hiding. My mother had to go to hospital to give birth. The rest of us were
locked in a terrible flat 24 hours a day, until it was our turn to get on the
boat. So we got into a little leaky fishing boat, more than 100 of us. I was
one of 30 children and babies on board.
‘reaction’ It took us 10 days to get to Australia – 10 days of horror, sadness, no food or
drink and so many worries about our future. The only music I heard in my
childhood in Bamiyan was the screaming with horror and mothers crying
for their children’s future and I heard it again on this boat. We were all
vomiting. My poor mother with a newborn baby was sick the whole way.
‘solutions’ Finally, in September 2000, our boat was guided by the royal Australian navy
and landed on Australian land safely.
‘reaction’ I was happy because my miserable life was over, and a new horizon with no
more death and killing was welcoming us.
Central to the realization of pathos across both experientially and evaluatively

oriented phases are the discourse semantic resources of appraisal (Martin and
White, 2005). While a detailed analysis of appraisal resources is beyond the
scope of this chapter (however, see Humphrey, 2008), explicit values of positive
or negative affect (e.g. horror, sadness, worries, happy, miserable) saturate
‘reaction’ phases and radiate these values across the less explicitly emotional
‘problems’ and ‘solutions’ phases.
5.5.1.6 Reassurance
The final Motivating element of political testimony, the Reassurance element
has a number of complex interpersonal functions, reinforcing the alignments
and rhetorical appeals made in earlier elements. One important function is
to release tension by bringing both the physical and emotional journey of the
rhetor to a happy ending in a ‘solution’ phase. The intensified positive emo-
tions associated with the present and future offer reassurance to the audience
that the difficulties and suffering encountered by the refugees are not insur-
mountable and cements the rapport initiated in the Identification element. For
example:
Australians heard my cry for refuge and with open hands delivered me from
the terror of my oppressors . . . Now that I am a permanent resident I can
look forward to the future without fear. [HSS1]
Figure 5.3 illustrates the typical structure of political testimony and the rela-
tionship of elements to the appropriated stages of autobiographical recount.
Figure 5.3 Orbital structure of political testimony
From this ‘orbital’ perspective (Iedema, 1997), the centrality of the Appeal,
whether textually or intertextually realized, can be modelled, with other ele-
ments functioning as supporting satellites to align the reader into shared
communities and to therefore motivate compliance.
5.6 Conclusion
The multilayered model of cultural context presented in this chapter allows for
the persuasive purposes of texts produced by adolescent social activists to be
made visible at the level of genre. While it is certainly clear that the Chilout
Ambassadors are still developing control of the range of genres needed to
engage fully in the civic domain, the analysis of their texts indicates that they
have developed a repertoire of resources which enable effective persuasion
across Affinity spaces of the Chilout campaign. In particular, by appropriating
the resources of recount genres, the activists were able to exploit their powerful
speaking position as insider victims to build solidarity with their audiences and
to ‘help sway public attitudes and opinions’ (Ozdowski, 2004).
References
Bakhtin, M. M. (1935: Reprinted in full 1981). The Dialogic Imagination: Four Essays
(C. Emerson and M. Holquist, trans.). Austin, TX: University of Texas Press.
Coffin, C. (2006). Historical Discourse: The Language of Time, Cause and Evaluation.
London: Continuum Discourse Series.
Eggins, S. and Martin, J. R. (1997). ‘Genres and registers of discourse’, in T. A. van

Dijk (ed.), Discourse as Structure and Process. London: SAGE publications,
pp. 230–56.
Gee, J. (2000). ‘Identity as an analytic lens for research in education’. Review of
Research in Education 25, 99–125.
Gee, J. (2005). ‘Semiotic social spaces and affinity spaces: from the age of mythology
to today’s schools’, in D. Barton and K. Tusting (eds), Beyond Communities of
Practice. New York: Cambridge University Press, pp. 214–32.
Habermas, J. (1979). Communication and the Evolution of Society (T. McCarthy, trans.)
London: Heinemann Educational.
Halliday, M. A. K. (1985). ‘Context of situation’, in M. A. K. Halliday and R. Hasan.
(eds), Language, Context, and Text: Aspects of Language in a Social-Semiotic Perspective.
Geelong, Victoria: Deakin University Press.
Halliday M. A. K. (1991). ‘The Notion of “context”’, in T. Le and McCausland M.
(eds), Language Education in Language Education: Interaction and Development (Pro-
ceedings of the International Conference held in Ho Chi Minh City, Vietnam,
1991), pp. 1–26.
Halmari H. and T. Virtanen (2005). Persuasion across Genres: A Linguistic Approach.
Amsterdam: John Benjamins.
Hood, S. (2006). ‘The persuasive power of prosodies: radiating values in academic
writing’. Journal of English for Academic Purposes 5, 37–49.
Humphrey, S. (2008). ‘Adolescent literacies for critical social and community
engagement’. Unpublished PhD thesis, University of New England.
Iedema, R. (1997). ‘The Language of administration: organizing human activity in
formal institutions’, in F. Christie and J. Martin (eds), Genre and Institutions: Social
Processes in the Workplace and School. London: Cassell, pp. 73–100.
Jehenson, M. Y (1995). Latin-American Writers: Class, Race and Gender. Albany:
State University of New York Press.
Kress, G. (2003). Literacy in the New Media Age. London: Routledge.
Macken-Horarik, M. (1996). ‘Construing the invisible: Specialized literacy practices
in junior secondary English’. Unpublished PhD thesis, University of Sydney.
Macken-Horarik, M. (2003). ‘Envoi: intractable issues in appraisal analysis?’ Text 23,
313–19.
Maddison, S. and Scalmer, S. (2006). Activist Wisdom: Practical Knowledge and Creative
Tension in Social Movements. Sydney: UNSW Press.
Martin, J. R. (1984). ‘Language, register and genre’, in F. Christie (ed.), Children
Writing: Reader. Geelong, Victoria: Deakin University Press, pp. 21–9.
Martin, J. R. (1995). ‘Interpersonal meaning, persuasion and public discourse:
packing semiotic punch’. Australian Journal of Linguistics 15 (1), 33–67.
Martin, J. R. (2001). ‘A context for genre: modelling social processes in functional
linguistics’. in J. de Villiers and R. Stainton (eds), Communication in Linguistics.
Toronto: Éditions du GREF, pp. l–41.
Martin, J. R. (2002). ‘A universe of meaning – How many practices? Response to
Grabe’, in A. Johns (ed.), Genre in the Classroom: Multiple Perspectives. Mahwah, NJ:
Lawrence Erlbaum Associates, pp. 269–78.
Martin, J. R. (2004). ‘Positive discourse analysis: solidarity and change’. Revista
Canaria de Estudios Ingleses 49 (Special issue on discourse analysis at work: Recent
perspectives in the study of language and social practice), 179–200.

Equinox.
McCormack, R. (2003). Common Units: Politics and Rhetoric Presentation to Indigenous
Minds Forum. Bachelor Institute, 22 July 2003.
Melucci, A. (1996). Challenging Codes: Collective Action in the Information Age.
Cambridge, UK: Cambridge University Press.
Ozdowski, S. (2004). ‘Australian human rights – inside and outside the wire’.
Address to the National Press Club, Canberra, 25 August 2004.
Rose, D. (2007). ‘Reading genre: a new wave of analysis’. Linguistics and the Human
Sciences 2 (1), pp. 184–204.
Rothery, J. and Stenglin, M. (1997). ‘Entertaining and instructing: exploring experi-
ence through story’, in F. Christie and J. R. Martin (eds), Genre and Institutions:
Social Processes in the Workplace and School. London: Cassell, pp. 231–63.
Veel, R. (1997). ‘Learning how to mean – scientifically speaking: apprenticeship
into scientific discourse in the secondary school’, in F. Christie and J. R. Martin
(eds), Genre and Institutions: Social Processes in the Workplace and School. London:
Cassell, pp. 161–95.
Chapter 6
‘News’ and ‘Register’:

A Preliminary Investigation1
Annabelle Lukin
A text is . . . the voice of its social process: it is through this voice that a social process is
known for the social process that it is.
(Hasan, 1999, p. 239)
A text is the form of the social relationship made visible, palpable, material.
(Bernstein, 1990, p. 17)
6.1 Introduction
The purpose of this chapter is rather simple: to bring Halliday’s notion of

‘register’ to bear on an instance of the category ‘news’. For Halliday, ‘register’
goes back nearly half a century, when he used it to refer to language ‘distin-
guished according to use’ (Halliday et al., 1964, p. 16). The category was consid-
ered necessary ‘to account for what people do with language’ (ibid.). In 1985,
Halliday described register as ‘a variety of language, corresponding to a variety
of situation’, with situation interpreted ‘by means of a conceptual framework
using the terms “field”, “tenor” and “mode”’ (Halliday, 1985, pp. 29, 38). This
claim located register in realizational terms; the term was, in addition, situated
in relation to the systemic functional linguistic (hereafter SFL) dimension of
stratification. Halliday located register at the semantic stratum, defining it ‘as a
configuration of meanings that are typically associated with a particular con-
figuration of field, mode and tenor’ (Halliday, 1985, p. 38).
Within the SFL framework, news discourse has been investigated by scholars
of the genre school (e.g. Iedema et al., 1994; Martin and White, 2005, among
others). For Martin, genre is a ‘staged, goal oriented social process’ (e.g. Martin,
1992). In this model, register ‘function[s] as the expression form of genre at
the same time as language functions as the expression form of register’ (ibid.,
p. 495). The implications of Martin’s notion of genre and his repositioning
of register for Halliday’s conception of language and context are discussed in
Hasan, 1995. A study showing the descriptive power of ‘genre’ compared with
‘register’ applied to instances of news discourse would be valuable, although
‘News’ and ‘Register’ 93
beyond the scope of a short chapter such as this.2 Approaching the study of
news discourse from a register perspective involves an explication of the
‘configuration of meanings’ typically associated with ‘a particular configuration
of field, mode and tenor’. It is an approach which enables us both ‘to under-
stand language . . . and to understand what people do with it’ (Halliday, 1985
[1989], p. 4), two preoccupations which are fundamental to the appliability
of Halliday’s linguistic theory.
6.2 News as Pedagogic Discourse
From a sociological point of view, news is important as a ‘fundamental social

context through which cultural reproduction-production takes place’ (Bernstein,
1996, p. 17). Although Bernstein did not give much direct attention to the
media in his account of code and pedagogic practice, his formulation was
intended to be wide enough to apply to any kind of socializing institution
(ibid.). His discussion of the ‘division of labour of symbolic control’, glossed as
‘the new professions which regulate mind, body, social relations, their special
contexts and temporal projections’ is relevant here (Bernstein, 1990, p. 133).
Symbolic control, he argued, is ‘the means whereby consciousness is given
a specialized form and distributed through forms of communication which
relay a given distribution of power and dominant cultural categories’ (ibid.,
p. 134). Bernstein makes a distinction between the field of production and the
field of symbolic control, the latter defined as ‘a set of agencies and agents that
specialize in discursive codes which they dominate’ (ibid.).
Bernstein located commercial forms of media in the field of production.
Media barons, such as Rupert Murdoch, are thus agents within the field of
production, engaged in the control over physical resources. Journalists within
agencies in the field of production are agents of symbolic control (ibid.,
p. 137). Commercial media are agencies which ‘market texts’; power is with the
marketeer(s), while the creator of the text – the journalist – is considered to
have ‘control over its expected realizations’ (ibid.). These agencies, Bernstein
argues, suffer the ‘intrinsic tensions arising out of the meeting of two differ-
ently specialized agents – one oriented to the production of physical resources,
the other to the production of discursive resources’ (ibid., pp. 137–8). In other
words, there is ‘a market principle between knowledge and the knower, between
the inner relation to, and the outer form of, knowledge’ (ibid., p. 2). Publicly
owned media, by contrast, were considered by Bernstein to be located within
the field of symbolic control. Important here is the degree of autonomy such
agencies have from government. Processes of symbolic control enacted by
agents in the field of symbolic control were considered by Bernstein, borrowing
from Foucault, to have an ‘explicit normalising function’ (ibid., p. 138).
The implications of considering news as a form of pedagogic discourse in
Bernstein’s terms are considerable, and yet to be laid out. If pedagogic dis-
course is the means by which ‘knowledge systems become part of consciousness’
(Bernstein, 1996, p. 17), then we need also to ask how media discourses act as
the carriers of external power relations, how ‘a dominating distribution of
power and principles of control generate, distribute, reproduce and legitimize
dominating and dominated principles of communication’ (ibid., p. 18)? Even
a perfunctory answer to Bernstein’s question awaits a considerably more expan-
sive forum than this chapter provides. But given Bernstein’s argument that the
structure of the discourse provides ‘the means whereby external power relations
can be carried’ (ibid., p. 18), register analysis has as its focus the ‘relevant mean-
ings’, through which such power relations can be carried.
6.3 Register Analysis
For lack of space, this chapter presents analysis of a single news item, broadcast
by the ABC (Australia’s public broadcaster), an agency within the field of sym-
bolic control. The news item was broadcast during the main evening 7 p.m.
bulletin, on 25 March 2003, five days after the invasion of Iraq (see Appendix 1
for the news item transcript). The ABC news at this time was overwhelmingly
focused on the invasion. This news item, the second of six devoted to Iraq, was
filed by the ABC’s embedded correspondent (on embedding, see Knightley,
2003). ABC news and current affairs sent 14 of its journalists to the Middle East
and Washington to report on the invasion of Iraq.
While a single text is, logically, an instance of a register, it clearly provides an
insufficient basis for claims about a type. A further difficulty is that, as Hasan
(2009) has noted, much of the SFL work drawing on field, tenor and mode has
done so in ‘the absence of “checkable” criteria’ and relying largely on ‘common
sense’. Including some of her own work in this critique, she argues ‘faced with
a text already there, the SFL linguists have largely been doing what any ordi-
nary speaker of language would do, i.e., construing from the language of the
text what the text is all about’ (ibid., p. 180). Hasan (1999) and Butt (2003) are
a response to this problem, in that they provide networks at the level of context.
Hasan refers to such network as ‘contextualisation systems’, and argues they
have the distinction that instead of taxonomising realized meanings, they

actually systemize the realization-instigating contextual features and attempt
to relate context to wording via meaning which acts as the interface between
the two. (Hasan, 2009, pp. 181–2)
Referring to her 1999 field network, Hasan suggests it is ‘not any worse than
the MOOD system networks drawn in the early 1960’s SFL’ which, through
‘continued use and discussion’ were gradually developed to the stage of today’s
networks (ibid., p. 185). My analysis will proceed from the three context
variables: mode, field and tenor, each treated as ‘a reservoir of “values”’ (ibid.,
p. 178). Hasan has used the term ‘contextual configuration’ (CC) to refer to
the specific values of the three variables, of which a given text – such as the one
to be considered in this chapter – is the realization. The statement of the CC
provides the context in which one can make statements about text structure.
A register description involves statements about which elements are obligatory,
and which optional, and the potential order and iteration of elements (Hasan,
1985). Again, a single instance precludes me from establishing a structure state-
ment for the register of which this text is an instance; I will offer no more than
an informal description of the apparent structural elements of this particular
instance. Nonetheless, the informal discussion of text structure will illustrate
one crucial distinction between ‘register’ and ‘genre’. Structure (and texture;
see below) from the perspective of register theory is a function of all three con-
textual variables; for Martin, genre is ‘made responsible for generating text
structure’ (Martin, 1992, p. 506). Genre itself is not a function of social context
(i.e. of field, tenor and mode), but rather ‘underlies’ it (ibid., p. 505). On this
basis Martin can argue that ‘to some extent, genres have a life of their own’
(ibid., p. 507).
Since register is a ‘configuration of meaning’, I have drawn on two kinds of
semantic analysis as evidence for claims I am making about the text. One is
Cloran’s Rhetorical Unit (RU) analysis (Cloran, 1994; see Appendix 1 for RU
analysis), a semantic unit which realizes ‘aspects of a text’s register’ (Cloran,
1994, p. 126). Since ‘units of all sizes at the level of semantics are responsive to
the contextual configuration as a whole’, then RUs express ‘the three kinds of
meaning realizing aspects of field, tenor and mode’ (ibid., p. 128). In addition,
I include a Cohesive Harmony analysis (Hasan, 1984) of the text, for its rele-
vance to texture, texture being, like structure, crucial to text unity (Hasan,
1985) (See Appendix 2 for cohesive harmony analysis; NB chain interaction has
not been displayed). Hasan has argued that texture, like structure, is a function
of the context of situation (e.g. 1984, 1985, 2004), specifically when the princi-
ple of ‘delicacy’ is brought into play. That is, ‘situation type, at a high degree of
specificity, is relevant to texture’. At the same time, ‘the facts of texture con-
strue the very detailed aspects of the situation in which the text came to life’
(Hasan, 1985, p. 115). I turn now to the discussion of the three parameters of
context, beginning with mode.
6.4 Mode
Mode is defined as ‘the nature of contact for the conduct of speaking’ (Hasan,
1999, p. 232), and elaborated in Hasan (1985) along the dimensions of role
of language, channel and process sharing. On the last of these, the text
is a kind ‘received in displacement from the location of [its] production’
(Hasan, 1999, p. 237). It comes to the viewer as an artefact. An implication
of the popular term ‘news’ is that the time between production and reception
is not inconsequential (‘today’s news is tomorrow’s fish and chips wrap’).
This time has consistently been reduced through technological development.

For instance, the embedded journalist who filed the news item analysed in this
chapter had the technology to shoot and produce a news item just 30 minutes
before it was to be broadcast. However, such technological change does not
obviate the fact the text lacks ‘process sharing’ (Hasan, 1985). The implication
of the displacement between production and reception is that the addressee is
not only ‘absent from the moment of production, but he is also an imaginary
being’, what Hasan terms a ‘virtual addressee’ (1999., p. 238). This has implica-
tions for tenor (see below).
In relation to channel, the text is met by its audience through both aural
and visual modes, via TV broadcasting. What is the relation between the two
modalities? A detailed response to this question requires an understanding
of the ‘functional nature of each code’ and how each code is ‘predispose[d] to
the encoding of certain relevancies rather than others’ (Hasan, 1996a, p. 38).
For instance, the visual by itself cannot encode the distinctions brought out by
a RU analysis (Cloran, 1994; see Appendix 1) between such units as Recount,
Report and Generalization. The ‘relevancies’ to which the visual mode is pre-
disposed clearly need consideration for analyses which claim to be multimodal.
I am not proposing to offer multimodal analysis here, although a rather obvi-
ous feature of the relations between the two modalities is their potential for
combination. Hartley (1982) describes the interactions of the modalities as
‘presentation modes’, distinguishing ‘voice over’ (where the journalist appears
outside of the material location presented in the visual mode), ‘vox pop’ (in
which some speaker is made relevant to the news item through the selection
of commentary interwoven with the text) and ‘stake out’ (where the journalist
speaks ‘on location’). See Figure 6.2.
A further point in relation to channel is that the text displays what Bateman
has called ‘cross modal resonances’ (Bateman, 2008, p. 42), not only in the
sense that the verbal is some construal of what is being presented concurrently
in the visual mode. The verbal text also directly indicates the visual mode, by
the use of what one might call ‘cross modal phoric reference’ (for instance, in
message 223, But this convoy wasn’t – a quick start foiled by a bad wrong turn, where
the ‘this’ is resolved by reference to the visual mode). The environments for
cross modal phoric reference, and its realization, is a matter for further
research. The text also displays instances of a kind of ‘recontextualized’ exo-
phoric reference. The environment for the use of such exophora is the ‘vox
pop’4 presentation mode (Hartley, 1982). In this presentation mode, one finds
a speaker making reference to what was, at the time of speaking, a material situ-
ational setting shared with his/her interlocuter. The function of such commen-
tary is ‘authenticity’ or ‘grounding’, that is, the appearance of connecting the
viewer to some material setting relevant to the news item. Thus, it is not surpris-
ing to find exophoric reference in this environment. Given that there is no
actual shared material situational setting (Hasan, 1973) between the text
interactants, the effect presumably is to recreate an ‘as if’ experience of a
shared setting.
Figure 6.1 Hartley’s presentation modes mapped onto Cloran’s continuum of

the role of language in the social process (Cloran, 1994, p. 132)
The RU analysis of the text (see Appendix 1) reveals some tendencies in mode
shift on the basis of the distinct interactions between the modalities described
by Hartley’s presentation modes. Figure 6.1 displays the RU types modelled
along a cline for the role of language in the social process, with the distribu-
tion of Hartley’s ‘presentation modes’ indicated along this cline (Hasan, 1999
considers role of language to be a function of field; it will therefore also be con-
sidered below in the discussion of field). The dominant RU for the ‘voice over’
is Report, the RU whose function is to ‘expres[s] the current non-habitual states
or activities of an absent person or object’ (Cloran, 1994, p. 109). For the ‘vox
pop’ it is Commentary, which realizes ‘a materially based role of language involv-
ing the monitoring of currently occurring states or activities of persons or objects
in the material here-and-now of the speech event’ (ibid., p. 94). When the jour-
nalist switches to ‘stake out’ (i.e. he speaks to camera from atop of the convoy in
which he is embedded), he provides a sense of the significance of what has been
told and shown in the news item, an explanation for why viewers should care about
the details of what has been reported. The RU at this point is Generalization.
6.5 Field
For a discussion of the field of the text – the nature of the social activity – I draw
on Hasan’s 1999 network for field (see Figure 6.3). Taking the system material
action first, the selection for this system is [non-present];* i.e. the experience
of the text is entirely semiotic. From the point of view of the system of [verbal
action], the text is [constitutive]; the ‘text’s language . . . acts as the essential
crystalliser’ of the experience of the context (Hasan, 1999, p. 239). The verbal
action, in consort with the visual modality, is all there is. In the choice that fol-
lows from constitutive, the text is [conceptual] rather than [practical], since
the option [practical] is oriented to the achievement of some material outcome
(ibid., p. 278). The option [conceptual] is the pathway to three further sets
of options: the choice between [relation based] or [reflection based], the
choice of [first order] or [second order] and the choice between [informing]
or [narrating].
*Square brackets denote a feature selection from a system network.

Figure 6.2 Hartley’s presentation modes (1982). Reading left to right 1. ‘voice
over’; 2. ‘vox pop’; 3. ‘stake out’5
Figure 6.3 Hasan’s 1999 network for field
Taking the first of these three options, the text is an instance of a [reflection-
based] activity. Note that [relation-based] activities tend to be those which
act as ‘tone setting sub-texts . . . run[ning] side by side like a prosody of the
on-going main activity’ (ibid., p. 289). [Reflection-based] activities produce
‘semiotic constructs such as explanations, generalisations, classifications and
descriptions of phenomena in the world of experience and imagination’ (ibid.,

p. 289). They position the speaker ‘as an observer vis a vis the already existing
material and social world’ (ibid., p. 290). Hasan further notes a common con-
ception of reflection-based activities is that they are ‘“about” something which
already exists and whose identity is independent of [reflection-based] verbal
actions as such’ (ibid., p. 290). Public debates about notions of journalistic
‘impartiality’ and ‘objectivity’ exemplify Hasan’s point. I suggest at this point
the text be considered an instance of a [reflection-based] activity, although
later I indicate evidence which could support an argument that the text is
[relation-based], though not in the sense of a ‘tone setting subtext’, but as the
main and single activity of the context.
Taking the option of [first order] and [second order], the text here would be
a [first-order] social activity; a [second-order] activity is one which depends
on an already existing text. Finally, there is the choice of [informing] versus
[narrating]. These options ‘capture respectively the significant division between
already experienced time and time that is in some sense present’ (ibid., p. 291).
The tendency in common parlance to describe a text of this kind as a ‘news
story’ might suggest that [narrating] would be the choice at this point of the
field network. The point of departure of the news item is oriented to past time,
and is a Recount in RU terms. But the specific event referred to in the opening
(the clash around Nasiriyah) is not picked up at all by the correspondent. It
receives no further attention in the text. Of the remaining four instances of
the RU Recount, only two function within the journalist’s ‘voice over’. The
remaining two are embedded within commentary from Sergeant Gomez. In
fact the ‘voice’ of the journalist is largely Report, that is, the discoursing on
‘the current but non-habitual states or activities of an absent person or object’
(Cloran, 1994, p. 134). Interestingly also, the text is rather muted with respect
to events or action. Only 20 of the total 55 messages of the text select [material]
as process type; and some of these processes analysed as material could also
have been called relational [. . . death now lies, The new day brings a new threat].
In addition, note the liberal use in the text of nominal group structures func-
tioning at clause rank (for instance, the correspondent’s point of departure,
Final preparations to ensure the security of a precious wartime cargo; see messages 7,
11, 31, 35, 50, 51, 52). Hasan refers to such structures as a ‘depictive’ nominal
groups (Hasan, 1964), a structure which can prefigure the narrating of events
by scene setting, but which in itself cannot construe an unfolding event, since it
lacks Finite and Predicator.
In addition, despite the cohesive conjunction construing time (Soon the break
in Iraq’s border approaches, in message 29), the text lacks temporal organization.
Indeed, what is being related temporally through Soon, in fact, is hard to inter-
pret: it would appear that ‘coalition invaders’ is a reference to the convoy with
which the correspondent is embedded, and which includes the character of
Sergeant Gomez. What precedes the use of Soon is an episode referring to a
distinct convoy, an episode internal to which there is a temporal frame (the
references to ‘night’ and ‘day’) but which overall has no temporal location.
In summary, the text lacks a focal event, or sequence that would one would
expect to be the basis for arguing the text construes the feature [narrating].
The cohesive harmony analysis reflects this absence. There are two chains
formed by continuity of material process. One, chain C takes in the material
processes of the text’s opening (messages 1–2), but then is picked up only
sporadically in the remainder of the text with four further central tokens, one
of which is a non-finite form, and two of which are the word ‘war’, an abstract
noun. Chain G is more extended, running from messages, 5–28, but then drop-
ping out of the text until the last two messages of the text. Chain G is therefore
a focal chain with respect to processness for the text. It shows that action for this
text is intransitive, since the processes essentially construe movement through
space. In addition, the chain develops largely by repetition and synonymy,
which I suggest could not support [narrating] as a contextual feature. While
there are two chains in the cohesion analysis which are features relating to time,
none of the terms in these chains form central tokens.
The alternative to [narrating] in Hasan’s network is the option [informing].
The option leads to the further choice between [commenting] and [describing].
The ‘currency of the states of affairs’ (Hasan, 1999, p. 296) is what distinguishes
these options. The selection [commentary] refers to the social activity of
‘constru[ing] states of affairs that are as it were located within the spatio-temporal
confines of the on-going interaction’ (ibid.). The option [describing] by con-
trast, construes states of affairs ‘whose currency goes beyond the here and now
of speaking’ (ibid.). That the journalist’s discoursing, from an RU point of view,
is largely Report (which, as noted above, expresses ‘the current non-habitual
states or activities of an absent person or object’ (Cloran, 1994, p. 109)), pro-
vides some evidence for arguing that with the combination of the verbal and
visual modalities, the text realizes the feature [commenting]. However, unlike
live commentary on a sporting match, the commentary in this case is organized
around visual material which has undergone selection. Elements in the text are
related not by an actual unfolding material event, such as in a football game.
Rather they are structured to display a kind of peripatetic development (see
below), as if there is some kind of natural link between the elements of the text.
Finally, with respect to Hasan’s field network, there is the question of sphere
of action, a system which provides the options [quotidian] versus [special-
ized]. The terms are considered end points of a cline, rather than as a categori-
cal distinction. The production of news, with its highly specialized division of
labour (e.g. Nanri, 1993) would indicate [specialized] is the appropriate term
from sphere of action.
6.6 Tenor
Three systems of tenor have been proposed: social distance, social/dyadic

relation and agent role (Hasan, 1985, 1999). Hasan (personal communica-
tion) suggests the terms the ‘text producing’ interactant and the ‘text receiving’
interactant for the interactants in the context of news presentation. As noted

above, the ‘text receiving interactant’ is virtual. As such, ‘the addressee is built
into the text as a prosody of its meaning and its structure’ (Hasan, 1999,
p. 237). Where an addressee is virtual, social distance must be maximal. The
social relation is non-hierarchic: viewers choose to watch the news, or not,
and which news to watch. The basis on which one comes to choose one news
source over another is not, however, accidental or arbitrary, but a function
of one’s life experience, in which social position is relevant (Hasan 1999).
In regards to agent role, the news item was broadcast in the evening bulletin,
so presumably the audience is largely adult. Typically, the ABC has 15 per cent
of the ‘prime time market share’ audience, compared with the 80 per cent
market share for the commercial stations.6 A niche market, one might say, to
whom the ABC ‘brand’ carries some meaning. The ABC audience, therefore,
opts for news produced within the field of symbolic control, rather than news
produced essentially as a marketing vehicle for commercial networks.
The ‘text producing’ interactant(s) are agents within the field of symbolic
control. The news anchor (who introduces the news item) and the correspon-
dent speak on behalf of an institution, the ABC. They have behind them the
authority of a public broadcaster, with a legislated requirement ‘to ensure that
. . . news and information is accurate and impartial according to the recognized
standards of objective journalism’ (Section 8, Australian Broadcasting Corpora-
tion Act, 1983). The institution’s independence from government has gener-
ated ‘a capital of trust’ and is the basis of ‘the Corporation’s legitimacy’ (ABC
Annual Report, 2002–2003, p. 13). In the year this news item was broadcast,
95 per cent of respondents to a News Poll survey believed the ABC 7p.m.
TV news to be ‘balanced’ and ‘even handed’ (Newspoll, 2003). While a public
broadcaster is not subject to the commercial pressures of private news net-
works/suppliers, news providers want an audience to stay switched on. The
ABC’s annual report for 2002–2003 reports its coverage of the Iraq war as one
of the ‘significant events’ for the year, reporting in particular the success of its
online reporting (‘Over 18 million pages of content were requested in just one
week following the announcement by the US of the commencement of the war
on Iraq’, ABC Annual Report, 2002–2003, p. 4). The unit of the news item itself
can be seen as a function of tenor; that is, a theory of what the viewer will bear
as a focused unit of reported news. Constant change from one topic to another
is a response to the function of keeping viewers tuned in to the news bulletin.
Each new item is a promise of ‘new’ information. And it must be consumable in
its own right. One does not typically find news presenters referring back to
events reported from previous bulletins, except in the situation where a correc-
tion is required.
A specific characteristic of the agent relation in the context of this news
item is that the correspondent was part of the US Department of Defense
embedding programme. Thus, another institution, the US military, is interpo-
lated into the agent relation. By becoming embedded, the ABC agreed to
their correspondent circumscribing aspects of his reporting.7 The architect of
the embedding arrangement, Deputy Assistant Secretary of Defense, Bryan

Whiteman, reported that the US military were very satisfied that their objectives
in establishing the programme were met.8
6.7 Register: Structure and Texture
It remains to be considered how the specific values of the three variables (mode,
field and tenor) can be seen to account for the text’s structure and texture. As
mentioned, my comments on structure will be informal. The cohesive harmony
analysis (see Appendix 2) has been annotated to indicate structural boundar-
ies, and the elements are numbered E1 to E7. The text’s opening unit (E1) is
the ‘hook’ for the story, although it does not function in the way the ‘lead’ ele-
ment is typically considered to function in theories concerned with structure
of news. Allen, for instance, defines the unit as ‘a summary or abstract of the
account’s essential “peg” or “hook” which projects, in turn, “the story” in a par-
ticular direction or angle’ (Allen, 2004, p. 83). E1 sets out from a purportedly
widespread issue of attacks on supply convoys, with a particular instance of an
ambush given to ground this claim. This element provides a preview only in the
most general sense with some continuity in chains A, B, D and E, although not
on the basis of co-referentiality; the chain in which the processes ‘attack’ and
‘ambush’ appear is very sparse, and contains no further instances of transitive
processes relating to this field. In E2, the presenter sets in location the
correspondent and his cameraman (giving the news item what Aristotle called
‘ethos’; note the correspondent is described as ‘travelling with’ rather than
‘embedded with’ US marines). This ‘scene setting’ continues as the correspon-
dent takes up the report (the cohesive harmony analysis showed no break here
despite the speaker change from presenter to correspondent). The correspon-
dent’s focus is the convoy whose function is to ‘keep marines fighting in Iraq’.
Messages 14 to 53 I have tentatively analysed as four elements (E3 to E6).
E3’s function might be described as ‘personalization’, with ‘Sergeant Gomez’
introduced as a token of the experience of young marines in Iraq. Note the
dominant process at this point is relational (Chain K).
E4 and E5 are weakly defined elements, on the basis of the cohesive
harmony analysis. E4 is a brief vignette of another convoy, based on a localistic
elaboration. This convoy is brought into the news item through a contrastive
link to Gomez’s excited assessment of the progress of the marines (We’re moving
pretty fast). The introduction of this new convoy brings into the text an instance
of a convoy reportedly under threat of snipers, a faint echo of the content of the
news report’s ‘hook’ element. The following message (#29) initiates element
E5. It begins with the news report’s one temporal cohesive conjunction Soon,
which purports to link what follows (i.e. the break in Iraq’s border approaches) by a
relation of time. But the viewer has now been returned to follow the events of
the convoy with which the correspondent is embedded, although one has to
infer this retrospectively on the basis that Gomez re-enters the news report.
Exactly what the event of the break in Iraq’s border approaching is ‘soon’ with
respect to is not clear. It is noteworthy that this ill-defined element construes
the process of the Coalition’s invasion of Iraq, an event already two days old. It
is as if the event has been recruited for dramatic/aesthetic purposes. Note that
the process of the invasion is construed through a middle voice (Halliday and
Matthiessen, 2004) selection, in which ‘Iraq’s border’ is the central participant.
In E6, the feelings of Gomez again take centre stage, in the reporting of his
reaction to the bodies the convoy has happened upon. Gomez’s range of reac-
tions is reported, as is his view that the killing must have been justified. The
links between E3 and E6 create structure through what seems a kind of peripa-
tetic relation: E3–E4 are connected through demonstrative deixis (this convoy);
E4–E5 are connected by time; E5–E6 are connected again through the deixis
of here, as if what is being reported is merely the stitching together of found
elements. Relations of this kind are potentially evidence for an argument that
the text is in fact not [reflection based], but rather [relation based], a point
requiring further investigation. The text closes with a statement of the signifi-
cance of what has been shown of this convoy’s experience (E7, messages 54–55).
The correspondent adopts the ‘stake out’ mode (i.e. he directly engages the
viewer by speaking to camera). The RU as noted is a Generalization, and con-
cerns the importance of the role of the convoys to sustain the American opera-
tion in Iraq.
Agencies and agents of the field of symbolic control ‘specialize in discursive
codes’ (Bernstein, 1990, p. 134). These discursive codes ‘specialize and distrib-
ute forms of consciousness, social relations and dispositions’ (ibid., p. 135).
With this in mind, what is the message of the text examined here?: the Coali-
tion is under threat; supply convoys are on a mission; the personal/emotional
reactions of individual marines constitutes a form of news; if threatened,
American forces will respond; American violence is a response to a threat, not
itself a threat; the act of invasion can be reported without specificity of time,
place or agency.9 The text appears to fulfil at least two of the functions of the US
Department of Defense embedding programme, including to ‘turn attention
back to the military’s role in the war, especially the part played by ordinary
American service men and women’, at the same time as ‘emphasiz[ing] the
dangers posed by the Iraqi regime’ (Knightley, 2003, pp. 530, 529).
6.8 Concluding Remarks
The analysis of one instance of one register of news presented in this chapter is
a preliminary attempt to understand the ‘social process’ of news. The analysis
has necessarily been brief and indicative only, but I hope sufficient to illustrate
the power of Halliday’s concept of register for the investigation of the social
process of news. The descriptive power of the concept comes from its openness
to the dynamism of social context, an openness which derives from it being in

a dialectic (i.e. realizational) relation to social context. No doubt much remains
to be done in elaborating the parameters of field, tenor and mode within the
SFL framework. But given the centrality of the notion of ‘functional variation’
to Halliday’s model, it would seem important to explore the full potential for
this variation to be captured through the parameters of social context, before
seeking explanations in supervening concepts, such as ‘genre’.
Notes
1
I am very grateful to Ruqaiya Hasan for fulsome feedback on an earlier draft of the
chapter. Carmel Cloran revised and corrected my Rhetorical Unit analysis. Wendy
Bowcher also gave me helpful feedback.
2
Note Nanri’s (1993) study argues Martin’s and Halliday’s conceptions of context
are compatible, and he draws on both for his argument. However, his argument
concerning the discourse structure of the news, in my view, is closer to Halliday’s
‘register’ than to Martin’s concept of ‘genre’.
3
Message is ‘a linguistic unit at the semantic stratum. Seen from above, it is the
smallest significant semiotic action that an interactant might take in the context of
an interaction so as to affect its character’ (Hasan, 1996b, p. 117).
4
As Hasan (personal communication) noted on an earlier version of this chapter,
Hartley’s term is not ideal for a text such as this one, in which all such ‘vox pops’
comes from US Marine Corp personnel of the Coalition invasion force – hardly
the ‘voice of the people’.
5
The images reproduced here comply with the copyright conditions under which
the data was purchased for this research project. See http://www.abc.net.au/
programsales/order.htm.
6
Since 2003 when this item was broadcast, the media landscape has changed
considerably, and options for news from a mode point of view have diversified.
7
See www.defenselink.mil/news/Feb2003/d20030228pag.pdf for the rules govern-
ing the embedding process.
8
See http://www.defenselink.mil/transcripts/transcript.aspx?transcriptid=2505.
9
Lukin (2008) reports the tendency across a corpus of 2 weeks of ABC TV news
reports (from which this instance is selected) to mitigate actual violence perpe-
trated by the Coalition, while presenting as highly potent the threat of violence by
Iraqi forces.
(Continued)
Appendix 1
(Continued)
Appendix 2
A B C D E G H I J K L M
E1 1 supply convoys attack Iraqi soldiers
clash Nasiriyah
2 column
ambush Iraqi forces
E2 3 coalition soldiers
4 convoy one
5 convoy border cross
6 [convoy] Basra head
7 keep fighting Iraq marines
8 convoy major
come
9 they border
across
10 sustainment convoy biggest
11 trucks > 70
12 [trucks] stretch
13 trucks disappear
E3 marines
14 young be
Gomez
15 we be
16 we wait
17 be
18 be
19 you
19 Iraq troops progress southern
20 be
21 we move fast
E4 22 convoy [move] tast

23 start quick
24
25 Iraqi southern
26 convoy fight
27
we
28 move
convoy
E5 29 border
coalition
30 invaders
point
31
32
33
E6 34 lies
35 BE
36 marines young BE
37 X
38 I seen
39
40
41 (I) seeing
42
43 you be
44 ME [be]
45 ME BB [be]
46 whoever do
they
47 do
they
48
49
50 war BE
51 BE
52 Gomez
53 war American soldiers become
E7 54 [supply] convoy marines move

Iraqi
55 resupply troops forward southern
US
N O P Q R S T U V W DD EE FF
KEY
1. square brackets denote retrieved ellipsis/lexicalisation of reference item
2. Underline denotes instantial relations (Hasan, 1985/89)
3. * denotes exophoric reference mediated by visual modality
4. Capitals denote interpretation of items not present
envy left behind

excited
anxiety
good
feel
good
roads
break
breach
left behind
death
bodies civilians
bodies civilian
feels disturbing bodies civilian Y
bodies civilian s-thing
[feel] gruesome Z AA
messed
sounds
up
feels motivating bodies civilian s-thing
contra-
sounds
dictory
trouble they* civilians
[trouble] [they] [civilians]
[trouble] [they] [civilians]
felt
mission accomplish
mission continue
captives
POWs
sombre
prisoners
roads
References
Allen, S. (2004). News Culture (2nd edn). Maidenhead: Open University Press.
Bateman, J. (2008). Multimodality and Genre: A Foundation for the Systematic Analysis of
Multimodal Documents. Basingstoke: Palgrave Macmillan.
Bernstein, B. (1990). Class, Codes and Control, Volume IV: The Structuring of Pedagogic
Discourse. London: Routledge.
Bernstein, B. (1996). Pedagogy, Symbolic Control & Identity: Theory, Research, Critique.
London: Taylor & Francis.
Butt, D. (2003). ‘Parameters of Context’. Mimeo. Dept of Linguistics, Macquarie
University.
Cloran, C. (1994). Rhetorical Units and Decontextualisation: An Enquiry Into Some
Halliday, M. A. K. (1985: Reprinted in full 1989). ‘Part A’, in M. A. K. Halliday and
R. Hasan (eds), Language, Context and Text: Aspects of Language in a Social Semiotic
Perspective. Geelong, VIC Oxford: Deakin University Press/Oxford University
Press, pp. 3–48.
Halliday, M. A. K. and Matthiessen, C. M. I. M. (2004). An Introduction to Functional
Grammar (3rd edn). London: Arnold.
Halliday, M. A. K., McIntosh, A., and Strevens, P. (1964: Reprinted in full 2007).
‘The users and uses of language’, in J. J. Webster (ed.), Language and Society.
Volume 10 in the Collected Works of M. A. K. Halliday. London and New York:
Hasan, Ruqaiya. (1973: Reprinted in full 2005). ‘Code, register and social dialect’,
in J. J. Webster (ed.), The Collected Works of Ruqaiya Hasan, Vol 1. London:
Equinox, pp. 160–193.
Hartley, J. (1982). Understanding News. London and New York: Metheun.
Hasan, R. (1964). ‘A Linguistic study of contrasting features in the style of two
contemporary English prose writers’. Unpublished PhD thesis, University of
Edinburgh, Edinburgh.
Hasan, R. (1984). ‘Coherence and cohesive harmony’, in J. Flood (ed.), Understand-
ing Reading Comprehension. Newark, DE: IRA, pp. 181–221.
Hasan, R. (1985: Reprinted in full 1989). Part B. In M. A. K. Halliday & R. Hasan
(Eds.), Language, context and text: Aspects of language in a social semiotic perspective.
Geelong, VIC./Oxford: Deakin University Press, Oxford University Press, pp.
52–118.
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. Norwood, NJ: Ablex, pp. 183–283.
Hasan, R. (1996a). ‘What’s going on: a dynamic view of context in language’, in
C. Cloran, D. Butt and G. Williams (eds), Ways of Saying: Ways of Meaning.
London: Cassell, pp. 37–50.
Hasan, R. (1996b). ‘Semantic networks: a tool for the analysis of meaning’, in
C. Cloran, D. Butt and G. Williams (eds), Ways of Saying: Ways of Meaning.
London: Cassell, pp. 104–131.
Hasan, R. (1999). ‘Speaking with reference to context’, in M. Ghadessy (ed.), Text

and Context in Functional Linguistics: Systemic Perspectives. Amsterdam/Philadelphia:
John Benjamins, pp. 219–328.
Hasan, R. (1999: Reprinted in full 2005). ‘Society, language and the mind: the meta-
dialogism of Basil Bernstein’s theory’, in J. Webster (ed.), Language, Society and
Consciousness. Collected Works of Ruqaiya Hasan, Volume 1. London and Oakville:
Equinox.
Hasan, R. (2004). ‘Analysing discursive variation’, in L. Young and C. Harrison
(eds), Systemic Functional Linguistics and Critical Discourse Analysis: Studies in Social
Change. London and New York: Continuum, pp. 15–52.
Hasan, R. (2009). ‘The place of context in a systemic functional model’, in
M. A. K. Halliday and J. Webster (eds), Continuum Companion to Systemic Func-
tional Linguistics. London and New York: Continuum, pp. 166–89.
Iedema, R., Feez, S., and White, P. R. R. (1994). Media Literacy. Sydney: Disadvan-
taged Schools Program, NSW Department of School Education.
Knightly, P. (2003). The First Casualty: The War Correspondent as Hero and Myth-Maker
from the Crimea to Iraq. London: Prion.
Lukin, A. (2008). ‘The explanatory power of the SFL dimensions in the study of
news discourse’, in C. Wu, M. Herke and C. M. I. M Matthiessen (eds),Voices
around the World. Proceedings from the 35th International Systemic Functional
Linguistics Congress, Volume 1, Macquarie University, pp. 143–171.
Martin, J. R. (1992). English Text. Philadelphia/Amsterdam: John Benjamins.
Martin, J. R., and White, P. R. R. (2005). The Language of Evaluation: Appraisal in
English. London and New York: Palgrave.
Nanri, K. (1993). ‘An attempt to synthesize two systemic contextual theories through
the investigation of the process of the evolution of the discourse semantic struc-
ture of the newspaper reporting article’. Unpublished PhD thesis, University of
Sydney.
Newspoll. (2003). ABC Appreciation Survey. June 2003.
Chapter 7
Constructing Sports Stars: Appliable

Linguistics and the Language of the Media
Shoshana Dreyfus and Sandra C. Jones
7.1 Introduction
High-profile sports people and their use of drugs and alcohol have featured
prominently in the news in the recent past. From AFL star Ben Cousins’ suspen-
sion for possession of illegal recreational drugs (March 2007), rugby league
player Andrew Johns’ admission to a long history of drug abuse (October 2007),
Nick D’Arcy’s charge for assaulting fellow sportsman Simon Cowley (March
2008), and cricketer Andrew Symonds’ admission of excessive drinking
(November 2008), the media has had no shortage of events to cover. Studies
(such as Biskrup and Pfister, 1999; Anderson and Cavallaro, 2002; White and
O’Brien, 1999; Stevens et al., 2003) have shown that sports stars are held up as
role models by some young people, particularly males. Other studies (such
as Paccagnella and Grove, 1997) have shown that sports stars’ transgressive
behaviour affects how young people perceive those sports stars, however, there
seem to be no studies to date that examine the way sports stars are portrayed in
the media the events of their transgressive behaviour. This chapter seeks to
begin to fill a gap in the literature on sports stars and their celebrity status in
the public eye, by focusing on the print media coverage of Ben Cousins during
the period of time when he was suspended from football due to drug taking
and erratic behaviour off the field. This chapter aims to begin to unpack the
way sports stars who transgress are packaged for the public, and in particular,
for adolescents. As such it uses an Appliable Linguistics Framework to address
a real-life problem, illuminating how the language choices that are made con-
strue particular versions of reality.
7.1.1 Background: a short biography of Ben Cousins

From 1996 to 2007, Ben Cousins played over 200 games for Perth’s
Australian Rules football (AFL) team, the West Coast Eagles, in the national
Constructing Sports Stars 115
football competition. He was appointed team captain in 2001and remained

captain until 2005. He was selected seven times for the Australian AFL team and
won numerous medals, including the Brownlow medal1, a Leigh Matthews
trophy2 and four ‘best and fairest’ awards. His off-field offences date from 2001
to 2007 when he was involved in a number of incidents involving recreational
drug use. These included a 2002 incident where Cousins was involved in a brawl
with a teammate, resulting in the breakage of Cousins’ arm; apparent associa-
tion with Perth ‘underworld’ figures in 2005; fleeing on foot after being stopped
by police for an alcohol breath test in February 2006 causing his resignation as
team captain; arrest and apprehension in jail for four hours after being found
in an uncontrolled state at Crown Casino in December 2006; missing a number
of training sessions, resulting in his suspension from the team in March 2007;
arrest for driving erratically and being found to be in possession of small quan-
tities of a number of drugs in October 2007, however no charges were laid;
admission to hospital in Los Angeles after an apparent cocaine binge at the end
of 2007. In November 2007 the AFL banned Cousins for 12 months.
When the story of Cousins’ drug use broke in late March 2007, there was a
flurry of media coverage around Australia, with 148 articles in an 11-day period
across a range of newspaper sections including news, sport and features. When
Cousins began negotiating his return to the AFL in late June and early July
2007, there were another 48 articles. By any measure of coverage, this is a large
number of articles. Lines (2001) attributes this kind of extensive media cover-
age to the treatment of sports stars as ‘celebrities’ – in news and entertainment,
as well as sports media – giving them highly visible and enduring images. Impor-
tantly, Lines (2001) also notes that while sports stars are real people, how they
are represented is a construction, a particular version of events:
Sports stars are real in the sense that they perform live under unpredictable
sporting conditions over which apparently the media has little control. Yet,
the nature of what the reader gets to see, hear and read about is determined
and amplified by camera angles, replays, gossip columns, photographic
images, chat shows and other such professional practices which ensure that
the sport star image develops through selected constructions of reality. (Lines,
2001, p. 287)
In the age of commodification of celebrities, it is not surprising that writers

have argued that celebrity status is not necessarily indicative of greater talent
than anyone else in the field but ‘that one has been more successfully pack-
aged, promoted, and thrust upon the hungry masses’ (Ferris, 2007, p. 374).
Thus a study of the portrayal of a transgressing sports star might shed light on
how they (and their identities) are packaged for young people, in order to
make a link, as it were, between studies that acknowledge the impact of sports
stars on young peoples’ behaviour and those that look at how those sports
stars are perceived by young people. Specifically, we were keen to see whether
Cousins was constructed as being responsible for his transgressive actions or
not. In terms of a possible hypothesis, it was thought that if the media deemed
Cousins responsible for the actions taking place, it would use language that
constructs him as being responsible.
7.2 Analytical Framework
The analytical framework used for this study comes from Systemic Functional
Linguistic theory (after Halliday, 1994), a comprehensive and powerful Appliable
Linguistics theory that can be used to address practical and theoretical prob-
lems that engage with language (Halliday, 2006). Within this model there are
two systems which allow for an examination of responsibility: Transitivity and
Ergativity. The details of these systems and how they can be applied are
explained below. A third system, Appraisal (Martin and White, 2005), is also
used to briefly examine how Cousins was portrayed.
7.2.1 Transitivity
The system of Transitivity (after Halliday, 1994) assigns roles to the partici-
pants in texts. In Transitivity, the roles are different depending on whether
the participants are active, that is, the doer in any given process, or whether
they are the recipient of that process. Whether a person is constructed as being
a doer shows whether they have an effect on things. When the process is a mate-
rial type of action, the doer is labelled Actor, and the done-to is labelled Goal.
For example, in the following clause, Cousins is the Actor (doer) and the ball is
the Goal (done-to):
Cousins kicked a goal
Actor Process: material Goal
This kind of construction shows Cousins as active in the event of kicking the
goal and can be contrasted with the following:
The club has failed Cousins
In this clause, Cousins is constructed as the receiver of an action done by

others. In being the receiver, he is constructed as not having responsibility or
part in the actions, other than receiving them, as the actions are done to him by
others. Transitivity therefore affords a view of whether someone is an active
participant in the actions in any given event.
7.2.2 Ergativity
The system of Ergativity complements Transitivity as it is about agency,
assigning roles to participants that are different from Transitivity roles. The
roles are Agent – which is the role that has a causative effect on others, or
Medium – the role that does not have an effect on others but is crucial to
the process taking place (see Halliday, 1994 for a fuller description). When a
clause is constructed with the feature of agency, someone or something is seen
to not just do an action, but cause an action to be done, and therefore has
responsibility for or power over the action. This can be seen in the following
example:
Cousins evaded the three-strikes drugs policy
Ergativity Agent Process Medium
Transitivity Actor Process: material Goal
In this clause, Cousins has evaded the policy, so he is the Agent. The policy is
the Medium as it is the thing being evaded. Therefore, the above clause con-
structs Cousins as being responsible for the evasion of the ‘three strikes’ policy.
This can be contrasted with:
The ball rolled
Medium Process
Actor Process: material
In this clause there is no feature of agency as there is only a Medium and no

Agent. There is no cause for the process of the ball rolling. It is constructed as
simply happening.
The above two examples depict two different ways of representing events in
two different types of clause: the first, with agency, is called an effective: active
clause, because something has an effect on something else. The second ‘the
ball rolled’ is called a middle clause, because it has no feature of agency. Con-
structing events as middle clauses is one way of removing or backgrounding
agency from events. However, there is a second way, which is via the effective:
passive clause, which backgrounds the Agent by either putting the Medium first
and the Agent second or omitting the Agent altogether, as in the clause below:
Cousins was nominated as a player deserving of specific scrutiny by drug

testers
Medium Process Circumstance: role

While a clause constructed in the passive is often constructed that way for pur-
poses of organizing the information in a particular way (Martin, 1993), what is
selected as Agent or Medium, and the presence and order of those selections in
the clause, sheds light on who is construed as being responsible for the actions.
7.2.3 Appraisal theory

Appraisal theory (Martin and White, 2005) was developed from within the
Systemic Functional model of language and is a framework for analysing how
speakers and writers evaluate phenomena. This framework has three systems:
Attitude, Engagement and Graduation, and it is Attitude that this chap-
ter is concerned with. Within Attitude there are three subsystems: Affect
(relating to emotions), judgement (relating to evaluation of people and their
behaviour) and Appreciation (relating to evaluation of artefacts). Within each
of these subsystems, evaluation can be either positive or negative, and it is this
positive/negative aspect of Appraisal that was used for the examination of the
portrayal of Cousins and his behaviour. As we were concerned with how Cousins
and his behaviour were evaluated, the system of judgement is particularly
relevant here.
7.3 Methodology
Two hundred and thrity-three articles from across Australia between March and
July 2007 that related to the Ben Cousins story were collected. This large corpus
was narrowed down to the coverage of two broadsheets: Melbourne’s The Age
(Melbourne being the ‘home of AFL’) and Perth’s The West Australian (Perth
being the ‘home of the Eagles’ – Ben Cousins’ team). The following analysis
reports on findings from the first eight articles from the coverage of The Age. As
some articles only mentioned Cousins briefly, using him as a springboard to
discuss larger issues, while others talked in more detail about Cousins and the
events surrounding his suspension, it was decided that instead of conducting
a detailed analysis of whole articles that were not about Cousins, every time
there was a mention of Cousins, the clause/s in which he was mentioned would
be examined for Transitivity and Ergativity structure, and Appraisal. If
Cousins was the main focus of the article, and therefore mentioned many times,
then these instances were viewed in terms of how the meanings built up cumu-
latively across the text.
7.4 Findings from Transitivity and Ergativity Analysis
The findings from the Transitivity and Ergativity analysis show that Cous-
ins was constructed both as Actor (doer) and as Goal (done-to) in the articles.
He was also sometimes constructed as Agent, but more frequently as Medium.

Examples showing these construals follow (examples are first written, then pre-
sented in tables with the Transitivity and Ergativity analysis beneath).
Example 1 shows Cousins in the active role – Actor, in 3 out of 5 clauses:
Example 1:
Fallen West Coast champion Ben Cousins remains in denial over his drug
addiction and has refused several attempts by his family and his club to con-
vince him to check into rehabilitation. The 28-year-old, who has been sus-
pended indefinitely by the Eagles, has continued to behave erratically and
has also resisted attempts by the club to make a public statement about the
severity of his health crisis. (The Age, 27.3.2007)
1.1.1 Fallen West Coast remains in denial over his drug addiction
champion Ben
Cousins
Trans Carrier Process: Attribute: Circumstance: matter

relational Circumstance
Erg Medium Process Range Circumstance
1.1.2 and (he) has refused several attempts by his family and his club
[[to convince him to check into
rehabilitation.]]
Agent Process Medium
1.2.1 The 28-year-old, . . . has continued to behave erratically
Behaver Process: behavioural Circumstance: manner
Medium Process Circumstance
1.2.2 . . . who has been suspended indefinitely by the Eagles . . .
Goal Process: material Circumstance: extent Actor
Medium Process Circumstance Agent
1.2.3 and (he) has also resisted attempts by the club [[to make a public
statement about the severity of his health
crisis.]]

In this example, where Cousins is mentioned five times in five clauses, three out
of the five (1.1.2, 1.2.1 and 1.2.3) construct Cousins as the active Participant in
the clause (although Agent in only two):
z refusing several attempts to get him into rehabilitation in 1.1.2;

z continuing to behave erratically in 1.2.1; and
z resisting attempts to make a public statement in 1.2.3.
In contrast to this, Example 2 is one of the instances where Cousins is not

construed as Actor and Agent:
Example 2:
Nathan Buckley feels for Ben Cousins, with his private life laid bare to public
scrutiny, but believes West Coast has done the right thing in suspending
him . . . ‘If you want to be a leader in society . . . You’re going to have things
come up and bite you on the arse’, the Collingwood captain said yesterday.
(The Age, 22.3.2007)
2.1.1 ‘Nathan Buckley feels for Ben Cousins,
Trans Senser Process: mental Circumstance: cause: behalf
Erg Medium Process Range
2.1.2 with his private life laid bare to public scrutiny
Goal Process: material Beneficiary
Medium Process Range
2.1.3 but (he) believes
Senser Process: mental
Medium Process
2.1.4 West Coast has done the right thing
2.1.5 in suspending Him . . .
Process: material Goal
Process Medium
2.2.1 ‘If you want to be a leader in society . . .
Carrier Process: relational Attribute
2.2.1 You’ re going to have things [[come up // and bite

(cont) you on the arse’,]]
Carrier Process: relational: possessive Attribute: possessed
2.2.2 the Collingwood captain said yesterday.’
Sayer Process: verbal Circumstance: time
In this example from the very early coverage of the events in The Age, Cousins is
not constructed actively, but has things done to him: he has his private life laid
bare, he is suspended, and things ‘come up and bite [him] . . . on the arse’. He
is neither Actor nor Agent in any of the clauses, as can be seen in the analysis
above, where Cousins features:
z in the Circumstance in clause 2.1.1, as the person on whose behalf we feel;

z as the Goal/Medium in clause 2.1.2, with his private life laid bare; and
z as Goal/Medium (being suspended) in clause 2.1.5.
Further, in the embedded clause ‘things come up and bite you on the arse’,
which is within clause 2.2.1, Cousins also features as Goal/Medium, a recipient
having things done to him, where agency is given to ‘things’:
things come up and bite you on the arse
Actor Process: material Goal Circumstance: place
Agent Process Medium Circumstance
Example 3 shows an instance of more explicit discussion of the issue of

responsibility, and comes from one of the two articles that completely deprived
Cousins of agency, instead, attributing responsibility for the footballer’s situa-
tion to the football club:
Example 3:
After Thursday’s confirmation, Eagles chief executive Trevor Nisbett said the
club had failed Cousins by not providing better help . . . AFL chief executive
Andrew Demetriou said the league would learn from this: ‘What we want to
do is to wrap our arms around Ben Cousins, get him rehabilitated, get him
healthy.’
(The Age, 24.3.2007)
3.1.1 After Thursday’s confirmation, Eagles chief executive said,

Trevor Nisbett
Trans Circumstance: time Sayer Process: verbal
Erg Circumstance Medium Process
3.1.2 the club had failed Cousins
3.1.3 by not providing better help . . .
Process: material Goal
Process Medium
3.2.1 AFL chief executive Andrew Demetriou said
Sayer Process: verbal
Medium Process
3.2.2 the league would learn from this
Senser/Actor Process: mental/behavioural Circumstance: location
3.2.3 [[‘What we want to do]] is [[to wrap our arms around Ben
Cousins, // get him rehabilitated,
// get him healthy.’]]
Value Process: Token

relational
Range Process Medium
In this example, it is the club, specifically in clauses 3.1.2 and 3.1.3, who is
Agent for failing Cousins by not providing help. Nowhere is Cousins given
agency over the events.
However, while Cousins is construed as lacking in agency over the events in

six out of the eight articles, there are construals of Cousins as becoming more
active and somewhat agentive in the events. This can be seen in Example 4,
which begins with agency being attributed to the club but shifts to Cousins
having some agency:
Example 4:
Although the Eagles hierarchy has made it clear to Cousins that he will not be
allowed back to the club until he fully accepts his addiction to the metham-
phetamine ‘ice’, and seeks rehabilitation, the 2005 Brownlow medallist as
recently as the weekend remained determined he could beat his drug
problem without full-time medical and psychiatric help . . . West Coast chair-
man Dalton Gooding confirmed yesterday that Cousins had resisted a series
of attempted interventions but added that some progress was being made.
‘The first step for Ben is accepting the problem,’ said Gooding, ‘and then
Ben making a statement and then going off somewhere for rehabilitation.’
(The Age, 27.3.2007)
4.1.1 Although the Eagles has made it clear to Cousins

hierarchy
Trans Initiator Process: Carrier Attribute Beneficiary

causative
Erg Agent Process Medium Range Range
4.1.2 that he will not be allowed back to the club
Goal Process: material Circumstance: location
4.1.3 until he fully accepts his addiction to the methamphet-

amine ‘ice’,
4.1.4 and (he) seeks rehabilitation,

4.1.5 the 2005 Brownlow medallist as recently as the weekend remained determined
Senser Circumstance: time/extent Process: mental
Medium Circumstance Process
4.1.6 he could beat his drug problem without full-time medical and
psychiatric help . . .
Actor Process: Goal Circumstance: accompaniment

material
Agent Process Medium Circumstance
4.2.1 West Coast chairman Dalton Gooding confirmed yesterday
Sayer Process: verbal Circumstance: time
4.2.2 that Cousins had resisted a series of attempted interventions
Actor Process: material Range
Medium Process Process
4.2.3 but added
Process: verbal
Process
4.2.4 that some progress was being made.
Goal Process: material
Medium Process
4.3.1 ‘The first step for Ben is [[accepting the problem,’]]
Value Process: relational Token
Range Process Medium
4.3.2 and then Ben making a statement
Actor Process: material Range
Agent Process Range

4.3.3 and then (Ben) going off somewhere for rehabilitation.’
Actor Process: Circumstance: Circumstance: purpose

material location
Medium Process Circumstance Circumstance
Example 4 shows the club as Agent first, not allowing Cousins back into the club
unless he performs particular actions. Therefore, the kind of doer the club
wants of Cousins is someone who first accepts his addiction or problem, then
makes a statement about it, and then goes off somewhere for rehabilitation.
From this example, it seems clear that the football fraternity do want Cousins to
be the Agent and Actor/doer, and take control of his life, however, they only
want a particular kind of Agent/Actor/doer, one who accepts his addiction,
seeks rehabilitation and makes a statement.
Finally, on 30 March, Cousins is depicted as being the kind of doer the League
wants:
Example 5:
Fallen Eagle Ben Cousins will head overseas today in a bid to overcome his
drug addiction. (The Age, 30.3.07)
5.1 ‘Fallen will head overseas today in a bid to

Eagle Ben overcome his
Cousins drug addiction.’
Trans Actor Process: Circumstance: Circumstance: Circumstance:

material place time cause
Erg Medium Process Circumstance Circumstance Circumstance
7.4.1 Findings from Appraisal analysis

The findings from the Appraisal analysis show two particularly salient points.
First, as mentioned above, it was noticed that there were differences between
how Cousins was represented on and off the football field. On the field, the
portrayal of Cousins was positive, lauding his skills and prowess as a footballer,
whereas off the field the portrayal was often negative, depicting him as an out-
of-control drug taker. In terms of the finer classification in the Appraisal sub-
system of judgement, the positive judgement relates to Cousins’ capacity (as
footballer), whereas the negative judgement relates to issues of ethics and pro-
priety. This can be seen in Table 7.1, which plots some of the attributes ascribed
to Cousins in The Age accompanied by the Appraisal values.
The second salient point that was observed with regard to Appraisal is the
use of the lexical metaphor ‘fallen’ to describe Cousins (as in examples 1 and
5 above). Within the Appraisal system one of the functions of lexical metaphors
is to provoke an attitudinal response in the reader (Martin and White, 2005).
Table 7.1 Appraisal values of Cousins’ behaviour

On-field Appraisal value
AFL champion positive judgement (capacity)
reigning Brownlow medalist positive judgement (capacity)
star midfielder positive judgement (capacity)
Off-field Appraisal value
. . . remains in denial over his drug negative judgement (propriety)

addiction
. . . has refused several attempts by his family negative judgement (propriety)

and his club to convince him to check into
rehabilitation
. . . has continued to behave erratically negative judgement (propriety)
Considering that this term was used to describe Cousins at least four times,
including fallen Eagle, fallen star and fallen West Coast champion, it seems worth
exploring.
Typically, the ‘fallen’ metaphor is used to mean a fall in status or prestige, and
it refers to the ‘original sin’ in the Judeo-Christian bible: the fall from God’s
grace by Adam and Eve in the Garden of Eden as a result of eating the forbid-
den fruit. Just as Eve was led into the temptation of eating the fruit by the
serpent, so Cousins was led into the temptation of drug taking (‘things come
up and bite you on the arse’ – perhaps referring to the serpent that accosted
Eve and coerced her into eating the fruit), the result for both being a loss of
innocence and a fall from grace. This is a theological frame of reference, a
biblical way of positioning Cousins in relation to his misdemeanour. The use of
the metaphor therefore refers to the fact that as a result of his actions, Cousins
has fallen from his highly regarded position into the lowly position of disgrace
as a result of succumbing to the temptation of drugs. However, whether this
provokes pity or blame (or both) in the reader is unclear.
7.5 Discussion
The results of the study of eight articles from The Age newspaper show that
even when there are instances where Cousins is construed as active and
agentive in the events, he is also regularly construed as not active and agentive,
in conjunction with not being negatively evaluated. This begs the questions of
why – when it is seemed to be fairly clear that Cousins had repeatedly behaved
inappropriately by breaking the law, using illegal drugs, as well as not turning
up to training and behaving erratically – did the representation of Cousins in

these newspaper articles frequently construct him as having little active involve-
ment and agency in the events? In other words, why was he frequently con-
structed as the Goal and Medium, not the Actor and Agent? And why did the
club and the football league take responsibility for many of the events (as Actor
and Agent) when it was clear that Cousins had transgressed (and should have
therefore been depicted as Actor and Agent of his own downfall)? The answer
to this can possibly be found in a number of factors:
First, if, as Whannel (1992) suggests, sports heroes are a source of national
identity and pride, then perhaps, the higher the profile of the sports hero, the
more their fall is cushioned. This study suggests that Cousins’ fall was cush-
ioned by the print media’s representation of him through the repeated con-
struction of him as non-active and non-agentive in the events, and clearly, as
can be seen from how he is positively appraised as a footballer in the articles, he
was seen to be a very highly valued sports star.
Second, unlike violent offences committed by other high-profile sports people
such as Nick D’Arcy (in early 2008), Cousins’ offences were about his personal
drug use and related behaviour. That is to say, other than let his team (his
family, his fans and his league) down, Cousins did not do anything that hurt
anyone else. The conjecture here is two-fold: the representation of Cousins as
not only negative can be partly attributed to the fact that Cousins is of extremely
high value as a sports person, and that his offences did not transgress into the
realm of violence. While it was not the brief of this project, it would be interest-
ing to compare the media representation of D’Arcy, who assaulted a fellow
sportsperson Simon Cowley in a bar, breaking his nose, with Cousins, to see
whether the media construction is as varied with D’Arcy as it seems to be with
Cousins.
This varied representation of Cousins in The Age articles fits into Lines’ (2001)
assessment of the complex role that male sports heroes occupy in media
narratives, simultaneously constructed as villains, heroes and fools, forming a
complex mix which he calls ‘damaged hero’ (2001, p. 285). There is evidence,
particularly from the Appraisal analysis, which shows Cousins depicted as
damaged hero: sports hero on the field and disgraced role model off the field.
There is a tension here, between the construction of Cousins’ positive capacity
on the field and his negative value in terms of ethics – referring to what is
socially sanctioned as the correct behaviour a high-profile footballer should
exhibit or engage in off the field.
To return to the original concerns motivating this study, we must ask what the
effect of this varied construction of transgressive sports stars like Cousins is on
young people, particularly those who hold sports stars in high regard. As Lines
(2001) points out, more research is needed to explore this.
7.6 Conclusion
This chapter sought to explore whether the representation of Ben Cousins in

the print media around the time he was banned for a variety of drug-related
offences used linguistic resources that constructed him as responsible for his
actions, with the view to shedding light on how transgressing sports stars
are packaged for the public. Using the Appliable Linguistics Framework of
Systemic Functional Linguistics, specifically the systems of Transitivity and
Ergativity, the chapter showed that in the eight articles from The Age, Cousins
was constructed in a variety of ways, sometimes as active and sometimes not,
and only occasionally as agentive. In other words, he was frequently constructed
in ways that deprived him of responsibility and agency in the events that culmi-
nated with his suspension from the football league. It was argued that the
construction of Cousins as a recipient of actions done by others rendered him
less responsible for the situation and therefore less blameworthy. It was conjec-
tured that Cousins’ extremely high value as a sportsman and the fact that his
misdemeanours did not include violence towards others afforded a generous
treatment of him by some sections of the media. As this chapter sought to begin
to unpack how transgressive sports stars are packaged for the public in the
events of their transgression, it is interesting to note that the varied depiction
of Cousins confirmed Lines’ (2001) analysis of the representation of the con-
temporary sports star as ‘damaged hero’, and we might, therefore, ask what
impact this kind of portrayal might have on young people.
Notes
* Acknowledgement: the work for this chapter was made possible by a $3000 faculty
grant from the Faculty of Health and Behavioural Sciences at University of
Wollongong.
1
The Brownlow medal is the leading individual award in AFL, awarded once a year
to the fairest and best player of the football season.
2
The Leigh Matthews trophy is awarded by the AFL players Association to the most
valuable player in the football league.
References
Anderson, K. J. and Cavallaro, D. (2002). ‘Parents or pop culture? Children’s heroes
and role models’. Childhood Education 78 (3), 161–8.
Biskrup, C. and Pfister, G. (1999). ‘I would like to be like her/him: are athletes role
models for boys and girls?’ European Physical Education Review 5 (3), 199–218.
Ferris, K. O. (2007). ‘The sociology of celebrity’. Sociology Compass 1 (1), 371–84.
Halliday, M. A. K. (2006). ‘Working with meaning: Working towards and appliable

linguistics’. Inaugural lecture to mark the Halliday Centre for Intelligent Appli-
cations of Language Studies at the City University of Hong Kong.
Halliday, M. A. K. (1994). An Introduction to Functional Grammar (2nd edn). London:
Edward Arnold.
Lines, G. (2001). ‘Villains, fools or heroes? Sports stars as role models for young
people’. Leisure Studies 20, 285–303.
Martin, J. R. (1993). ‘Technicality and abstraction: language for the creation of
specialized texts’, in M. A. K. Halliday and J. R. Martin (eds), Writing Science:
Literacy and Discursive Power. London: Falmer Press, 203–220.
Paccagnella, M. and Grove, J. R. (1997). ‘Drugs, sex and crime in sport: an
Australian perspective’. Journal of Sport and Social Issues 21, 179–88.
Stevens, J. A., Lathrop, A. H. and Bradish, C. L. (2003). ‘Who is your hero? Implica-
tions for athletic endorsement strategies’. Sport Marketing Quarterly 12, 103–10.
Whannel, G. (1992). Fields in Vision: Television Sport and Cultural Transformation.
London: Routledge.
White, S. H. and O’Brien, J. E. (1999). ‘What is a hero? An exploratory study of
students’ conceptions of heroes’. Journal of Moral Education 28 (1), 81–95.
Chapter 8
Naming Culture in Convivial

Conversational Humour
Naomi K. Knight
8.1 Introduction
This chapter attempts to explain how friends create humour in conversation by

identifying and bonding around shared and unshared values that may be inter-
preted in the linguistic text. As conversation is a genre motivated by a prosody
of appraisal (Martin, 2000b) (the sharing of attitudes), it is proposed that inter-
actants bond in humour by playing with values that represent their relational
identities. While these values are negotiated in the function of solidarity, they
also construct ideological distinctions that create tensions in participants’ soci-
ality. Through humour, friends laugh off tensions together in order to bond
around what they share in a process of affiliation. This dynamic negotiation
will be presented through a systemic functional discourse analysis (Martin and
Rose, 2007) of five separate texts of humorous phases of conversation between
Canadian friends.
Through the framework of an appliable linguistics, this chapter takes the
everyday context of casual conversation between friends and focuses on the
problem of humour, developing the theoretical model of Systemic Functional
linguistics (SFL) to respond to this problem. In particular, the model of affilia-
tion is presented to account for the social relations at stake in this type of
humour, particularly in terms of how we negotiate who we are as members of
communities of the culture. SFL is especially appliable in this context, since it
offers a social semiotic perspective on language (Halliday, 1978), and ‘affilia-
tion’ is developed as a social semiotic model of communities. Furthermore, the
SFL resource of naming offers insight into the construal of identities in conver-
sational humour, and will be employed in the analysis of texts in this chapter.
8.2 Conversational Humour

Casual conversation is a rich source for engaging with the construal of socio-
cultural identity, particularly in the humour used by friends to manage their
Naming Culture in Convivial Conversational Humour 131
social complexity through laughter. Researchers who have studied conversa-

tional humour (e.g. Norrick, 1993; 1994; Hay, 2000; 2001; Eggins and Slade
1997; Priego-Valverde, 2003; Tannen, 2005) exhibit its role in construing both
solidarity (or rapport) between interlocutors and power (or aggression) func-
tions at the same time. In fact, Priego-Valverde (2003) describes a dialogic rela-
tion between the functions of cooperation and competition, so that it may be
argued that conversational humour is an important strategy for the manage-
ment of social relations between interactants.
However, particularly when conversational humour is shared by friends, a dis-
tinct orientation towards solidarity becomes apparent, and interactants use
humour to reinforce existing relations of familiarity and informality (Kotthoff,
1996). According to Norrick (1994, p. 423), solidarity may be perceived as the
main function of the ‘meta-message’ of conversational humour between friends,
and even those utterances perceived as locally aggressive achieve solidarity
globally due to a ‘customary joking relationship’ (Norrick, 1993). Hay (2000,
p. 721) also found that negative categories of the use of humour in the function
of power, such as control and creating conflict, seldom occurred in her study
on humour in friendship groups (see also Holmes and Marra, 2002).
Furthermore, friends bond together through humour by identifying with
values that they share as members of communities. According to Boxer and
Cortés-Conde (1997, p. 282), humour is used by conversational participants in
a ‘negotiation of a relational identity with others and through others’, leading to
a sense of group membership and an effect of bonding, and especially amongst
friends, ‘maximum RID [relational identity display] can be accomplished
through joking and teasing that bonds’ (Boxer and Cortés-Conde, 1997, p. 293).
Archakis and Tsakona (2005) also argue that humour ‘reveals information
regarding the humorists’ shared beliefs and values’ (p. 62) by which they may
construct their social identity, as they bring out shared ideals of ‘appropriate’
behaviour. Even when humour is targeted within the in-group, the bonds
between group members are confirmed and strengthened by the ‘safety’ their
relationship offers. These functions are reiterated in studies of laughter as well,
as Glenn (2003) argues that laughter in conversation is used to construe affilia-
tion and to negotiate interpersonal relationships and identity, and it ‘plays an
essential role in building and expressing affiliation, alignment, identity and
relationships’, according to Partington (2006, p. 229). Thus, affiliation and
bonding, as well as the construction of identity and solidarity between conver-
sational participants may be considered as intrinsically related and significant
functions of conversational humour between friends, and these aspects will be
explored in this chapter.
This chapter aims to demonstrate, through the analysis of conversational
texts between Canadian friends, how conversational humour between friends is
a strategy for the construal of affiliation (Knight, 2008). By negotiating values
that are more or less shared between them, speakers not only share a laugh but
also reaffirm their roles in society as complex members of communities that are
more or less negotiable. In this way, conversational humour between friends is
a tool for managing the tensions that occur in an ongoing negotiation of com-
munal, relational identities, and for reinforcing the bonds that friends share.
The chapter will also demonstrate the role of ideology and culture as a constant
feature of this negotiation, as participants constantly display their positioning
as members of a culture that is organized by ideology or ‘coding orientations’
(Bernstein, 2000). I will further suggest that humour allows speakers to make
relevant and negotiable various levels of affiliative identities by presenting them
through couplings (Martin, 2000a) in the text, and this is informed by realiza-
tions of the involvement resource of ‘naming’ (cf. Martin and White, 2005). To
contextualize the exploration of conversational humour in this chapter, the fol-
lowing section describes the analytical approach to the data, while Section 8.4
describes the social process of affiliation. Section 8.5 then describes the resource
of naming and discusses how it informs the interpretation of humour and affili-
ation. The use of conversational humour and the role of naming are exhibited
in the section 8.6 through discussions of the analysis of the humorous texts.
8.3 Approach
The data for this study consist of five texts of humorous phases of casual conver-
sation (totalling 12 hours of recorded data) between Canadian friends. These
friends ranged from ages 20 to 30 and most participants were undergraduate
university students, or were in post-university jobs.1 Casual conversation is speci-
fied according to Eggins and Slade’s (1997, p. 19) definition as ‘talk which is
NOT motivated by any clear pragmatic purpose’. The talk was recorded in dif-
ferent contexts with different sets of participants, both male and female, and
the participants were unaware that humour was of particular interest to the
study (since the topic had not yet been decided). From a total of 12 hours of
recorded conversation, I chose five phases of humour for this analysis.
The humorous phases were identified by the occurrence of laughter. Laugh-
ter served as a signal that the participants were negotiating a ‘laughing’ kind of
affiliation, and together dealing with a tension in the ongoing conversation.
The surrounding meanings were then analysed, while laughter along with the
naming of identity categories informed the analysis and the identification of
communities that were construed.
Most of the recordings included myself as researcher-participant, in both
friendship and additional roles with most of the interactants. While Labov
(1972) finds that this presents an ‘observer’s paradox’, risking the naturalness
of the discourse, Eggins (1990, p. 136) has argued that the risks are offset by
the advantage of having an insider take part in the interaction and thereby
contribute to the naturalness and comfort of the participants, since the
researcher shares a friendly relationship with them. Furthermore, my participa-
tion in the instigation of the humour proved critical for interpreting the often
unfunny texts.
The study is undertaken in an SFL perspective (cf. Halliday, 1978; also see
surrounding chapters, this volume, for elaboration). In particular, a discourse
analysis following Martin and Rose (2007) was completed, focusing on the
systems of ‘appraisal’ and ‘ideation’. The discourse semantic system of
appraisal2 is located in the interpersonal metafunction, which includes those
sets of meanings by which we enact relationships. Appraisal includes choices of
attitudes, how we intensify them and their sources. Attitudinal meanings are
divided into three subtypes: affect (emotions), judgement (evaluations of peo-
ple’s character and behaviour) and appreciation (evaluations of things both
concrete and abstract). Focusing on attitudes, I coded both inscribed (or
explicit) and invoked (or implicit) attitudinal meanings that were expressed in
the texts. Then, I focused on the targets of these evaluations through an analy-
sis of the system of ideation. Ideation concerns ‘what kinds of activities are
undertaken, and how participants undertaking these activities are described
and classified’ (Martin and Rose, 2007, p. 25). The system of ideation focuses
on the construal of experience in discourse, including the ‘people and things
involved in them, and their associated places and qualities, and on how these
elements are built up and related to each other as a text unfolds’ (Martin and
Rose 2007, p. 109). The ideation analysis involved classifying the ideational
meanings into taxonomies and identifying their role in the clause as well as
their relation to surrounding clauses. The relationship between attitude and
experience could then be more clearly interpreted, particularly in terms of
identifying ‘couplings’, which I will introduce in the next section. Also, in the
discourse analysis, the choices that speakers made were considered in relation
to the meanings that they construed in the social interaction, informing the
affiliation model that I will introduce in this chapter.
8.4 Background to Affiliation
While there are many types of humour that have been studied across disciplines
for centuries (cf. Attardo, 1994), the linguistic study of conversational humour
is still relatively new, and it lacks a consistent and unified definition (Coates,
2007). Therefore, the type of humour under focus in this chapter is given
its own definition, informed by the social functions that it achieves. ‘Convivial
conversational humour’ (Knight, 2010) is characteristic of conversation between
friends in which laughter proliferates, and is oriented towards solidarity and
connecting around shared social networks of bonds. This is the type of humour
that conversational outsiders would consider ‘unfunny’ and difficult to inter-
pret, but it is a highly informative feature of talk that is closely related to how we
construe who we are and how we relate as members of a culture on an ongoing
basis.
Furthermore, there has not yet been a full study of humour in the SFL frame-
work. Eggins and Slade (1997) offered a promising start to this topic of research
by connecting conversational humour between friends with the sharing of

attitudes in text (captured in the SFL system of ‘appraisal’ (Martin and
White, 2005)), and they argue that humour enables interactants to ‘negotiate
attitudes and alignments, and . . . degrees of “otherness” and “in-ness’’ ’ (Eggins
and Slade, 1997, p. 155). Knight (in progress) attempts to advance these insights
in a complete study of conversational humour by developing a model of ‘affili-
ation’. This model accounts for the sharing of attitudes in text and the partici-
pants’ construal of the social functions of bonding and co-identification. When
speakers communicate attitudes in convivial conversational humour, they tie
them together with particular ideational experiences, variously valuing differ-
ent persons, things and experiences in their world. These ‘couplings’ of atti-
tude with experience (as introduced by Martin, 2000a; see also Martin, 2008a,
2008b; Zappavigna et al., 2008; Zhao, 2010) bring the participants together,
who construe affiliation by variously communing around, rejecting or laughing
off these couplings as more or less ‘acceptable’ bonds between them. Couplings
are exhibited in the following excerpt:
C: == This was an awesome pie party guys

N: I love pie parties
Example 1: attitude + ideation couplings
[Transcription conventions adapted from Eggins and Slade (1997);
Appraisal Coding: negative attitude underlined, positive attitude double
underlined, target of evaluation in bold.]
In Example 1, the two speakers present couplings of attitude with the ideational
range ‘pie party/ies’. The inscribed positive appreciation (‘awesome’) and pos-
itive affect (‘love’)3 binds together with this ritual that they often participate in
as a group of friends, and creates a coupling that they can share to affiliate
together as members of a friendship community. By sharing this coupling in
the text, these interactants construe a social bond between them as members of
a ‘pie party’ community in the culture.
‘Bonds’ in this sense may be compared to the concept of ‘membership cate-
gorization devices’ (MCDs) as introduced by Sacks (cf. 1992), and reformu-
lated in Schegloff (2007) and in Housley and Fitzgerald’s (2002) terms. In this
model, speakers can be distinguished by a particular social categorization based
on relevant aspects of the linguistic text that are ‘consequential to the interac-
tional business’ (Schegloff, 2007, p. 474), such as ‘category-bound’ activities,
attributes and predicates, which signal particular social categories and collec-
tions of categories (i.e. MCDs). The feature for naming bond categorizations,
however, is specified here as couplings in the theory of affiliation, which may
involve ideational meanings in any category (i.e. processes, participants, quali-
ties, etc.) that are tied to particular attitudes.
The coupling is in this way a linguistic unit construing a bond as a social semi-
otic unit of a different order in the social dimension of affiliation, and it is by
negotiating bonds that conversational participants co-construct who they are.

Conversational participants construe affiliation through the presentation of
couplings in text, sharing them as bonds that connect them as members of
communities, or social networks, of more or less ‘like-minded’ persons (cf.
Stenglin, 2004). Social networks are perceived in this model of affiliation as
clusters of bonds4 negotiated ongoingly through couplings (e.g. ‘awesome + pie
party’) in the text, so that when participants present couplings they are constru-
ing bonds that connect them into a clustering of a particular community (e.g.
‘Fun Pie Party’). As social persons, however, our affiliations can be highly com-
plex. Those social networks that we affiliate with in different contexts
are not always harmonious and can at times come into conflict. Humour is one
way to manage this complexity, as it provides a mechanism for ‘laughing off’ the
tensions.
In convivial conversational humour, a speaker presents a coupling that con-
strues a bond of some unshared (but acceptable) social network, creating a
tension with what the participants at talk have been aligning around together.
Because this tension is humorous, interactants can laugh it off in order to affili-
ate together based on what they can share, and at the same time acknowledge
the intruding bond to be an acceptable, though unshared, one. This is exhib-
ited in the following text, which occurs after that presented in Example 1. One
interactant presents a coupling around which the others cannot align, creating
tension with the pie party bond they have just shared:
C: I can’t wait to have another one

F: “Next year in (LV) Jerusalem”
F,N: (L) ==
C: == What?
F: I mean next year more (LV) pie!
N, F: (L) …
F: It’s-it’s a Jewish thing y-ye-at the end of every Passover you say next year
in Jerusalem in Hebrew.
Example 2: Laughable tension created
by a coupling presented in humour
[Transcription conventions as above.]
F humorously presents a coupling of a positively appreciated phrase about

Jerusalem (‘Next year in Jerusalem’) shared by the Jewish community (and
uttered at the end of the Passover holiday dinner); this creates a tension with
the previous affiliation around pie parties. N laughs off this tension with F, and
they continue to align around pie afterwards (‘I mean next year more pie!’).
Affiliation is negotiated in this phase through humour by laughing off those
social networks that interactants might otherwise affiliate with in different con-
texts (in this case, the Jewish network in the context of Passover). It is through
couplings that conversational participants negotiate affiliation, and laughter
signals that this social process is going on by showing that tensions that will
inevitably affect how we co-construct ‘who we are’ can be laughed off together
for the sake of affiliating.
8.5 Humour and Naming
It is often difficult to determine where couplings are presented in conversa-

tions between friends, as they share a great deal of background knowledge and
values that they do not need to make explicit to share (see, for instance, the
implicit couplings in Example 2). Also, depending on the degree to which the
interactants have been involved in the communities they construe, they may
not need to inscribe the appraisal to express that meaning towards an experi-
ence that is well-known to community members. The value must instead be
extracted from ideational meanings to which they are implicitly bound.
However, these may be made evident through realizations of the system of
involvement (cf. Martin and White, 2005; Eggins and Slade, 1997), specifically
the resource of ‘naming’.
Involvement is situated as a system of discourse semantics alongside
appraisal and negotiation5 according to Martin and White (2005, pp. 34–5),
and involves non-gradable resources that negotiate the interpersonal relations
of ‘tenor’ (Halliday, 1985), or the roles and relationships, between interactants.
These resources are said to be ‘used to negotiate group identity and so co-
operate with appraisal and negotiation in the realization of tenor relations’
(Martin and White, 2005, p. 34), and are particularly oriented towards soli-
darity. Eggins and Slade (1997) incorporate involvement into their study of the
discourse semantics of casual conversation, and find that naming, technicality,
swearing and anti-language are used to construct degrees of group member-
ship and solidarity while also controlling and excluding other interactants.
Naming has a particular role to play in the construction of humour and affilia-
tion in the conversational texts that will be presented in Section 8.6, as it affords
(i.e. implies, see Martin and White, 2005, pp. 61–8) attitudinal meanings in
ideation, marking the presentation of couplings in text. Specifically, these
resources can afford attitude in couplings in convivial conversational humour
by setting up opposing perspectives on the part of named individuals, groups
and communities. As social networks are constructed by the clustering of bonds
in affiliation, naming these networks in discourse indicates that couplings are
being negotiated to construe their bonds.
When we look to realizations of naming in convivial conversational humour,
the construal of higher-level networks becomes salient, illustrating the negotiated
co-construction of identities on the personal, ideological and cultural levels of
affiliation (see Figure 8.1).
These realizations will be discussed in the following section along with their
role in the presentation of couplings in humorous phases of conversation. The
analyses will exhibit how conversational participants use humour to construe
bond (e.g. Fun Pie Party; Sacred Jerusalem)
Figure 8.1 Levels of affiliation with examples of their construal
affiliation as members of culture by variously negotiating their friendships,

ideological positionings and cultural subgroupings, made evident and relevant
by their naming in text. These ‘levels’ of identification may be aligned
with Tracy’s (2002) categorization of ‘master’, ‘interactional’ and ‘relational’
identities. That is, as interactants identify in different ideological networks,
including those of gender, age, ethnicity, capacity, class (as outlined by Martin
and White, 2005, p. 29), religion and sexual orientation (cf. Livia and Hall,
1997), as well as by national and regional origins (Tracy, 2002, p. 18), they con-
strue ‘master’ identities that constantly intercede the more local, personal
social networks they share. Lower-level affiliative identities are those such as
friendships and communities of interest that are both ‘interactional’ (situation-
and relationship-specific roles) and ‘relational’ (highly variable and negotiated
moment-to-moment with particular conversational partners). In convivial con-
versational humour, participants often interactionally negotiate local subgroup
identities, while at the same time construing overarching global identities
including Canadian and North American cultural categories along with ideo-
logical positionings. The different levels of participants’ ‘communal identities’
will be exhibited through their construal in convivial conversational humour in
the following section, informed by realizations of naming, and it will be shown
that humour enables interactants to make identities negotiable in affiliation.
8.6 Naming Culture in Humour: Results
By tracking instances of naming that occurred around laughter in the humor-

ous phases, a number of different categories of both personal and master iden-
tities were identified. Bonds were negotiated through humour by the interactants
to construe the following affiliations:
z National master identities as Canadians, negotiated through regional mem-

berships to ‘Toronto’, ‘Quebec’, and ‘St. Catharines’ (Ontario)
z Heterosexual, young female master identities, negotiated through ‘SNL’

television show watchers membership (including naming of actors ‘Chris
Farley’, ‘Patrick Swayze’ and American association ‘Chippendales’)
z Religious master identities as non-denominational (in relation to ‘Catholic’)
and as ‘Jewish’, negotiated through local theatre memberships
z Male and female master identities, referred to as ‘they’ and ‘girls’, negoti-
ated through membership to a personal group of friends
z National master identities as Canadians, negotiated in relation to master
identity as Thai (in naming of ‘Thailand’) and to female identities (‘they’)
These affiliations were made particularly salient by the naming in the text. For
instance, in the third text, religious master identities are made evident when
the speakers name the Catholic religion as one that they do not share but that
interrupts their theatre rehearsal time (‘we can’t have rehearsal on Easter
Sunday because mm-Adriana’s all Catholic’); and further, one speaker names
the Jewish religious network to associate another interactant with this member-
ship (‘Why can’t everyone be Jewish’). The identities thus emerged from the
text as the interactants made them relevant to the ongoing talk, and negotiated
the bonds of their social networks through couplings of attitude with ideation.
Coupled meanings surrounding laughter bouts (Owren, 2007) were inter-
preted and informed by the resource of naming.
The analysis of the texts confirmed that friends use conversational humour
to negotiate affiliation by identifying around shared and unshared bonds, and
laughing off the tensions that occur between them in their communal member-
ships to social networks. In convivial conversational humour, interactants nego-
tiate their similarities by laughing off their differences, so that any conflicting
coupling presented in discourse is laughable due to its relation to an underly-
ing shared bond. Through the convivial conversational humour in the texts,
the participants construed affiliation by laughing off the bonds of those identi-
ties that were not shared between them (e.g. ‘Catholic’), managing the tensions
they caused (e.g. ‘we can’t have rehearsal’), while reinforcing bonds that they
did share (e.g. rehearsing in theatre group). Moreover, while often negotiating
personal bond networks locally, the interactants also brought in identities on an
ideological and cultural level, accessing the macro-social structure through
humour (Eggins and Slade, 1997).
I will present the analysis of the humorous texts in this section, exhibiting
how the speakers present couplings both inscribed and afforded by surround-
ing named regions, associations and individuals. These names set up opposing
perspectives, pointing towards the values around which the interactants posi-
tion themselves at various levels of affiliative community; both in terms of local
networks, and in ideological and cultural networks of gender, age, ethnicity,
capacity, class, religion, sexual orientation, national and regional origins.
8.6.1 Canadian weather humour

The first humorous phase is a discussion between two participants about
Canadian weather, one from the province of Quebec and one from Ontario,
both living and talking in Toronto, Ontario after their holidays spent at home.
Together the participants laugh about the harsh winter weather conditions they
have experienced in their home regions:
N: It’s nice out!

P: Yeah!
[pause 2 secs]
P: So uh gaw… in Quebec it’s so frickin cold (N laughs) like ugh “I hate
this!”
(N continuous laughter) “I want to go back to Toronto” == yeah
N: == (laughing) where == it’s mild.
P: == Now everybody’s like “yeah [Toronto’s] so cold” I’m like “[Toronto’s]
not == cold”
N: == (laughs) [Toronto’s] not cold at all we had a deep freeze though
P: == Yeah?
N: == We had like a really cold few days I- I was in St. Catharines but really
really cold like you went outside and all of a sudden it was like whoa ==
P: == Oh really?
N: ( ) cold yeah so um . . . and . . . we ha- still had this big snowfall and then
like rain and stuff, I got stuck twice in my car (laughs)
P: Oh == (laughing) no!
N: == in the driveway (laughs) So [Ontario] got like really like sort of
(laughing) bad conditions (P laughs) for a couple days around Christmas.
Example 3: Canadian weather humour
[Transcription conventions as above. Implied targets
have been rendered in square brackets.]
These friends manage the tension presented by bad weather conditions in both
provinces through humour, laughing it off to align together as Canadians. The
first laugh signals this tension after P’s utterance ‘in Quebec it’s so frickin cold’,
which presents a laughable coupling of intensified negative appreciation for
Quebec’s weather. P construes herself both as a member of the Quebec provin-
cial community, and as one who had to endure intensely cold weather because
of this association, also toning down her offence with a euphemism rather than
its dysphemistic synonym (Allan and Burridge, 2005). The naming in this text
informs us of the construal of couplings, as P first set up the regions of Quebec
and Toronto as opposing each other, coupling negative appreciations with
Quebec weather and positive appreciations with that of Toronto. N turns this
around, however, by describing the ‘deep freeze’ that occurred in Ontario,
once again signalling the presentation of couplings around regions by specify-

ing her circumstance of location as ‘in St. Catharines’ and so expanding the
region beyond Toronto to Ontario (or at least Southern Ontario). By doing so,
she construes herself as an Ontarian and at the same time presents a negative
coupling for its weather and the consequences of it on her daily life (specifically
in driving). The opposition is made between the currently mild weather of
Toronto and previous weather conditions in both Quebec and Ontario, so that
the participants must laugh off both of these construals in order to align
together as Canadians who must endure this tension in the winter.
This social process of affiliation involves co-identification at the levels of
municipal, provincial and national communities, and these interactants affili-
ate together around enduring the weather to reinforce a Canadian master
identity as members of this culture, while laughing off their differing regional
memberships within it. The realizations of naming in the text make the opposi-
tion between these laughable couplings apparent, and also lay bare the over-
arching national identity that is underlying the humour. Weather is discursively
construed as something forced upon the speakers as inhabitants, and so by
making this consequential to the unfolding interaction, they cope in their cur-
rent situation as enduring Canadians. They are thus able to laugh off the ten-
sion and also identify their local regional memberships as unshared but
acceptable differences.
8.6.2 Attractiveness ideals humour

In the following phase, four female interactants discuss male attractiveness ide-
als following from a conversation about ‘male stripping’, and laugh around
what constitutes acceptable standards of body size and age for male strippers:
M: The funniest thing I’ve ever seen is- do you ge- it was like a really
really old SNL skit and Patrick Swayze was on and they were doing
this um
C: Chippendales?
M: The Chippendales ==
C: == With Chris Farley? == Yeah
M: == with Chris Farley!
G: == Oh::: yeah::: ew:::
L: == Oh my (laughing) god! …
M: == Chris Farley’s like rollin off his shirt == too and like he’s got
G: == Ew::: I think I’ve rented that one
C: [Chris Farley]’s all like this [non-verbal action]
M, G, L: (laugh)
C: And you realize like in that moment that Patrick Swayze’s kinda hot
and then you’re like “Oh!* Patrick Swayze!” (L,G laugh) and you’re
like == “Wait wait” . . .
M: == And then you look at the hair and you’re like “Oh wait [Patrick
Swayze’s] kinda old”
L: == (laughs)
Example 4: Attractiveness ideals humour
M opens up the humour by prefacing the phase with ‘the funniest thing I’ve
ever seen is’, and follows by describing a comedy skit on the American tele-
vision programme, Saturday Night Live (or SNL). The interactants then co-
construct the premise of this funny skit as one in which the actor, Chris Farley,
portrays the role of a male stripper in the American association ‘Chippendales’.
The laughter and expressions of disgust (e.g. ‘ew:::’) signal that this creates a
tension to be laughed off by these interactants, and while the attitude is not
explicitly inscribed, couplings are afforded by realizations of naming through-
out the phase.
First, ‘SNL’ as a specialized term for the television show sets up a community
of SNL-watchers by which these interactants may share in a communal identity,
and this is followed by the naming of the individual actor ‘Patrick Swayze’; the
community of American male strippers, the ‘Chippendales’ (a group of men
with ‘ideal’ body types); and the individual actor ‘Chris Farley’ (an actor with
an overweight body type). In this naming, M and C together set up an opposi-
tion6 between Chippendales and Chris Farley, affording a positive aesthetic
appreciation (cf. White, 2001) for the male strippers and a negative aesthetic
appreciation for the actor. By interpreting this opposition, the interactants co-
identify both as SNL-watchers, construing themselves as familiar with American
programmes and associations, and as heterosexual females with an interest in
‘attractive’ body ideals for men. At the same time, they laugh off the tension
created by Chris Farley in playing the role of an attractive male stripper through
such activities as ‘rollin off his shirt’, which would otherwise be desirable by
positively appreciated members of the Chippendales.
The underlying bonds that they share as they affiliate together are further
explored when C presents a coupling of positive aesthetic appreciation for the
actor Patrick Swayze as a Chippendale, which is met with laughter and an admis-
sion of their shared negative appreciation for him as too old. These couplings
are inscribed in ‘kinda hot’ and ‘kinda old’, but the naming again makes clear
the opposition of perspectives and exhibits the interactants as familiar with
popular American culture and the named actor ‘Patrick Swayze’. In the ongo-
ing humorous discourse around attractiveness ideals, they laugh off ‘oldness’
as an unshared bond in their heterosexual female identities. Thus, through the
local community of watchers of the American television show SNL, these con-
versational participants negotiate their ideological positioning in the gender
network and in terms of sexuality through different bonds around male attrac-
tiveness ideals.
8.6.3 Religion humour

Religious affiliations are brought into the humorous negotiation through local
theatre networks shared by the interactants in the following phase. Two non-
religious friends, C and G, joke about Catholicism with a Jewish friend, F, as it
intrudes into their theatre rehearsals during the Easter holidays. G also indi-
cates her close association with the Jewish religion as one who is familiar with its
values:
C: Do they have school on Easter Monday?

G: Yeah. == But I’m surprised.
F: == Some.
C: Yeah but, what’s the deal
G: I don’t know but I just ==
C: == Oh right
G: realized today that we can’t have rehearsal on Easter Sunday because
mm- Adriana’s all Catholic
C: Yeah nobody wants to rehearse on mine Friday either
F: Hey I have no problem with [rehearsing at Easter]
C: (laughs)
G: (laughs)
C: You and me Farley ==
G: == Why can’t everyone be Jewish (laughs) … There’s a good reason
for that
Example 5: Religious humour
C and G mark F’s utterance as creating a tension (in the same way as Example
2) by reacting with laughter and continuing the humour (Hay, 2001). While at
first the naming of the ‘Catholic’ network sets up an opposition between the
bonds of this network, named as the Easter holidays, and those of the speakers’
theatre play networks (implied in their ‘rehearsing’), this is then countered by
F by asserting his own affiliation with the Jewish community. When G names this
social network following her laughter (‘Why can’t everyone be Jewish’), she
exposes this underlying social network and creates further humour around it
by imagining all people as fitting into the Jewish category, noting that this is not
necessarily an affiliative possibility. Specifically, her naming makes clear the
underlying bond on which F’s coupling is made laughable, as he proposes a
positive appreciation coupling with rehearsing that is in fact due to his Jewish
membership (and not to his dedication as a play member). The couplings
inscribed around the play members’, and in particular Adriana’s, disinterest in
attending rehearsals are first associated with the quality of being ‘Catholic’,
affording negative attitudes towards the bonds that cluster together in this net-
work such as the Easter holidays. F’s assertion of a positive appreciation coupled
with rehearsing (‘I have no problem with that’) is funny because he is in fact
asserting a Jewish identity that is not here shared, but one that excludes any
potential bond around Easter, thus making his positive appreciation of rehears-
ing less ‘genuine’ than he makes it out to be.
These conversational participants negotiate the tension caused by master
identities in this phase, as membership to a religious network such as Catholi-
cism cancels the possibility for them to affiliate with their play members in
rehearsals during the Easter holidays. In humour they co-construct their own
ideological positionings as non-religious and as Jewish, and bond together as
theatre play members who, in any case, will not allow Easter to cancel their
rehearsing. By presenting a coupling to construe his Jewish identity, F seems to
licence the humour around religion7, and the naming makes the opposition,
and the humour, more apparent.
8.6.4 Gender humour

Ideological networks of gender are made relevant to the interaction through
their naming, and through the use of the exclusive pronoun ‘they’ (cf. Fairclough,
1989; Fairclough and Wodak, 1997) by the speaker in the following phase. This
begins with a discussion of the group of friends of one interactant, C, which is
then specified to the males within her group who she associates with feminine,
rather than masculine, values:
C: == They don’t like him

M: Why?
C: Cause they’re ga- they’re such girls! They’re the girliest boys I’ve ever met
in my life ==
N: == Really?
M: == Why don’t they like him
C: They all talk shit about each other like == woo::: they’re really girly
N: == Oh ma:::an …
M: == Oh my god
C: Dale was always like “Do you know what Irene said about Alan” and == I’m
like “I don’t care [about gossip]”
L: == (laughs)
Example 6: Gender humour
C first dissociates herself from her group of friends by referring to them as

‘they’, then specifies the males in her group and refers to them as ‘the girliest
boys I’ve ever met in my life’. In these utterances, C names the female gender
category (‘girls’) as a quality now associated with the alternate gender category
(in a ‘standardized relational pair’ according to Membership Categorization,
cf. Sacks, 1992) of ‘boys’, signalling opposing social networks and the presenta-
tion of couplings as bonds of these networks in the text. Gossip is then con-
strued as a bond typically associated with a female social network but one that
is shared by the males in her group, and one which C herself dissociates with
(‘I don’t care’).
Gender is brought into this conversation to negotiate the bonds that typically
cluster within these networks, and through humour these interactants laugh off
the tension created when males take on a ‘female’ bond. At the same time,
these interactants negotiate themselves as ones who do not share this bond, so
that physical associations with this ideological network do not prevent these
women from co-constructing their own ‘non-girly’ identities. So, as members of
the female gender they can affiliate together by establishing known valued attri-
butes and activities that are part of this category, while construing one bond
within it as too ‘girly’ for them, and laughing it off when presented as a posi-
tively appreciated coupling by ‘boys’. The use of exclusive pronoun ‘they’ first
sets up that gossip is presented as a coupling but that this is one that the speaker
dissociates herself from, and the opposition between ‘girls’ and ‘boys’ is made
clear through their explicit naming. Tracy (2002) argues that master identities,
such as this ideological network of gender, are frequently conceived of as ‘con-
trastive sets’ (p. 18), so that ‘gossip’ is defined by being a non-male bond. How-
ever, not only do these speakers acknowledge the tension created when males
take on this non-male bond, but they make it negotiable as something shared
by ‘girly’ females, and not necessarily other females such as them. This negotia-
tion is presented through the personal network of C’s group of friends, and
allows the interactants to co-identify by sharing a ‘non-girly’ subgroup within
the female identity.
8.6.5 National humour

Participants also identify values that vary across cultures at the national level,
acknowledging the tension they may cause by laughing off the assertion of these
values by others. In the following phase, the female gender network is negoti-
ated across the Canadian and Thai cultural communities as one Canadian male
speaker projects the speech of a Thai woman as funny:
CO: I just hope one day it does find its way to be like we [men and women]’re
equal; but we’re totally different sexes and it’s like y’know both [men
and women] are celebrated for their own thing and whatnot. …
CO: I don’t know this one girl sh- (laughs) I think we bo- Jess’ girlfriend this
one girl . . . with my friend, he had a bunch of clothing that was ripped
he’s like “Ah I have to take this to get this sewed up I don’t know how
to do that” and his girlfriend’s like was like uh “Oh I could do it” and
he’s like “Oh you know how to sew?” And her answer to that was- to the
question “d’you know how to sew,” she was like “You know I am a
woman of course!”
K: == (laughing) Oh my gah
N: == (laughing) (A woman?)
CO: == He was just like . . . yeah.
K: == (laughs) . . . Where was that?
CO: In Thailand
K, N: (laugh)
CO: She’s like “I’m a woman of course I can sew!” (K,N laugh) “I can clean
too you know” (K,N laughing) You know it’s like (laughing)
T: You know and they [women] can cook!
CO: Yeah they [women] can cook! (N laughing) == It’s like you know
K: == (laughing) Oh my god (laughs)
Example 7: National humour
[Transcription conventions and appraisal coding as above.]
While the projected speaker, ‘Jess’ girlfriend’, asserts her positive appreciation
for sewing as a bond that is necessarily shared between women, these interac-
tants interpret it as creating a tension and laugh it off in favour of the values
they had previously shared in the discourse as Canadians (around ‘equality’ of
the sexes). In her realization ‘of course’, Jess’ girlfriend presents the sewing
value as universally shared and excludes Jess’ potential alternative viewpoint as
‘at odds with what is purportedly generally agreed upon or known’ (Martin and
White, 2005, p. 124). This relegates sewing to being a woman’s duty, creating
tension with the shared bond of ‘equality’. The humour is then furthered by
T and CO who use the exclusive pronoun ‘they’ to both widen the network to
include all women, and to exclude themselves as male in the duties of cleaning
and cooking. The interactants show that they are negotiating social networks at
the national level by the naming of ‘Thailand’ as the circumstance of location
in which this coupling was presented by Jess’ girlfriend, and this seems to
licence the female interactants’ laughter as they construe themselves as differ-
ent from that network, instead aligning with the males in the conversation
around ideals of equality. The tension is drawn out by the projected speaker’s
naming of the female network in ‘I am a woman of course’, but this is laughed
off as distinctly Thai and as culturally different from the female network as
constructed in the Canadian culture. These interactants affiliate together on
a cultural level by naming the national network (‘in Thailand’), setting up an
opposition with their own implicit Canadian network, and laughing off the ten-
sion created by Jess’ girlfriend in her coupling around sewing as a bond of the
female network. This is instead construed as something unshared by the females
in the current conversation, and ungeneralized by the males as well.
8.7 Conclusions
The phases of convivial conversational humour that have been presented in

this chapter exhibit the way that friends use humour to manage the tensions
that occur in their ongoing negotiation of a communal, relational identity. In
the dynamic unfolding of the text, interactants present values both explicitly
and implicitly in order to laugh off the tensions created between them as bonds,
which tie social persons together as members of communities. These bonds are
realized in the linguistic text by couplings of attitude with ideational meaning,
and are often highly implicit so that it is difficult to interpret why the interac-
tants might find something funny in convivial conversational humour. How-
ever, it was demonstrated in this article that realizations of the involvement
resource of naming (and the use of the exclusive pronoun ‘they’) serve as tools
pointing towards the presentation of couplings in text, as they set up opposing
perspectives of communities constructed by clusters of bonds. By naming the
social networks under negotiation in their discourse, these interactants indicate
the comparison of bonds between them and present a tension to be laughed off
in humour.
A focus on naming in convivial conversational humour also exposes the con-
struction of higher level networks of ideological and cultural positioning by the
interactants, as these are named alongside the negotiation of their lower level
personal groupings. However, by bringing these positionings into the humor-
ous text, participants make these intruding hierarchically based roles negotiable,
rather than ‘slipping out of the non-hierarchic relations of the casual context
and into the hierarchic patterns of formal, public contexts’ to construe power
relations as proposed by Eggins and Slade (1997, p. 167). That is, these identi-
ties are always discursively available, and when taken up in convivial conversa-
tional humour, power relations can be laughed off rather than acted out as in
more formal contexts. This type of humour then offers a space for momentary
release from the tension created by these roles through laughter.
This chapter has demonstrated that through a process of affiliation in conviv-
ial conversational humour, culture is variously negotiated in the conversational
text. By applying an SFL perspective to the context of everyday conversation,
this negotiation becomes clear and a model of communities through ‘bonds’
has been developed to account for it. Conversational participants present ten-
sions to laugh off in order to align with their interlocutors and reinforce their
common identities. Both local and global concerns are made relevant in dis-
course, and relations of solidarity are reinforced and shaped by what the inter-
actants co-construe as ‘us’ and laugh off as the acceptable other. In interpreting
why these participants may find their shared talk funny, it is useful to engage
with the implicit meanings that inform us about the complexity of negotiation
hidden beneath a surface of laughter. Naming is one resource for the inter-
pretation of implicit attitude in couplings, and has presented an interesting
area for research into how ideology in particular is made relevant, negotiable,
and even laughable, in even the most common ‘casual’ conversation.
Notes
1
These categories will necessarily affect the data, but the shorter range of variation
arguably minimizes the effect of categories of age and education (see Hay, 1995,
pp. 35–8 discussion of control of variables).
2
That is, the system of appraisal that is situated in the stratum of discourse, above
lexicogrammar and below the social context, in the SFL architecture of language
(cf. Martin and White, 2005, pp. 9–12).
3
See Martin and White (2005) for further description of these systems of Attitude.
4
This is a view of affiliation from a systems perspective. Zhao and Knight (2009),
however, present a complementary process perspective on affiliation.
5
The system of ‘negotiation’ is introduced here in a technical sense as a discourse
semantic system, but throughout this chapter when I refer to the negotiation of
affiliation, it is meant in a non-technical sense.
6
This is something to be expected due to its being part of a comedy skit.
7
The utterance ‘Adriana’s all Catholic’ was not explicitly laughed off before but
expressed with a contrastive sarcastic intonation (cf. Attardo et al., 2003), indicat-
ing humour.
References
Allan, K. and Burridge, K. (2005). Forbidden Words: Taboo and the Censoring of
Language. Cambridge: Cambridge University Press.
Archakis, A. and Tsakona, V. (2005). ‘Analyzing conversational data in GTVH terms:
a new approach to the issue of identity construction via humor’. Humor 18 (1),
41–68.
Attardo, S. (1994). Linguistic Theories of Humor. Berlin: Mouton de Gruyter.
Attardo, S., Eisterhold, J., Hay, J., and Poggi, I. (2003). ‘Multimodal markers of irony
and sarcasm’. Humor 16, 243–60.
Boxer, D. and Cortés-Conde, F. (1997). ‘From bonding to biting: conversational
joking and identity display’. Journal of Pragmatics 27, 275–94.
Bernstein, B. (2000). Pedagogy, Symbolic Control and Identity: Theory, Research Critique.
Lanham: Rowman and Littlefield.
Coates, J. (2007). ‘Talk in a play frame: more on laughter and intimacy’. Journal of
Pragmatics 39, 29–49.
Eggins, S. (1990). Keeping the conversation going: a systemic-functional analysis of conver-
sational structure in sustained casual talk. Unpublished PhD Thesis, University of
Sydney, Sydney.
Eggins, S. and Slade, D. (1997). Analysing Casual Conversation. London: Cassell.
Fairclough, N. (1989). Knowledge and Power. London: Longman.
Fairclough, N. and Wodak, R. (1997). ‘Critical discourse analysis’, in T. van Dijk
(ed.), Discourse as Social Interaction. London: Sage, pp. 258–84.
Glenn, P. (2003). Laughter in Interaction. Cambridge: Cambridge University Press.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of
Language and Meaning. London: Edward Arnold.
Halliday, M. A. K. (1985). ‘Context of situation’, in M. A. K. Halliday and R. Hasan
(eds), Language, Context and Text. Victoria: Deakin University Press, pp. 3–14.
Hay, J. (1995). ‘Gender and humor: beyond a joke’. Unpublished MA thesis, Victoria
University of Wellington, Wellington, NZ.
Hay, J. (2000). ‘Functions of humor in the conversations of men and women’. Journal
of Pragmatics 32, 709–42.
Hay, J. (2001). ‘The pragmatics of humor support’. Humor 14, 55–82.
Holmes, J. and Marra, M. (2002). ‘Over the edge? Subversive humor between col-
leagues and friends’. Humor 15, 65–87.
Housley, W. and Fitzgerald, R. (2002). ‘The reconsidered model of membership
categorization analysis’. Qualitative Research 2 (1), 59–83.
Knight, N. K. (2008). ‘“Still cool . . . and american too!”: an SFL analysis of deferred
bonds in internet messaging humour’, in N. Norgaard (ed.), Systemic Functional
Linguistics in Use. Odense: Odense Working Papers in Language and Communi-
cation, vol. 29, pp. 481–502.
Knight, N. K. (2010). ‘Wrinkling complexity: concepts of identity and affiliation in
humour’, in M. Bednarek and J. R. Martin (eds), New Discourse on Language: Func-
tional Perspectives on Multimodality, Identity, and Affiliation. London: Continuum,
pp. 35–58.
Knight, N. K. (in progress). ‘Laughing our bonds off: Conversational humour in
relation to Affiliation’. PhD thesis in progress, University of Sydney, Australia.
Kotthoff, H. (1996). ‘Impoliteness and conversational joking: on relational politics’.
Folia Linguistica 3–4, 299–325.
Labov, W. (1972). Sociolinguistic Patterns. Philadelphia, PA: University of Pennsylvania
Press.
Livia, A. and Hall, K. (1997). Queerly Phrased: Language, Gender and Sexuality.
New York: Oxford University Press.
Martin, J. R. (2000a). ‘Beyond exchange: appraisal systems in English’, in S. Hunston
Discourse. Oxford: Oxford University Press, pp. 142–75.
Martin, J. R. (2000b). ‘Factoring out exchange: types of structure’, in M. Coulthard,
J. Cotterill and F. Rock (eds), Working with Dialogue. Tubingen: Niemeyer,
pp. 19–40.
Martin, J. R. (2008a). ‘Innocence: realization, instantiation and individuation in a
Botswanan Town’, in A. Mahboob and N. K. Knight (eds), Questioning Linguistics.
Newcastle: Cambridge Scholars Publishing, pp. 32–76.
Martin, J. R. (2008b). ‘Tenderness: realization and instantiation in a botswanan
town’, in N. Norgaard (ed.), Systemic Functional Linguistics in Use. Odense: Odense
Working Papers in Language and Communication, vol. 29, pp. 31–62.
(2nd edn). London and New York: Continuum.
Norrick, N. (1994). ‘Involvement and joking in conversation’. Journal of Pragmatics
22, 409–30.
Norrick, N. (1993). Conversational Joking: Humor in Everyday Talk. Indianapolis:
Indiana University Press.
Owren, M. J. (2007). Understanding Acoustics and Function in Spontaneous Human
Laughter. Proceedings of the Interdisciplinary Workshop on the Phonetics of
Laughter, Saarbrucken, Germany, 4-5 August.
Partington, A. (2006). The Linguistics of Laughter: A Corpus-Assisted Study of Laughter-

Talk. London and New York: Routledge.
Priego-Valverde, B. (2003). L’humour dans la conversation familière: Description et
analyse linguistiques. Paris: L’Harmattan.
Sacks, H. (1992). In G. Jefferson (ed.), Lectures on Conversation, Vols. 1 and 2. Oxford:
Blackwell.
Schegloff, E. A. (2007). ‘A tutorial on membership categorization’. Journal of
Pragmatics 39, 462–82.
space’. Unpublished PhD thesis, University of Sydney, Sydney.
Tannen, D. (2005). Conversational Style: Analyzing Talk among Friends. New York:
Oxford University Press.
Tracy, K. (2002). Everyday Talk: Building and Reflecting Identities. New York: Guilford
Press.
White, P. R. R. (2001). ‘Attitude/Appreciation’. Manuscript (word processor
version). www.grammatics.com/appraisal (accessed 9 October 2008).
Zappavigna, M., Dwyer, P. and Martin, J. R. (2008). ‘Syndromes of meaning: explor-
ing patterned coupling in a NSW youth justice conference’, in A. Mahboob and
N. K. Knight (eds), Questioning Linguistics. Newcastle: Cambridge Scholars
Publishing, pp. 164–85.
Zhao, S. (2010). ‘Intersemiotic relations as logogenetic patterns: towards the resto-
ration of the time dimension in hypertext description’, in M. Bednarek and J. R.
Martin (eds), New Discourse on Language: Functional Perspectives on Multimodality,
Identity, and Affiliation. London: Continuum, pp. 195–218.
Zhao, S. and Knight, N. K. (2009). Coupling and genesis: what is the ‘process’ in affilia-
tion? Paper presented at the SFL Seminar Series, 24 April, University of Sydney,
Australia.
Chapter 9
Visualizing Appraisal Prosody

Michele Zappavigna, Chris Cléirigh, Paul Dwyer and J. R. Martin1
9.1 Introduction: Appraisal Prosody
Appraisal prosody refers to the patterning of evaluation in a text. Evaluative

patterns are prosodic in the sense that they are not reducible to constituent
parts but instead resonate across the text as it unfolds in time (Halliday, 1981;
Martin and White, 2005). This means that their boundaries are fuzzy rather
than distinct (Macken-Horarik, 2003a). Appraisal patterns construe meanings
that are interpersonal, that is, they are about negotiating the nature of the rela-
tionship of interactants. Cléirigh (in preparation) suggests that interpersonal
meaning is always on standby for performing this function:
The interpersonal dynamics of the clause can then be understood as intrud-

ing on demand as an onrush, a surge, a billowing of the chemical ‘sea within’.
Interpersonal potential is, in a sense, vigilant in the background, waiting for
opportunities to burst out through the surface of form into the foreground
of meaning-making.
Along with the wave metaphor deployed above, metaphors that have been used
to characterize interpersonal patterning include crescendo and diminuendo
of a musical prosody (Martin and Rose, 2003), chord progressions (Macken-
Horarik, 2003b), movement of motifs (Halliday and Mattheissen, 2004), propa-
gation (Lemke, 1998) and radiation (Hood, 2006).
Appraisal prosody is both local and global in nature, being visible in a single
clause and in extended portions of discourse. Poynton (1984) suggests two
types of prosodic structure: diffuse and compact prosodies. For example, con-
sider the following:
Christ they bloody well beat the living daylights out of those LOUSY ROTTEN
STINKING bastards!
Poynton (1984, p. 23) argues that the items marked in italics above are part of
a diffuse prosody of negative attitude, while the items in capitals are ‘a subset of
Visualizing Appraisal Prosody 151
this attitudinal lexis which seems more constituent-like insofar as it is localized’.

Diffuse prosodies are involved in long-range text patterns. For example, Macken-
Horarik (2003, p. 307), in work very relevant to the prosodic structure explored
in this chapter, has suggested the role of ‘opposing appraisal choices’ in creat-
ing contrasts in phases of narrative discourse.
However, if we are to talk convincingly about long-range evaluative pattern-
ing then we face the problem that our evidence is somewhat intractable: the
patterns we are interested in extend beyond a single page or screen, in essence
they extend beyond what we can hold in consciousness in a given moment.
Thus, when working with our Youth Justice Conference (YJC) data we found
that we required visualization strategies to supplement the work of close
discourse analysis. These visualizations were intended to make sense of the
appraisal patterns annotated in the Affray YJC transcript following the system
networks proposed by Martin and White (2005). In contrast to most descriptive
statistics, this visualization technique preserves logogenesis, the sequencing of
the text, while achieving a synoptic perspective on the prosodic structure.
This chapter aims to explore the long-range appraisal prosody involved in
the discourse of an Ethnic Liaison Officer (ELO) talking in a NSW YJC. We
begin by introducing the context of situation (Halliday and Mattheissen, 2004),
a conference convened to sentence a young person (YP) who has committed an
affray. We explain the system of attitude2, a component of appraisal theory
(Martin and White, 2005) that was used to annotate the transcribed video
of the YJC. The ELO’s intervention in the stage of the conference where the YP
is prompted to describe and evaluate his offending behaviour is the focus of the
data analysis. While we begin by exploring the ELO’s verbiage, we also draw
upon gestural and phonological evidence to support our claims about the pro-
sodic structure of his evaluative language. The second part of the chapter
explains a visualization strategy using Stacked Area Graphs to interpret the
unfolding of this evaluative language. We conclude by making some prelimi-
nary suggestions about the role of the prosodic structure in an overall ‘push-
pull’ affiliation strategy that is part of the integrative function of the ELO’s
discourse.
9.2 NSW Youth Justice Conferencing
We take as a case study of a prosody talk between an ELO and a YP in a NSW

YJC convened to sentence the YP for an affray offence. YJCs are meetings held
to sentence a YP who has been convicted of an offence as an alternative to
attending the Children’s Court. They are typically held in a circle configuration
(See Figure 9.1) in which participants are afforded the opportunity to talk
about their experiences and feelings surrounding the offence. These partici-
pants may include the victim of the crime, support persons for both the victim
and the YP, police officers such as Youth Liaison Officers and ELOs. For brevity
the conference will be referred to as the Affray YJC.
Figure 9.1 Seating configuration of the ‘Affray YJC’ (blurred to maintain

anonymity)
The texts in our dataset include both observed and video-recorded confer-
ences. We thus adopt a multimodal text analysis strategy that considers both
talk and gesture. In this chapter we focus on evaluative meaning in a single YJC,
the Affray YJC, and use textual, phonological and gestural evidence to support
our claims about the structure of the ELO’s appraisal prosody. The study is
an example of ‘appliable linguistics’ concerned with understanding linguistic
patterns in real-world contexts.
The Affray YJC was one hour and a half in duration. The discourse was tran-
scribed using Transcriber (Boudahmane, Manta, Antoine, Galliano and Barras,
2001) and annotated using ELAN (Max-Planck-Institute for Psycholinguistics,
2008). In undertaking our analysis of the video-recorded YJC we were particu-
larly motivated by Bateman’s (2008) focus on finding ways to manage dynamic,
‘multichannel’ data so that our multimodal analysis would remain empirical:
When we move to multimodal documents, we also need to be able to find

concretely identifiable empirical evidence to motivate particular structures
and interpretations rather than others. Only then do we have a foundation
sufficiently firm for further theory building. (Bateman, 2008, p. 187)
Thus, ELAN was chosen as an annotation tool as it allowed us to track multiple

data series while concurrently tracking the time series of the video text. These
data series took the form of different ‘annotation tiers’ (see for example the
tier labelled ‘tone group’ near the bottom of Figure 9.2. Participants have been
blurred for anonymity).
Visualizing Appraisal Prosody
Figure 9.2 An example of tone group analysis performed in ELAN
153
Figure 9.3 Configuration for recording a YJC
The Affray conference was video recorded and the researchers also sat in
the conference circle as silent observers. To minimize the visual presence of
recording equipment in the hope of reducing their impact upon conference
participants, video-recordings were made using a camera mounted on a micro-
phone stand recording to DVD. The cameras were positioned in the corners of
the room to further reduce their visual presence (See Figure 9.3).
The manual used in NSW as part of convenors’ training includes a scripted
outline (Youth Justice Conferencing Directorate, 2005) detailing the structure
of a typical conference. This notional script indicates that the convenor should
invite the YP to ‘tell their story’. The YP, however, in the conferences that
we have observed, will rarely produce a packaged recount of events. Both the
details of the offence and related emotion talk are usually jointly constructed
through prompting by the convenor. For this reason we refer to the story that
the YP tells as a Commissioned recount. It typically occurs in the generic
structure of the conference after the gathering and legal framing, and before
the responses by third parties to the YP’s ‘story’ (For an account of the Commis-
sioned recount stage see Martin, Zappavigna and Dwyer, this volume). The
Commissioned recount tends to have the following structure:
Orientation ^ Record of events ^ (Re-orientation) ^ (Extension) ^ Interpre-

tation ^ Ramifications
The talk by the ELO analysed in this chapter is part of the Interpretation
phase of the Commissioned recount stage. We will refer to his contribution to
the Interpretation stage informally as the ‘ELO’s intervention’.
9.3 The System of Attitude
Within appraisal theory (Martin and White, 2005) attitude is defined as

‘the region of meaning construing our attitudes to people and the way they
behave – their character (how they measure up)’ (Martin and White, 2005,
p. 52). System networks are used within SFL to model the semiotic choice
involved in creating meanings of this kind. The ELO’s talk was annotated using
the system network in Figure 9.4 (this network is abridged with the judgement
system shown at a greater level of delicacy). Square brackets represent a choice
between different options, while braces indicate choices that may be selected
simultaneously. Diagonal arrows are used to show example realizations of
each path in the network and were taken from the corpus of the ELO’s dis-
course. As this figure indicates, in addition to attitude there are two other
semantic regions that may be occupied by evaluative language, engagement
and graduation.
The ELO’s use of the judgement system is of particular interest due to the
claims that have been made in restorative justice that a conference should cri-
tique the YP’s behaviour rather than their identity in an effort to avoid labelling
the YP as a deviant (Braithwaite, 1989). The ELO explicitly references this kind
of objective:
We’re not saying, ‘YP, you’re a this and you’re a that’. What we’re saying is
your behaviour is getting you into trouble, man. You’re hurting your family,
brother.
9.4 The ELO’s Intervention
The ELO’s intervention in the Commissioned recount comes after the Conve-
nor has attempted to elicit, extend and finally begin to jointly evaluate the YP’s
recount of the events of the affray. In the Interpretation phase proceeding the
intervention it appears that the Convenor has not been able to leverage shared
networks of value that we see the ELO deploy in his wielding of the judgement
system. Instead she relies on questions relating to emotions about family mem-
bers in the hope of eliciting an evaluative response. The Convenor often packs
her evaluation into questions for the YP to confirm (or deny), instead of state-
ments (C = Convenor, YP = Young Person):
C: But do you think it’s reasonable if you’re doing the wrong thing that
she’s telling you off?
YP: Yes.
C: Why is that?
YP: I’m in the wrong. She has the right to.
156
Figure 9.4 Situating judgement within the appraisal system (based on Martin and White, 2005)
C: Because she is your . . .?

YP: My mum.
C: Yep. And she got upset?
YP: Yes.
C: Was she crying?
YP: I can’t remember.
C: Screaming and yelling?
YP: I don’t know.
C: What about the rest of your family? What did they think about it? What
did your sister think? Your older sister?
YP: She didn’t tell me what she thinks. I don’t know-
C: Didn’t she? What about your uncle?
YP: I haven’t spoken to him.
Deciding that the proceeding phases have been ineffectual in deploying the
appropriate ‘evaluative levers’, the ELO declares that he will deploy a different
strategy:
ELO: Listen, I want to take, with you permission, I wanna take a different
angle. OK? Alright. Mate, what’s your mum wearing on her head?
YP: Scarf.
Here the ELO begins by invoking a bonding icon, or ‘bondicon’ (Stenglin,

2004) of the Islamic culture: the Hejab. This bondicon is infused with positive
evaluation and representative of two important values in the culture: respect
for Islam and respect for the mother. These values become to motifs to which
the ELO returns repetitively in his intervention.
After opening, somewhat dramatically, with this bondicon that distinguishes
the ELO and the YP from the other conference participants, the ELO then
begins what will be a sustained contrastive rhetoric by dividing the participants
into members and non-members of the Islamic culture:
ELO: Yeah. OK. What a- where is she now? In the presence of who?
YP: Me.
ELO: No, who’s sitting here? Who’s sitting here right now? Have a look
across.
YP: Men.
ELO: Have a- m- have a look across what uniform are they wearing?
YP: Police uniform.
ELO: OK. Where are these guys from? They’re from a certain place. OK.
What’s the perception going to be?
YP: Think bad of me.
The division made at the level of discourse semantics is also in the ELO’s
gesture as he tilts his head towards the different participants in the circle. We
now turn to exploring the prosodic structure of the kind of evaluative contrasts
that commenced in this phase.
9.5 The Prosodic Structure of the ELO’s Intervention
The evaluative language deployed by the ELO gives his discourse, from a
layperson’s perspective, the feel of a ‘good cop – bad cop’ routine, while the
Convenor sounds like a stern but caring mother. The ELO’s talk is constituted
by a relatively high frequency of negative judgement. A sample of this
appraisal is provided in concordance layout in Table 9.1. The judgement is
usually about social sanction and the target of this evaluation is typically the
YP’s behaviour.
From a corpus-based perspective the discourse involves an overall ‘halo’
(Bednarek, 2006) of negative judgement (negative judgement underlined):
You have no respect for your mum whatsoever, brother. You have no respect
for what your mum’s got on her head. You have no respect for our commu-
nity. You have no respect. You tell me, brother, how it’s a part of our culture
or our religion or our tradition to do things like that. You tell me when.
However if we re-factor in time, we see an oscillation of evaluation rather than

a constant level of negative judgement. The ELO alternates between negatively
judging YP’s behaviour and positively judging the cultural identity, of which
the YP is an instance. So, considering invoked3 positive judgement, ‘our
community’, ‘our culture’ and ‘our tradition’ are deployed as positive evaluated
tokens of the Lebanese Islamic culture, in which the ELO locates himself (‘I’m
Muslim background myself’). It is this community into which the ELO is calling
the YP to integrate.
Logogenetically, the patterning of judgement is typically construed as the
following proposal (positive judgement marked in capitals; negative judge-
ment underlined):
Table 9.1 Examples of negative judgement in the Affray YJC

Think BAD of me.
your saying to me and DISRESPECT your mother.
everybody here we can
You have NO RESPECT for your mum whatsoever, brother.
You’ve gotten off on the WRONG FOOT by pretty much abusing the rest of
us here by being late
It’s what you’re doing that’s no good. You’re probably a GOOD PERSON.
What you’re doing is not good. You understand the difference?
It is also construed as the following rhetorical questions:
what you’re doing does that HELP OUR COMMUNITY at the moment or
does it make our community look worse?
This type of contrastive prosody is constituted by two evaluative prosodies of

opposing polarity running in parallel: one positively evaluating tokens of the
Islamic culture and the other negatively evaluating the YP’s behaviour. So in
effect, what we see is a prosody of prosodies (for a related account of patterns
of patterns see Lemke (1998)) (Figure 9.5).
The ELO uses the contrast between these two evaluative prosodies for rhe-
torical impact. The effect is particularly visible when we consider both inscribed
and invoked judgement, that is, evaluation that is explicitly encoded and eval-
uation that is implied (positive judgement marked in capitals; negative
judgement underlined):
Do you know how easy it is to break the law? It’s so easy. [clicks fingers] It’s
so easy. [clicks fingers] It is so easy. Get in your car, do a u-turn on double
yellow lines, you’re breaking the law. You TRY STAYING OUT OF TROUBLE.
You find out HOW MUCH OF A MAN it takes, hey?, TO STAY OUT OF
TROUBLE. You tell me which one takes MORE OF A MAN to do. How does
it take MORE OF A MAN to listen to your friends or TO SAY TO YOUR
FRIENDS, ‘No, what you’re doing is wrong’. You tell me.
In the above, there is contrastive play of negatively judged, unlawful conduct

and positively judged ‘masculine’ behaviour (which is unpacked as ethical
conduct).
Figure 9.5 Distribution of evaluation across the gesture space

9.6 Gestural and Phonological Evidence
The ELO’s paralanguage accords with the kind of contrastive prosodic struc-
ture explained in the previous section. Both his gestures and phonological pat-
terns emphasize the rhetoric of opposites. For example, the ELO tends to divide
his gesture space into halves, with gestures relating to negative judgement on
his right, and positive judgement on his left (the following screen captures
are cropped and blurred for anonymity).
The gestures of ‘left-good’ are made towards the YP, and the ‘right-bad’
align away from the YP. The ELO is associating the YP with good, so valuing him
as a person; he is also urging him to put aside the bad, and rendering those
behaviours as spatially distinct from the YP as a person. The to-and-froing seems
to have a pedagogic function, contrastively evaluating types of behaviour.
The evaluative meanings being contrasted correspond with tonic promi-
nence, making them the informational focus; for example, consider the ELO’s
refocusing of the target of evaluation away from the YP’s identity to the YP’s
behaviour. The excerpt is annotated using Halliday’s (1970) notation scheme,
where each foot begins with the beat, either salient syllable or silent beat and /
= foot boundary, // = tone group boundary:
// 1 Not be/cause of / you as a / person // 1 be/cause of / something

you’re / doing that’s not / right. // 13 We’re / not / targetting you / person-
ally / brother. // 1 It’s / what you’re / doing that’s no / good. // 1 You’re /
probably a / good / person. // 1 What you’re / doing is not / good. //
The instances of tone 1 in this excerpt are tone 1+ (wide drop in pitch on the
tonic) which, according to Davies (1992), typically has a contrastive textual
function.
The ELO’s talk has a ‘pulsed’ character involving a fairly low number of
syllables per foot, generally between one and three. He tends to stress words
which in other contexts might remain unstressed. The feet have fewer than
usual syllables. This means that more words than usual are salient as the ELO
attempts to emphasize the importance of the contrasts that he is making. For
example, the following tone group analysis shows the rhythm of the ELO’s
discourse as he critiques the YP’s behaviour. In this analysis the beat falls on the
next syllable to the right of the single slash.
// You have / no re/spect for your / mum / whatso/ever / brother. // You

have / no re/spect for / what your / mum’s got on her / head. // You have
/ no re/spect for our com/munity. // You have / no re/spect. // You / tell
/ me / brother // when was it / part of /our / culture or // our re/ligion
or // our tra/dition to // do / things / like / that. // You / tell me / when.
// You / show me / where in the ko/ran it / says we can / do things / like /
that. // You / show me / where. // Tell me / where. // Does it? //
9.7 Using Visualization to Understand the

Global Patterning of Prosodies
When dealing with extended texts or with corpora linguists face the problem of
making their analysis tractable so that they can view long-range prosodic pat-
terns in order to comment about their structure. Visual aids for appreciating
such patterning need to provide to features: an ability to track the unfolding of
the text as a time series and the capacity to represent multiple annotation series.
This is because the linguist is interested in how patterns unfold logogenetically
and how multiple features work together to make meaning. A potential source
of assistance is the field of Text Visualization, an area related to Information
Visualization and Scientific Visualization that attempts to leverage the charac-
teristics of human visual perception to present data in ways that make salient
particular kinds of information or patterns.
A metaphor used within Text Visualization for representing time series data
is the ‘stream’. This metaphor has been applied in visualization methods such
as ThemeRiver and StreamGraph (Havre et al., 2002; Byron, 2008; Byron and
Wattenberg, 2008; Clark, 2008). The StreamGraph and ThemeRiver systems
have been used for representing the unfolding of sets of lexical features that
can be automatically extracted from large corpora using natural language pro-
cessing techniques. For example, ThemeRiver has been used to visualize lexis
in Associated Press news wires with some prominent features in the context
marked above the graph (Figure 9.6).
Stacked Area Graphs, a primitive of the StreamGraph technique, is a
technique more appliable to small datasets such as a transcribed text of a YJC.
Figure 9.6 Themeriver visualization of Associated Press news wire stories

(Havre et al., 2002, p. 12)
This is because it is more amenable to working with smaller, annotated corpora

where we are interested in discourse semantic features that cannot, as yet, be
identified automatically in a text by a machine and must be hand-annotated.
Stacked Area Graphs are a form of area graph, a graphing technique where
the space between the data curve and the horizontal axis is shaded (See
Figure 9.7). The vertical axis represents the frequency of the feature and the
horizontal axis the time series. The shaded region represents the number of
instances within a time period.
While an area graph usually shows a single data series, Stacked Area
Graphs represent multiple data series by stacking one on top of the other (See
Figure 9.8). The height of the curve at a given point represents the total fre-
quency of all features at that point and thus each data series should be read as
starting at zero rather than as their accumulative height. This makes the graph-
ing technique most useful to a linguist interested in the general trend of a data
series, or in other words, the qualitative ebb and flow of the annotated appraisal
over the time series. It is also a useful technique for appreciating the relation-
ships between the data series as they unfold by the overall impression of the
relative amount of colour.
The ordering principle applied to the horizontal axis will depend on the
kind of data series that are going to be represented in the graph. In the case of
video data we may use the time series in the video encoding to represent tem-
poral sequencing. In other cases it may be more appropriate to use ‘text time’.
Figure 9.7 An example of a simple area graph
Figure 9.8 A simple example of a stacked area graph

Figure 9.9 Positive and negative judgement (invoked and inscribed) in the ELO’s
intervention stage
For example, if we are interested in showing the unfolding of negotiation in a

text, we may use exchanges as the unit along the x-axis, with the y-axis showing
the frequency of particular exchange types.
Figure 9.9 is a stacked area graph showing the unfolding of judgement in the
ELO’s intervention. Negative judgement is shown on the lower data series in
black and positive judgement on the upper series in grey. The graph depicts
only ELO–YP talk and the blank areas in the graph (e.g. 0:04:00–0:05:30) are
phases in the text when these participants were not talking, typically because
another participant such as the Convenor or Arresting Officer was speaking.
The ELO’s intervention begins with an accumulation of negative judge-
ment (Figure 9.9, 0:00:00–0:01:50) such as the following:
You wanta be tough. [inaudible] You’re not. Number one. Number two, man.
When I see someone of my own background bringing their mum in wearing
a Hejab, OK, honestly man inside I feel sick. You understand?
After this initial burst of negative judgement, the patterning in Figure 9.9
shows a ‘to-and-fro’ structure, with negative judgement generally being fol-
lowed by positive judgement, shown in the graph as grey and black regions
co-occurring. This provides us with a synoptic perspective on the following kind
of structure (invoked negative judgement underlined; invoked positive
judgement in capitals):
Mate, if you – if there’s a fire do you walk straight into it or do you WALK
AROUND IT?
However the opposite configuration was also possible:
. . . what it sounds like is if you get a job, you’ll be out of trouble but if you
DON’T GET A JOB, you’ll GET INTO TROUBLE.
The stacked area graph provides us with a view of the evaluative tendencies of
the ELO without losing information about how the appraisal unfolds in terms
of its sequencing.
9.8 Conclusion: Appraisal Prosody and Affiliation
The patterning of evaluative language in a text construes ‘the kind of commu-

nity that is being set up around shared values’ (Martin and Rose, 2003, p. 54).
The final section of this chapter takes up Martin and White (2005) and Knight’s
(2008) ideas about the role of evaluative language in affiliation to make some
preliminary assertions about the integrative function of the ELO’s talk. While it
is beyond the scope of this chapter to deal thoroughly with different theories of
affiliation, we wish to briefly comment on the relationship of the back-and-forth
movement in the evaluative language to the ELO’s attempts to create solidarity
with the YP.
The appraisal prosody that we have identified appears to be involved in the
broader contrastive rhetoric employed by the ELO. The instances where nega-
tively and positively evaluated behaviours are set in opposition to one another
seem to have a pedagogic function, laying out the values that the YP should
hold as part of the Lebanese Islamic community. For example, this community
is contrasted with the YP’s mates network:
ELO: Where’s your mates now?

YP: They’re at home
ELO: At home. Why aren’t they with you, supporting you, brother?
Knight (2008) has suggested that couplings of ideational and evaluative

meanings are involved in the process of affiliation. The following is a coupling
of the YP’s behaviour, which has previously been elaborated as breaking the law
and disrespecting his mother, with negative judgement (Figure 9.10):
The accumulation of coupling such as ‘respect + mother’, ‘respect + Hejab’,
‘respect + community’ has an instructional function. The ELO is attempting to
make transparent the evaluative bonds (Knight, 2008) that the YP should hold
if he is to be part of the local Lebanese Islamic community. The alternation
Figure 9.10 An example of coupling

between presenting appropriate and inappropriate bonds is part of the rhetoric

aimed at integrating the YP into this community.
In tandem with this alternating is a to-and-fro in involvement, that is, in the
negotiation of power and solidarity. The ELO employs vocatives of solidarity of
three main kinds: generation (I’m thirty six, man. I’ve never been spoken to once by
the coppers), ethnicity (‫( ﺍﻩﻝﻙ ﺓﻕﻁﻥﻡﻝﺍ‬Kelaha almnteaha bet’heen um), brother [You are
insulting all the area, brother]) and gender4 (Mate, ya- your mate didn’t have to go
down there did he?). ‘Mate’ generally occurred at the beginning of the clause as
a Tone 3 (low rising) (Halliday, 1970) functioning as a warning. ‘Brother’ often
occurred as a Tone 13 (fall + low rise) functioning as a call to pay attention:
Mate (Tone 3), what’s your mum wearing on her head?

You tell me brother (Tone 13) when was it part of our culture or our religion
or our tradition to do things like that.
The overall effect of this kind of patterning seems a kind of push–pull affi-
liation where the ELO aligns himself with the YP while concurrently using
the reduction in interpersonal distance as an opportunity to judge the YP’s
behaviour. It is in this sense that the vocatives act as levers to open up the
discourse and let the evaluative prosody in.
In this chapter we have reported on an appraisal prosody that was concurrent
with contrastive gestures, phonological contrast and with a rhetoric of opposi-
tions. We have used multimodal evidence and visualization of extended text
patterns to make these claims about a prevalent back-and-forth structuring
visible across strata. The ELO’s discourse is an example of persuasive language
wielded for the common good. His attempt to integrate the YP into the local
community is an effort at civic rehabilitation and restorative justice.
Notes
1
We wish to acknowledge the support of the Australian Research Council in fund-
ing this research project.
2
Technical terms relating to appraisal are presented in small caps to distinguish
them from common terms.
3
Following Martin and Rose (2003), we make a distinction between inscribed
appraisal that is explicitly construed, such as ‘bad person’, and invoked appraisal
which is indirectly construed such as ‘didn’t get a job’.
4
Mate, however, is becoming more frequently used by female speakers of Australian
English.
References
Bateman, J. (2008). Multimodality and Genre: A Foundation for the Systematic Analysis of
Multimodal Documents. London: Palgrave Macmillan.
Bednarek, M. (2006). Evaluation in Media Discourse. London: Continuum.

Boudahmane, K., Manta, M., Antoine, F., Galliano, S. and Barras, C. (2001).
Transcriber version 1.5.1. Retrieved 5 November 2008, from http://trans.
sourceforge.net.
Braithwaite, J. (1989). Crime, Shame and Reintegration. Cambridge and Sydney:
Cambridge University Press.
Byron, L. (2008). Last.FM listening history – What have I been listening to? Retrieved
31 July 2008, from Lee Byron : http://www.leebyron.com/what/lastfm/.
Byron, L., and Wattenberg, M. (2008). Stacked graphs – Geometry & aesthetics. Retrieved
8 July 2008, from Lee Byron: http://www.leebyron.com/else/streamgraph/.
Clark, J. (2008). ‘Twitter topic stream’. Retrieved 31 July 2008, from Neoformix:
Discovering and illustrating patterns in data: http://www.neoformix.com/2008/
TwitterTopicStream.html.
Cléirigh, C. (in preparation). The life of meaning.
Davies, M. (1992). ‘Prosodic cohesion in a systemic perspective’, in P. Tench (ed.),
Studies in Systemic Phonology. London: Pinter, pp. 221–6.
Halliday, M. A. K. (1970). A Course in Spoken English: Intonation. London: Oxford
University Press.
Halliday, M. A. K. (1981). ‘Text semantics and clause grammar: some patterns of
realization’, in J. Copeland, and P. W. Davis (eds), The Seventh LACUS Forum.
Columbia: Hornbeam Press, pp. 31–59.
Halliday, M. A. K. and Mattheissen, C. M. I. M. (2004). An Introduction to Functional
Havre, S., Hetzler, E., Whitney, P. and Nowell, L. (2002). ‘ThemeRiver: Visualizing
thematic changes in large document collections’. IEEE Transactions on Visualisation
and Computer Graphics 8 (1), 9–20.
Knight, N. K. (2008). ‘“Still cool . . . and american too!”: An SFL analysis of deferred
bonds in internet messaging humour’, in N. Norgaard (ed.) Systemic Functional
Linguistics in Use. Odense Working Papers in Language and Communication,
vol.29, pp. 481–502.
Lemke, J. (1998). ‘Resources for attitudinal meaning: Evaluative orientations in
text semantics’. Functions of Language 5 (1), 33–56.
Macken-Horarik, M. (2003a). ‘APPRAISAL and the special instructiveness of
narrative’, Text 23 (2), 285–312.
Macken-Horarik, M. (2003b). ‘Envoi: intractable issues in appraisal analysis’, Text 22
(2), 313–19.
Martin, J. R. and Rose, D. (2003). Working with Discourse: Meaning beyond the Clause.
London and New York: Continuum.
Max-Planck-Institute for Psycholinguistics (2008). ELAN Version 3.6.0. http://www.
lat-mpi.eu/tools/elan.
Poynton, K. (1984). ‘Names as vocatives: forms and functions’. Nottingham Linguistics
Circular 13 (Special Issue on Systemic Linguistics) (1), 1–34.

space’. Unpublished PhD thesis, University of Sydney.
Youth Justice Conferencing Directorate. (2005). ‘Conference convenor training
package’. Unpublished materials distributed at workshops for new convenors
conducted by staff from the NSW Department of Juvenile Justice. NSW.
Chapter 10
Evaluative Stance in Humanities:

Expectations and Performances
Erika Matruglio
10.1 Introduction
This chapter reports on research into literacy in the senior secondary school
which aims to explore the nature of the literacy requirements for success in the
final years of schooling in New South Wales, Australia. In so doing, it also
explores how an ‘Appliable Linguistics’ can contribute to the understanding
of disciplinary difference as reflected in end of school examinations in this
context and points towards future directions in applying linguistics to the study
of school discourse.
10.1.1 Rationale and background

The nature of literacy teaching and its place in schools continues to be a source
of contention and debate in both academic and public spheres with periodic
arguments about falling literacy standards and debates about appropriate
approaches to literacy development (Green et al., 1997). There are also ques-
tions about whose responsibility it should be to address literacy needs of stu-
dents, especially in the secondary school context. If literacy development in this
context is to be the responsibility of the subject teacher, as advocated in the
literacy across the curriculum movement (Cumming and Wyatt-Smith, 2001),
then there is an urgent need for further research into the language demands
of different subjects at different levels in the curriculum. To date, some very
significant contributions towards this end include studies in the language of
history (Coffin, 1996), English (Rothery, 1994b), science (Veel, 1993), the cre-
ative arts (Rothery, 1994a), maths (Veel, 1999) and geography (Humphrey,
1996). In many cases these contributions have resulted from strong collabora-
tions between Systemic Functional Linguists and teachers. While supporting
greater understanding of the language of schooling, they have also functioned
to continue to build Systemic Functional Linguistics (SFL) Theory, for example
Evaluative Stance in Humanities 169
in the areas of genre and appraisal theory (Christie, 1986; Martin, 1986). To
date, however, much of this work has been carried out in the primary or junior
secondary sectors or in the tertiary context. Less work has been targeted specifi-
cally in the senior secondary school context (Cambourne, 2001; Joy Cumming
and Claire Wyatt-Smith, 2001) although a recent study by Christie and
Derewianka (2008) does make some reference to literacy across the curriculum
in the senior years. There remains, therefore, a significant unresolved area
around literacy in the senior secondary context which merits investigation.
Arising from my work as a literacy educator in a senior high school, I began
to find it increasingly important to investigate literacy in the context of the final
two years of schooling, especially considering the high-stakes nature of the final
Higher School Certificate (HSC) examinations in NSW, which serve as a gate-
keeper for students’ entry into the tertiary sector. In their final year of second-
ary schooling students across the whole range of senior subjects have different
literacy development needs than junior students, as they are learning more
complex written genres and are learning to incorporate and evaluate multiple
voices and opinions within their written texts. Furthermore, anecdotal accounts
from teachers in the humanities in particular seemed to indicate that the
way students were required to write their ‘essays’ and the language they were
expected to use differed substantially across humanities subjects. It would there-
fore benefit both students and teachers to understand more clearly how stu-
dents are expected to write if they are to succeed in their final examinations.
10.1.2 Literature review

One important issue that emerges in the literature in relation to literacy in
schooling concerns the question of how writing is modelled, taught and used in
the classroom. Research into student writing conducted by Eggins, Martin and
Wignell in Sydney in 1986 (Wignell, 1987) found that although writing was
used to assess students’ knowledge in their school subjects, students rarely
engaged in writing in the classroom. Instead, class time consisted mainly of oral
interaction and extended pieces of writing were produced almost exclusively
for assessment purposes. Furthermore, the language and structural features of
the writing tasks themselves were often not explicitly explained to students
when they were given their tasks. Student writing in the classroom, if it occurred,
was mainly limited to short answers to comprehension-type questions or to
copying notes from the textbook with longer, sustained pieces of writing rele-
gated to homework or examination-style purposes (Wignell, 1987).
These findings were mirrored in Applebee’s national study of writing in the
secondary school in the United States (1984). He found that only 3 per cent
of students’ school time (including homework) was spent on writing texts of
paragraph length or longer, and when students were asked to write at length
the writing ‘served merely as a vehicle to test knowledge of specific content’
(Applebee, 1984, p. 2). This is reflected in his analysis of textbooks from across
the curriculum which shows that roughly 90 per cent of the tasks in textbooks
assumed the audience to be the ‘teacher as examiner’ (Applebee, 1984).
It seems, therefore, that students’ writing is valued only as an assessment tool
insofar as it provides an opportunity to communicate ‘subject knowledge’ and
that the form of the writing is unimportant as very little time is dedicated to
teaching students how to write.
Although these studies are now over 20 years old, there does not seem to be
an overall change in the treatment of writing within schools. This is hard to
determine definitively, as there have not been any more recent studies compa-
rable to Applebee’s (Hillocks, 2008). Hillocks (2008) comments briefly on his
own 2002 study into the impacts of state writing tests on the teaching of writing
in the United States, however his study did not include teachers across disci-
plines, but only teachers of English, who could be expected to focus more on
teaching writing than teachers of other subjects. With reference to this study,
Hillocks concludes that
there is an underlying similarity in the way writing is taught during the two
periods. In both periods, teachers and curriculum makers assume that the
knowledge necessary for effective writing is general knowledge of a few prin-
ciples that are applicable to all or most writing: knowing the form that the
piece of writing is to take; brainstorming for ideas before writing; knowing
that effective writing requires more than one draft, and so forth. (Hillocks,
2008, p. 316)
Anecdotal evidence and personal experience of teaching in a senior high

school would also support this conclusion, as the curriculum is so crowded that
many teachers struggle to ‘get through the content’ to prepare their students
for the final end of school examinations and do not feel they have time for the
teaching of literacy, which is often perceived as the domain of the English
subject teacher (or the English as a Second Language (ESL) specialist).
In addition to the problems of very little class time being devoted to teaching
writing, and the use of writing almost exclusively as a tool for assessing the
learning of ‘content’, feedback on students writing also does not appear to give
students much help with regards to improving their writing. When students do
receive feedback on their writing, teacher comments are mostly limited to com-
ments on the accuracy or adequacy of the content of the writing (Langer, 1984)
and when teacher comments do refer to form it is mostly at sentence or word
level without any explanation for how to make the writing more appropriate
(Marshall, 1984). The notion of appropriateness is in itself difficult, as the syl-
labus documents for the subjects included in this study make several references
to the use of appropriate language, but fail to indicate what ‘appropriate’ is in
the context of each particular subject (Matruglio, 2007). While syllabus out-
comes requiring students to ‘communicate a knowledge and understanding of
historical features and issues, using appropriate and well-structured oral and
written forms’ (NSW Board of Studies, 2004b, p. 11) seem to acknowledge
some form of link between disciplinary learning and the literacy skills necessary
to display it, they do little to elucidate this link for the students or the teachers.
Thus, many of the subject-specific requirements for student writing in different
subjects still remain part of the ‘hidden curriculum’ (Christie, 1985) in schools.
Such issues have prompted a strong assertion from researchers in educational
linguistics that the teaching of language and the teaching of content cannot be
separated (Columbi and Schleppegrell, 2002; Gee, 2002; Kress, 2001; Merino
and Hammond, 2002). Further, there is an ever-increasing agreement that
this language teaching must be explicit if students are to achieve the types of
‘advanced literacy’ that is demanded in secondary and post-secondary school-
ing today (Columbi and Schleppegrell, 2002; Scarcella, 2002). Kress states that
literacy ‘is not one thing evenly spread across curriculum areas. It varies with
the kinds of disciplinary practices and forms of knowledge that are at issue in
a school subject’ (Kress, 2001, p. 22).
Another significant issue impacting on the teaching of literacy in schooling
concerns the relationship between popular or common-sense knowledge and
erudite knowledge. According to Muller (2000) this distinction between ordi-
nary and formal knowledge is the basis of modern schooling. The growth of
capitalism and the split between mental and manual activity has led to the
commodification and professionalization of knowledge, which in turn has led
to knowledge specialization, removing knowledge further and further from
everyday meanings and resulting in discourses which are elaborated and highly
technical and which exclude those who have not yet learnt them (Muller, 2000).
It could be argued that schooling is the beginning of the process of initiation
into these specialized groupings, and that through their study of various school
subjects students begin to be apprenticed into the different ways that these
subjects relate to and discuss knowledge. Students need to be aware that each
school subject represents a different perspective on knowledge and that these
differences have become codified to such an extent that they affect the way
one reads and writes during the study of such subjects. Without such ‘insider’
awareness, students will struggle to write in a way that is deemed ‘acceptable’
and may transfer ways of writing which are highly valued in one subject to
another where they are not as highly valued. To enable teachers to adequately
guide and support students towards success, there is a crucial need for greater
explicitness in descriptions of what sort of knowledge is valued in different
subjects and the ways in which such knowledge is expected to be expressed.
Students need to understand what ‘appropriate’ means in the context of each
of their subjects, and whether this notion varies substantially from subject to
subject if they are going to succeed in their HSC and progress to further study
at the tertiary level.
Here too, we need to consider not just what knowledge is to be represented
and how but also what kind of interpersonal stance is considered ‘appropriate’
towards that knowledge. In her work on student written responses to narratives

in the School Certificate examination (at the end of year 10), Macken-Horarik
(2003) found that students who achieved A-range responses were able to
identify, understand and discuss both implicit and explicit evaluation in the
narrative texts. As students progress towards year 12 it would be increasingly
expected that such resources would be used by students in their own writing,
especially in the context of expository essays; whereas Schleppegrell (2004,
p. 102) argues, its effective use ‘indicates that the writer is interpreting and
arguing for a position’.
Recent studies of school genres (Christie and Derewianka, 2008; Martin and
Rose, 2008) have illustrated some of the differences in the literacy demands of
school subjects and this study aims to further contribute to the understanding
of school genres by asking how the literacy demands of subjects vary at the level
of discourse semantics. Previous studies (Matruglio, 2007; 2004) found that
syllabus requirements to ‘analyse’, ‘evaluate’, ‘synthesise information from a
range of sources’ and ‘assess the significance’ (NSW Board of Studies, 2004a)
of ideas, theories or events require sophisticated control of resources for
managing interpersonal stance and that the dual requirements of constructing
an ‘objective text’ which also evaluates different sources in terms of reliability
and accuracy can be problematic for senior-school students, as can be the
requirement to construct an argument integrating multiple viewpoints negoti-
ating the same knowledge space. This study aims to investigate how attitudes
are expressed in high-stakes examination writing across four humanities sub-
jects in the final year of secondary schooling and to see whether different
subjects have different ‘appraisal profiles’ in an attempt to make the evaluative
work that students need to do for each subject more explicit.
As Elkins (2001, p. 145) suggests, ‘[o]ne cannot help but be impressed (and
worried) by the differences in reading and writing across post-compulsory sub-
jects. It must be difficult for students to adapt to the different roles for literacy
in different classes.’ It is this requirement of students to navigate and master
differential literacy practices across subjects that constitutes the major motiva-
tion for the study reported here.
10.2 Methodology
The data for the study include the syllabus documents for the subjects Modern
History (hereafter MH), Ancient History (AH), Society and Culture (SAC) and
Community and Family Studies (CAFS) as well as student texts produced under
examination conditions for the Trial Higher School Certificate Examination
in each of the four subjects. Although four successful and two to three less
successful texts have been collected for each subject, I will focus here on the
analysis of just one text in each subject, which has been graded by the subject
teacher in each case and is considered to be a highly successful text by the

teacher/examiner. The syllabus documents for each subject were first read to
identify specific reference to language and to ascertain whether a particular
orientation to interpersonal language was salient in any part of the syllabus.
The syllabus rationales, a section of the syllabus which occurs close to the begin-
ning of the syllabus documents and which argues for the relevance and impor-
tance of each subject, were then analysed using appraisal theory in order to
determine which interpersonal meanings were prominent in what is essentially
each subject’s introduction to its syllabus. Student texts were then analysed
using appraisal theory to investigate whether the patterns in the use of interper-
sonal resources in the syllabus documents were reflected in the student texts.
This study takes as its theoretical basis a social orientation to language and
literacy pedagogy, beginning from an SFL definition of language as a ‘social
semiotic’. The appraisal system within SFL focuses on the expression of inter-
personal meaning and offers a theoretically sound way to explore the expres-
sion of evaluation in a text at the level of discourse semantics. In particular
I have used the sub-systems of Attitude and Graduation in my analysis of
both syllabus documents and student texts to explore expectations for encod-
ing writer values and stance. A brief explanation of the categories within the
systems of Attitude and Graduation follows. Examples provided below are
taken from the student texts unless otherwise noted.
10.2.1 Appraisal
The Appraisal system (Martin and White, 2005) is the system within SFL used
to express feelings, attitudes and judgements about people or things (Atti-
tude), to grade the intensity of these evaluations (Graduation) and to indi-
cate the source of these evaluations (Engagement). Attitude can be either
expressed as emotion, referred to as Affect:
CAFS 1: Parents in early stages can feel isolated from everyone as they aren’t
able to get out of the house as they are looking after the baby – feel sad,
depressed, lonely
or as an evaluation about human behaviour, referred to as Judgement:
SAC 1: They regard all Aborigines as alcoholic and not trustworthy tenants to
have in a house.
or as an evaluation of things, referred to as Appreciation:
AH 1: the last large & beautiful pyramid is that of Pepi II.

The system of Graduation may then be used to grade these feelings, either by
amount or intensity (Martin and White, 2005) referred to as Force:
AH 2: The lack of discriminating evidence has led to various interpretations,

many scholars suggesting that a build-up of events led to the once mighty and
centralized government’s collapse
or by strengthening or blurring the boundaries between categories, referred to

as Focus:
MH 1: However, many non-governmental responses contributed to the col-

lapse or at least the modification of Apartheid in some ways.
10.3 Analysis
The discussion below begins with an examination of the syllabus documents to

investigate the kinds of interpersonal meanings privileged in the syllabus of
each subject. These results are then compared to an analysis of a highly rated
student text in each subject to determine similarities and differences in the
pattern of interpersonal meanings between subjects and between student text
and syllabus documents.
10.3.1 Syllabus documents

An initial examination of the syllabus documents for MH, AH, SAC and CAFS
showed that syllabus documents oriented more strongly to one of the three
metafunctions of language at different stages. The key competencies section of
the syllabuses seemed to focus on the textual metafunction of language, that is
the organization and structure of language, reflected in key competencies such
as ‘collecting, analysing and organising information’ (NSW Board of Studies,
2004a, p. 13, emphasis mine). The syllabus outcomes seemed to focus on both
the textual and the ideational (that is the subject content) functions of lan-
guage, reflected in outcomes such as ‘communicate a knowledge and under-
standing of historical features and issues, using appropriate and well-structured
oral and written forms’ (NSW Board of Studies, 2004b, p. 11). A reflection of
the interpersonal metafunction of language, however, seemed to be woven
more implicitly through the syllabus documents as a whole with terms such as
debate, justify, evaluate and critically analyse being commonly used in the sylla-
buses and with the word appropriate cropping up repeatedly in phrases such as
‘. . . using appropriate written, oral and graphic forms’ (NSW Board of Studies,
1999, p. 13). Furthermore, outcome H3.3 from both the Modern and Ancient
History Syllabuses, requires students to be able to ‘analyse and evaluate sources
for their usefulness and reliability’ (NSW Board of Studies, 2004a, p. 11; 2004b).
This outcome requires students to comment on the reliability of certain histo-
rians’ work and thus explicitly judge these historians while simultaneously
maintaining an academic or ‘objective’ voice in their writing style. Another
interpersonal element intrinsic to the above syllabus documents is apparent
in outcome H3.4 in both Modern and Ancient History requiring students to
‘explain and evaluate differing perspectives and interpretations of the past’
(NSW Board of Studies, 2004a; 2004b, p. 11). This outcome reveals the impor-
tant part that negotiating multiple viewpoints has in the history syllabuses.
Students are called upon not only to integrate multiple voices into their texts,
but also to evaluate these differing voices and come to conclusions about them.
A partial Appraisal analysis using the systems of Attitude and Graduation
was then carried out on the syllabus rationales of the four subjects in order to
determine how each individual subject presents itself and makes its claims
about its own relevance. As the rationale section argues for the importance and
relevance of the subject and the benefit that studying it will have for the stu-
dent, it would be expected that the values expressed there would be the same
values to be learnt by the students. Analysis of the syllabus rationales can thus
provide an insight into what values students need to reflect in their own writing
and what would be deemed ‘appropriate’ writing by an HSC marker. The results
are presented in Table 10.1.
The analysis revealed that all four subjects appeared to have similar patterns
of Attitude and Graduation with Attitude split evenly between Judgement
and Appreciation and graded exclusively through the use of Force, with
two notable exceptions. The first of these exceptions is the use of resources of
Affect in the CAFS syllabus. While the amount of Affect is small, the result
is still significant as it is the only example of Affect in any of the syllabus
documents. The second exception is that the SAC syllabus rationale contains
significantly more resources of Judgement than any other rationale.
10.3.2 Student texts

The same analysis was then carried out on one student text in each of the four
subjects included in the study. As indicated above, these texts were all produced
Table 10.1 Comparison of Appraisal between syllabus rationales

CAFS SAC MH AH
Affect 2 0 0 0
Judgement 51 67 50 49
Appreciation 47 33 50 51
Force 100 100 100 100

Focus 0 0 0 0
under examination conditions for the Trial Higher School Certificate and were
marked and rated by the teachers as highly successful texts. These texts were
first analysed to discover the patterns of Attitude and Graduation irrespec-
tive of whether these were inscribed or invoked, in order to determine whether
the student texts corresponded with the distribution of interpersonal meanings
expressed in the syllabus rationale documents. A second analysis was then car-
ried out to investigate how much of the Attitude was explicitly stated and
whether this varied between subjects. The question of variation between types
of attitude and the impact of possible variations on the ‘formality’, ‘objectiv-
ity’ or ‘appropriateness’ of the writing was also of interest.
10.3.2.1 Overview of attitude and GRADUATION

When looking at the expression of Attitude and Graduation as a whole, dis-
tinct differences began to emerge between the four subjects. One of the most
striking findings was the large amount of Affect found in the CAFS text, which
accounted for 40 per cent of the total Attitude in the text. As mentioned
above, CAFS was the only subject rationale containing any Affect at all, and of
the students’ texts, CAFS contained the most Affect by quite a large margin.
The SAC student text also mirrored the use of Attitude found in the syllabus
rationale as 72 per cent of the Attitude in the text was Judgement. However,
there were also significant differences between the student texts and the sylla-
bus rationales as is evident from the figures in Table 10.2.
In the syllabus rationales, there was almost even distribution of Attitude
between Judgement and Appreciation in all of the subjects except for SAC,
however this is not the case in the student texts. In the MH text, the majority of
the Attitude was Judgement, and in AH there is more Appreciation than
Judgement or Affect. These findings mirror earlier findings from analysis
of values statements in MH and AH syllabus documents, pointing towards an
apparent focus on capacity and knowledge building for students in the AH
syllabus and a focus on ethics building in the MH syllabus (Matruglio, 2007).
Another difference between the student texts and the syllabus documents is the
use of resources of Focus in the student texts, most notably in MH. For
example:
Table 10.2 Comparison of attitude and graduation in student texts: percentages

CAFS SAC MH AH
Affect 40 2 6 11
Judgement 51 72 71 39
Appreciation 9 26 23 50
Force 95 97 69 85
Focus 5 3 31 15
MH 2: The campaign to Free Mandela, started by the British World Cam-

paign in 1978, effectively became an international solidarity movement
against Apartheid. So while it may be argued that . . .
It appears from this increased use of Focus, which is mostly used in the student
text to blur boundaries rather than strengthen them, that MH is less categorical
than the other subjects.
10.3.2.2 Inscribed ATTITUDE

In order to determine how overt the evaluation in the student texts was, the
Attitude in the texts was then coded according to whether it was explicitly
stated or inscribed, or whether the Attitude was implied, or invoked. Of particu-
lar interest was whether some subjects had greater amounts of inscribed Atti-
tude than others, as this could suggest that overtly evaluative language is more
acceptable in some subjects than others. For reasons of comparison between
subjects, inscribed Attitude is presented in Table 10.3 as a percentage of the
total words for each particular text.
As can be seen from Table 10.3, the differences in the amount of inscribed
Attitude in the four subjects are small. There does not seem to be a significant
difference between the two histories, however there is a fair difference between
the amount of inscribed Attitude in Modern History (3.5 per cent) compared
to SAC (5.1 per cent) and CAFS (5.9 per cent). In addition to the percentage
of the texts devoted to explicit Attitude, the percentage of the total Attitude
which was inscribed was calculated. These results are shown in the fourth col-
umn in Table 10.4.
Table 10.3 Inscribed Attitude in student texts as percentage of whole text
Subject Inscribed Attitude as percentage of total words
Modern History 3.5

Ancient History 4
Society and Culture 5.1
Community and Family Studies 5.9
Table 10.4 Inscribed Attitude in student texts

Subject Inscribed Attitude as Total Attitude as Percentage of Attitude
percentage of total words percentage of total words inscribed
MH 3.5 8.25 42
AH 4 9 44
SAC 5.1 10.66 48
CAFS 5.9 9.8 60
These figures reveal a significant difference between the amount of the

Attitude that is inscribed in MH, AH, SAC, and CAFS. Almost 60 per cent of
the Attitude contained in the CAFS text is explicitly stated, while only 42–44
per centof the Attitude is inscribed in the history texts and a slightly larger
amount (48 per cent) is inscribed in SAC. When considered in conjunction
with the types of attitude most commonly expressed in these texts, CAFS
stands out even more strongly from the other subjects as a subject in which
more congruent and ‘common-sense’ evaluative language is acceptable. This is
exemplified by the text below, which contains a large amount of inscribed
Affect.
CAFS 2: Gay + Lesbian couples with children can cause uncertainty in the
child and they may resent their ‘parents’ or ‘carers’ as they are not like every-
one else – cause tension within family can prevent them from wanting to
enter into social situations as feel embarrassed or scared of being teased –
negative impact on social wellbeing as well as emotional.
10.3.2.3 The interplay between inscribed and invoked ATTITUDE

It was also important to ascertain whether different types of Attitude were
more likely to be inscribed than others and whether this varied significantly
between subjects, as this would give further indications to the type of evaluation
deemed acceptable in each of the four subjects. There is less at stake when
appreciating things than there is when judging people’s behaviour, and aca-
demic texts therefore tend to invoke Judgement more often than inscribing it.
High levels of Affect such as those found in the CAFS text are even rarer in
academic texts in most disciplines, and a congruent or inscribed realization of
this Affect would therefore be highly significant. Tables 10.5–10.7 show the
number of instances where resources of Attitude were used in the texts,
and these figures enable an investigation of how much Affect, Judgement
and Appreciation is either inscribed or invoked in the case of each subject.
Table 10.5 shows the total instances of Attitude in the student texts, Table 10.6
shows instances of inscribed Attitude only and Table 10.7 shows the instances
of invoked Attitude in student texts.
Table 10.5 Total Attitude (inscribed and invoked) in student texts

Subject Affect Judgement Appreciation
MH 5 64 21
AH 9 31 40
SAC 3 76 28
CAFS 35 46 8
A number of interesting points emerge from these data. The first is that when
Appreciation occurs in a text, it is more likely than other categories of Atti-
tude to be inscribed. Out of the 21 instances of Appreciation in MH for exam-
ple, 17 of these were inscribed, with only 4 instances of invoked Appreciation,
and in SAC a similar pattern holds with 22 inscriptions out of the 28 instances
of Appreciation. In opposition to this, Judgement in a text is more likely to be
invoked. In fact, 62 per cent of the Judgement is invoked in SAC (47 instances
out of a total of 76), 67 per cent is invoked in MH (43 instances out of a total of
64) and 74 per cent in AH (23 instances out of a total of 31).
Another interesting point is that much of the Affect in CAFS is actually
inscribed, with only about 17 per cent (6 instances out of a total of 35) of the
Affect being invoked. These differences further accentuate the differences
between CAFS and the other three subjects, with CAFS relying on much more
explicit expression of Attitude than any of the other subjects.
In order to be able to obtain a clearer picture of what was happening across
the different subjects, the results were also calculated as percentage figures of
both total Attitude and inscribed Attitude only. The figures in Tables 10.8–
10.9 below show the distribution of the Attitude in the text according to the
categories of Affect, Judgement and Appreciation.
Table 10.6 Number of instances of inscribed Attitude in student texts

MH 0 21 17
AH 2 8 25
SAC 0 29 22
CAFS 29 21 4
Table 10.7 Number of instances of invoked Attitude in student texts

MH 5 43 4
AH 7 23 15
SAC 3 47 6
CAFS 6 25 4
Table 10.8 Percentages of inscribed Attitude in student texts

MH 0 55 45
AH 6 23 71
SAC 0 57 43
CAFS 54 39 7
Table 10.9 Percentages of total Attitude in student texts

MH 6 71 23
AH 11 39 50
SAC 2 72 26
CAFS 40 52 9
Both the MH and the SAC texts are relatively balanced in their use of
explicit resources of Attitude, with approximately half of the inscribed Atti-
tude expressing Appreciation and half expressing Judgement, while the AH
text contains more evaluative language explicitly indicating Appreciation
(Table 10.8).
Table 10.9, however, shows what percentages of the total Attitude (inscribed
and invoked) in each text is expressed as Affect, Judgement or Appreciation,
and this makes it clear that the patterns in the distribution of Attitude vary
significantly when invoked Attitude is considered together with the inscribed
Attitude. The most common type of Attitude expressed in MH, SAC and
CAFS then becomes Judgement, and the tendency of AH to orient strongly
towards Appreciation is moderated somewhat.
Significantly, most of the invoked Attitude is afforded (Martin and White,
2005) and therefore arises out of the particular orientation of the subjects
themselves. The following examples of invoked Judgement from MH demon-
strate this:
MH 3: As early as the 1950s , the domestic resistance movement had high-

lighted the Apartheid regime in SA . . .
MH 4: . . . resulted in another 86 deaths by police shooting . . .
MH 5: The International Defence and Aid Fund helped to fund lawyers for
the ANC and tried to counter the propaganda coming out of SA.
MH 6: The international response also reached businesses, with the Sullivan
Principles, a set of equal opportunity/right codes for Blacks in SA workplaces
...
MH 7: Ultimately, the fact that de Klerk did not use the military power that
Botha used to suppress Apartheid . . .
and from Society and Culture:
SAC 2: Australia likes to think of itself as an egalitarian [inscribed] society . . .

SAC 3: In a society riddled with social class and prejudice [inscribed] . . .
SAC 4: After the Europeans arrived in Australia, they took their land, and
introduced alcohol.
SAC 5: The majority of Aborigines are in semi-skilled and unskilled labour
which is quite insufficient when they have a family to support.
These instances from the student texts give some insight into what the appro-
priate language and concepts (NSW Board of Studies, 2004b, p. 11) for each
subject may be. Both these subjects are focused around concepts such as Apart-
heid, prejudice and propaganda that are still charged with evaluative meaning in
middle-class white Australian society and so are not completely technicalized as
subject-specific lexis. The texts naturalize a reading position that is highly criti-
cal of such policies and ways of thinking and therefore indicate that the values
considered important in each subject are those in opposition to such policies.
While SAC and MH appear to be oriented more towards the expression of
values and judgements about the behaviour of society and individuals, AH is
more concerned with Appreciation of artefacts, written evidence and histori-
cal empires. For example:
AH 3: The lack of discriminating evidence has led to various interpretations,

many scholars suggesting that a build-up of events led to the once mighty and
centralized government’s collapse.
The interplay of inscribed and invoked Attitude also works differently with
AH. Whereas in both SAC and MH it is the invoked attitude which orients the
text more strongly towards one particular category of Attitude (in their case,
Judgement), in the AH text, it is the inscriptions which orient the text strongly
in one particular direction. Table 10.8 shows that 71 per cent of the inscribed
Attitude for AH is Appreciation, however, when looking at the total Atti-
tude (Table 10.9), only 50 per cent is Appreciation. While Appreciation is
still the most frequently used type of Attitude, the invoked Attitude in AH
serves to create more of a balance in the overall type of Attitude expressed,
whereas in MH and SAC the invoked Attitude serves to orient the text more
strongly towards one particular type of Attitude.
To summarize the above findings, CAFS seems to operate completely differ-
ently from the other subjects. CAFS shows the most consistent distributions of
Attitude across the categories when looking at inscribed Attitude only and
inscribed and invoked Attitude together. While the other subjects use affor-
dances to give them their particular ‘flavour’, in CAFS it is more a case of ‘what
you see is what you get’. CAFS not only contains much more Affect than any
of the other subjects, but 83 per cent of this Affect is inscribed. When consid-
ering inscriptions only, CAFS is oriented towards Affect, with 54 per cent of
the inscribed Attitude in the text expressing Affect, however when consider-
ing both inscribed and invoked Attitude, the text is more oriented towards
Judgement, with 52 per cent of the total Affect expressing Judgement and
40 per cent expressing Affect. Despite these slight changes in orientation, the
distribution of Affect across the three categories varies much less sharply in
CAFS than in the other texts. Furthermore, the CAFS text contains hardly any
Appreciation at all, which is also a significant difference from the other texts.
10.4 Conclusion
Once the expectations and orientations of different subjects are made clear,
teachers and students can work more effectively towards developing ‘appropri-
ate’ literacy for each subject. Although these results are preliminary, arising
from the analysis of only one highly rated text in each subject, interesting dif-
ferences between the subjects of MH, AH, SAC and CAFS are beginning to
emerge. Although an Appraisal analysis of the syllabus rationale section did
not seem to indicate differences between the subjects, the analysis of student
texts has resulted in the emergence of different ‘appraisal profiles’ for the four
subjects included in this study with each displaying a distinctive pattern of Atti-
tude and Graduation usage. While MH and SAC are alike in their orientation
towards the use of resources of Judgement, MH makes greater use of Focus to
grade these resources. On the other hand, AH appears to be oriented towards
Appreciation, while CAFS has almost equal amounts of Affect and Judge-
ment and is much more explicit in its expression of Attitude than the other
subjects. CAFS therefore appears to be more grounded in the ‘everyday’ and
‘commonsense’, a conclusion which would be echoed by many teachers of
other humanities subjects in schools, who often view CAFS as a ‘soft option’ and
as ‘less academic’ or ‘less rigorous’ than other humanities subjects.
As this research is ongoing, it is hoped that analysis of more student texts
in each subject will strengthen these results. It is, however, beginning to become
clear that there are significant differences in the way that students of these
subjects construct stance through their uses of interpersonal language resources.
Making these differences explicit should help elucidate what ‘appropriate and
well-structured oral and written forms’ (NSW Board of Studies, 2004b, p. 11)
may be for these subjects.
References
Applebee, A. N. (1984). Contexts for Learning to Write: Studies of Secondary School
Instruction. Norwood: Ablex.
Cambourne, B. (2001). ‘Literacy and learning in senior schooling: the legal studies
classroom as an instructive case’, in J. Cumming and C. Wyatt-Smith (eds),
Literacy and the Curriculum: Success in Senior Secondary Schooling. Melbourne: ACER,
pp. 72–83.
Christie, F. (1985). ‘Language and schooling’, in S. Tchudi (ed.), Language, School-

ing, and Society. Upper Montclair, NJ: Boynton/Cook, pp. 21–40.
Christie, F. (1986). ‘Writing in the infants grades’, in J. R. Martin and C. Painter
(eds), Writing to Mean: Teaching Genres across the Curriculum. Sydney: ALAA,
pp. 118–35.
Christie, F. and Derewianka, B. (2008). School Discourse: Learning to Write across the
Years of Schooling. London and New York: Continuum.
Coffin, C. (1996). Exploring Literacy in School History (Write it Right Resources for Literacy
and Learning). Sydney: Metropolitan East Disadvantaged Schools Program.
Columbi, M. C. and Schleppegrell, M. (2002). ‘Theory and practice in the develop-
ment of advanced literacy’, in M. Schleppegrell and M. C. Columbi (eds),
Developing Advanced Literacy in First and Second Languages: Meaning with Power.
Mahwah, NJ: Lawrence Erlbaum Associates, pp. 1–19.
Cumming, J. and Wyatt-Smith, C. (2001). Literacy and the Curriculum: Success in Senior
Secondary Schooling. Camberwell, Victoria: ACER Press.
Cumming, J. and Wyatt-Smith, C. (2001). ‘A multi-theoretical and multi-disciplinary
approach to literacy education and curriculum research’, in J. Cumming and
C. Wyatt-Smith (eds), Literacy and the Curriculum: Success in Senior Secondary
Schooling. Melbourne: ACER, pp. 2–10.
Elkins, J. (2001). ‘Some comments on what is read and written’, in J. Cumming
and C. Wyatt-Smith (eds), Literacy and the Curriculum: Success in Senior Secondary
Gee, J. P. (2002). ‘Literacies, identities and discourses’, in M. Schleppegrell and
M. C. Columbi (eds), Developing Advanced Literacy in First and Second Languages:
Meaning with Power. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 159–75.
Green, B., Hodgens, J. and Luke, A. (1997). ‘Debating literacy in Australia: History
lessons and popular f(r) ictions. – Responses to claims that literacy standards are
falling are provided through an examination of Australia’s sociocultural history’,
Australian Journal of Language and Literacy 20 (1), 6–24.
Hillocks, G. (2008). ‘Writing in secondary schools’, in C. Bazerman (ed.), Handbook
of Research on Writing. New York: Lawrence Erlbaum Associates, pp. 311–30.
Humphrey, S. (1996). Exploring Literacy in School Geography. Sydney: Metropolitan
East DSP.
Kress, G. (2001). ‘“You’ve just got to learn how to see”: Curriculum subjects,
young people and schooled engagement with the world’, in J. Cumming and
C. Wyatt-Smith (eds), Literacy and the Curriculum: Success in Senior Secondary
Langer, J. A. (1984). ‘Where problems start: the effects of available information on
responses to school writing tasks’, in A. N. Applebee (ed.), Contexts for Learning to
Write: Studies of Secondary School Instruction. Norwood: Ablex, pp. 135–48.
Macken-Horarik, M. (2003). ‘Appraisal and the special instructiveness of narrative.’
Text, 23 (2), 285–312.
Marshall, J. D. (1984). ‘Schooling and the composing process’, in A. N. Applebee
(ed.), Contexts for Learning to Write: Studies of Secondary School Instruction. Norwood:
Ablex, pp. 103–19.
Martin, J. R. (1986). ‘Intervening in the process of writing development’, in
J. R. Martin and C. Painter (eds), Writing to Mean: Teaching Genres across the Cur-
riculum. Sydney: ALAA, pp. 11–43.
Martin, J. R. and Rose, D. (2008). Genre Relations: Mapping Culture. London: Equinox.
Matruglio, E. (2007). ‘Values and attitudes in ancient and modern history’. Paper
presented at the Australian Functional Linguistics Congress, Wollongong, NSW.
Matruglio, E. (2004). ‘Genre and literacy development in senior ancient history’.
Unpublished Research Project Report for MA in Applied Linguistics, University
of Technology, Sydney.
Merino, B. J., and Hammond, L. (2002). ‘Writing to learn: science in the upper-
elementary bilingual classroom’, in M. Schleppegrell and M. C. Columbi (eds),
Developing Advanced Literacy in First and Second Languages: Meaning with Power.
Mahwah, NJ: Lawrence Erlbaum Associates, pp. 227–43.
Muller, J. (2000). Reclaiming Knowledge: Social Theory, Curriculum and Education Policy.
London: Routledge Falmer.
NSW Board of Studies (1999). Society and Culture Stage 6 Syllabus.
NSW Board of Studies. (2004a). Ancient History Stage 6 Syllabus.
NSW Board of Studies. (2004b). Modern History Stage 6 Syllabus.
Rothery, J. (1994a). Exploring Literacy in School Creative Arts. Sydney: Metropolitan
East DSP.
Rothery, J. (1994b). Exploring Literacy in School English. Sydney: Metropolitan East DSP.
Scarcella, R. (2002). ‘Some key factors affecting english learners’ development of
advanced literacy’, in M. Schleppegrell and M.C. Columbi (eds), Developing
Advanced Literacy in First and Second Languages: Meaning with Power. Mahwah, NJ:
Lawrence Erlbaum Associates, pp. 209–26.
Schleppegrell, M. J. (2004). The Language of Schooling : A Functional Linguistics
Perspective. Mahwah, NJ: Lawrence Erlbaum.
Veel, R. (1993). ‘Exploring literacy in school science’. Unpublished Research Report.
Metropolitan East DSP.
Veel, R. (1999). ‘Language, knowledge and authority in school mathematics’, in
F. Christie (ed.), Pedagogy and the Shaping of Consiousness: Linguistic and Social
Processes. London: Cassell, pp. 185–216.
Wignell, P. (1987). ‘In your own words’, in S. Eggins, J. R. Martin and P. Wignell
(eds) Working Papers in Linguistics, Vol. 5. Department of Linguistics, University of
Sydney.
Chapter 11
The 3×3: Setting Up a Linguistic Toolkit

for Teaching Academic Writing1
Sally Humphrey, J. R. Martin, Shooshi Dreyfus and
Ahmar Mahboob
11.1 Introduction
Over the past 20 years, Systemic Functional Linguistics (SFL), especially its genre
theory, has informed the development of an influential suite of literacy pedago-
gies within and beyond Australia. In higher education, linguists and educators
have drawn on these perspectives to scaffold students’ academic reading (Rose,
2005); to describe the verbal and visual demands of a range of disciplines (Chen
and Foley, 2004; Hood, 2006; 2008; Jones, 2007; Lee, 2009; Ravelli, 2004;
Wignell, 2007); and to support students in gaining control of the genres valued
in these disciplines (Bonanno and Jones, 2007; Ellis, 2004; Coffin and Hewings,
2004; Mahboob, Dreyfus, Humphrey and Martin, this volume; Taylor and Drury,
2007; Woodward-Kron, 2005). The development of SFL-informed academic
literacy pedagogies is hardly surprising, given the extended coverage of SFL
systems and the long history of this ‘appliable’ linguistics in successfully address-
ing tasks and problems related to literacy learning and teaching (Halliday and
Hasan, 2006; Halliday, 2007).
One consequence of the expanding applications of SFL across disciplines
within the tertiary sector is richer descriptions of text/context relationships
and a growing reservoir of resources available to teachers and students for
learning to control the genres and registers of the academic domain. In addi-
tion to ongoing research describing the genres of specific tertiary discipline
areas, educational linguists have given a great deal of attention in recent years
to describing linguistic realizations of the complex status relationships between
students and markers (e.g. Lee, 2009), the relatively implicit persuasiveness of
academic texts (Coffin and Hewings, 2004; Hood, 2006; 2008) and the increas-
ingly multimodal nature of texts used for building knowledge (Jones, 2007;
Stenglin and Iedema, 2001). However, while the expanded semiotic toolbox
made available to educators from this research has enabled valued textual prac-
tices in the academic domain to be made far more explicit and accessible to
students, there is a need to consider the implications of these elaborations
in training the growing numbers of teachers needed in academic literacy

programmes.
Increasingly, SFL specialists are working collaboratively with subject teachers
and literacy tutors who, while expert users of academic language themselves,
have not had specialist training in SFL and have not developed a metalanguage
for making understandings explicit to students. For those working in defined
disciplines, an effective knowledge base can be relatively quickly built by identi-
fying the genres which students most urgently need to access learning and pro-
duce texts in their particular courses. Collaborative projects undertaken by
the Learning Centre and the Biological Sciences department at the University
of Sydney (e.g. Taylor and Drury, 2007), for example, have built a repertoire of
linguistic resources for teachers and students through modelling the stages
of genres such as laboratory reports and focusing on linguistic features which
realize meanings in each stage.
However, while genre has proved to be a very effective starting point for
building knowledge of language in contexts where communicative goals are
shared, it is not always a viable ‘way in’ to apprenticing English as a second lan-
guage (ESL) teachers and academic literacy advisors who are working with stu-
dents across a number of discipline areas. This is because the range of genres
used across disciplines makes it difficult to develop a deep knowledge of how
choices from linguistic systems at each level or stratum and from each meta-
function realize meaning within each genre. As a number of educational lin-
guists working in school-based literacy contexts have found (e.g. Polias and
Dare, 2006; Derewianka, 2001), without such knowledge, a focus on genre can
be limited to superficial understandings of the global organization of particular
text types.
In this chapter we propose a framework for apprenticing literacy tutors that
begins with the more generalized concept of academic domain (Macken-
Horarik, 1996; Humphrey, in press) to characterize the clusterings of social
purposes, social activities, social relationships and semiotic functions privileged
in tertiary learning. Such a perspective allows for the contribution of meanings
from each of SFL’s metafunctions (ideational, interpersonal and textual) to be
adequately accounted for while also encouraging the tutors’ gaze on text to
encompass multiple strata and ranks. Significantly, the framework allows for a
principled selection of resources from the elaborate SFL systems which can be
immediately used by teachers in building shared understandings of academic
literacy practices. It is hoped that this framework can also be used to inform the
development of genre-specific frameworks for analysing, modelling and assess-
ing texts in particular disciplines.
11.2 The 3×3 Framework
The 3×3 is a framework for describing key linguistic resources needed to con-
struct texts across academic disciplines. The instrument was initially developed
The 3×3: Setting Up a Linguistic Toolkit 187
in response to the needs of literacy tutors in the SLATE literacy project (see
Mahboob, Dreyfus, Humphrey and Martin, this volume) to engage analytically
with students’ draft texts and to make explicit the expectations of academic
writing across a range of undergraduate courses at City University Hong Kong.
These tutors, though skilled in producing and recognizing valued academic
language, had varying levels of expertise in diagnosing language use from a
functional perspective and in providing support to students in developing con-
trol of the genres and registers relevant to their disciplines. The 3×3 discussed
here (see Figure 11.1) was thus devised as a principled overview of resources
identified by educational linguists within SFL in their analysis of academic
writing.
The framework is called 3×3 because it forms a 9 square matrix from inter-
secting features of language from each of the three metafunctions of language
(ideational, interpersonal and textual) and features from three strata of
language (social activity, discourse semantics and grammar & expression2). As
is shown in Figure 11.1, the matrix can also be understood in terms of rank, that
is, a kind of constituency hierarchy moving from the level of whole text, to its
phases (e.g. paragraphs) and on to clauses and smaller units of grammar.
11.3 Theoretical Underpinnings of the 3×3
Three important perspectives on language use underpin the 3×3 framework.

These are: that language resources can be characterized according to the type
of learning privileged in particular sites or domains of people’s lives; that mean-
ings of texts are realized through resources at different levels or strata of
language and that any stretch of language functions simultaneously to enact
relationships, represent experience and organize text (interpersonal, ideational
and textual meanings respectively). Each of these perspectives will be further
explicated below.
11.3.1 Domains of literacy and learning: genres and registers

The semiotic space of learning domain was developed initially by Macken-
Horarik (1996) to articulate ‘the nature of commonalities’ shared by all ‘sec-
ondary school subjects’ (p. 235), and the relationship of literacy and learning
at school to other ‘places’ or sites of learning. The model of domains underpin-
ning the 3×3 draws on this theoretical work to explore literacy practices within
and beyond formal educational contexts (e.g. Humphrey, in press). It proposes
four loosely bounded domains of learning within the wider culture, each of
which privileges its own set of genres and registers. These domains and their
communicative purposes are outlined in Figure 11.2.
The academic domain, the site of formal education at the primary, secondary
and particularly tertiary level, is characterized across disciplines by language
188
Figure 11.1 The 3×3: a framework for describing linguistic resources of student writing in the academic domain
generalized
Figure 11.2 Four domains of learning (adapted from Macken-Horarik, 1996)
which is increasingly technical and grammatically metaphorical and which

reconfigures (i.e. generalizes and systematizes) everyday experience (Halliday,
1991, pp. 22–3). Importantly, roles in the academic domain are traditionally
hierarchical, with teachers as socially distant experts, instructing and assessing
students, who are expected to construct themselves in objective terms (Macken-
Horarik, 1996, p. 238).
Genres which are privileged in the academic domain are those which are
written (in either electronic or print form) and which allow students to both
access and demonstrate their learning of the reconfigured experience of par-
ticular disciplines. Persuasive genres which demonstrate a critical orientation
to learning, such as analytical exposition, discussion and challenge are privi-
leged in part because the evidence included in these genres displays knowledge
built from authoritative sources. The focus on displays of learning distinguishes
persuasive genres in the academic domain from those produced in the civic
domain (e.g. hortatory exposition) which are more directly oriented towards
producing social change (Humphrey, in press).
For students moving into the academic domain, it is particularly important
to extend their repertoire of language resources beyond those needed in
the everyday, ‘here and now of you and me’, domain (Macken-Horarik, 1996,
p. 247). While recognizing fuzzy boundaries between domains, as well as varia-
tions in register across disciplines, the 3×3 focuses on linguistic resources which
have been identified as privileged across tertiary disciplines (e.g. Bonanno and
Jones, 2007; Hood, 2008). Where required, more specialized 3×3 matrices can
be developed, to focus on particular disciplines or sub-disciplines, including
those parts of disciplines which are recontextualized in technical or civic
domains.
11.3.2 Realization of meanings at different strata

As is evident from the horizontal axis in Figure 11.1, language resources
included in the 3×3 framework are represented in terms of three levels or strata.
Sydney school models of stratification (e.g. Martin and Rose, 2007), which are
widely used in educational contexts, include three levels of language which are
particularly relevant for higher education (i.e. social activity, discourse and
lexicogrammar).
Stratification can be conceptualized in terms of the way we focus on text.
Beginning with a global perspective, social activity refers to the context in which
texts are produced (i.e. the manifestation of cultural purposes as staged goal-
oriented genres and the manifestation of the particular situation as register –
field, tenor and mode). Subject lecturers often assess student texts at the level
of social activity, with global comments concerning how well the text relates to
the task or question set (e.g. ‘your essay is not critical enough’; ‘you haven’t
answered the question’, ‘you need a conclusion’). Students and lecturers with-
out a shared understanding of how language works to make these global mean-
ings may be frustrated by the abstract nature of comments at this level.
At the level of discourse semantics, the descriptors included in the 3×3
describe resources which students need to control in order to fulfil their lectur-
ers’ expectations of genre and register. Linguistic patterns at the discourse
semantic level are typically those which construe meaning within temporal or
rhetorical sequences called phases and can be loosely mapped onto structures
such as paragraphs. Lexicogrammatical resources include functional and struc-
tural elements within clauses and those which link clauses within sentences.
While mapping strata onto constituent parts such as clauses and phases
suggests a relationship of composition between levels (i.e. whole texts made up
of parts of text3), it is the relationship of realization which is most pertinent to
the development of academic discourse. This relationship is one where choices
in staging whole texts to achieve their social goals are realized or ‘brought to
light’ through choices at discourse semantic level, which are further realized
through lexicogrammatical resources.
11.3.3 A metafunctional perspective

The third fundamental understanding about language informing the 3×3 frame-
work is a metafunctional or trinocular perspective (Halliday and Matthiessen,
2004), represented in Figure 11.1 by the vertical dimension of the matrix.

A metafunctional perspective recognizes that any stretch of language functions
simultaneously to make three meanings: ideational, interpersonal and textual.
Ideational meanings are expressed by clusterings of resources which represent
experience and connect events; these meanings realize the dimension of regis-
ter called field. Interpersonal meanings negotiate relationships and attitudes,
realizing the register variable of tenor. And textual meanings weave ideational
and interpersonal meanings into a coherent whole – thus realizing mode.
A metafunctional perspective can also be used to explain the different ways
that language is structured at different strata (Martin, 2000). Ideational mean-
ings at the level of genre are organized as bounded parts or stages. Interper-
sonal meanings, however, are structured as prosodies, which map across and
typically accumulate meaning across parts of the text. Textual meanings are
organized periodically as waves of Theme and New. An understanding of these
structures is important for recognizing the full role of what are often referred
to as introductory, body and conclusion stages in essay writing.
A metafunctional perspective encourages discipline teachers to expand their
concept of meaning in texts beyond the ‘content’ or ‘field knowledge’ which is
typically seen as their domain. It also encourages literacy teachers to more fully
account for the contribution of field and tenor relationships in unfolding text
and to track the development of resources as students expand their repertoire
of genres. Something that is often underestimated by both discipline and liter-
acy teachers is the important contribution of interpersonal meanings to texts
across disciplines. While resources for negotiating the objective and impersonal
relationships between expert teacher/markers and apprentice students have
often been limited to word- and clause-level realizations such as declarative
mood structures and an absence of emotional words and personal pronouns,
emerging descriptions of systems at the level of genre and discourse semantics
have allowed for meanings which have ‘such enormous power to shape style’
(Bakhtin, 1935/1981, p. 279) to be made far more explicit.
At the level of social activity, recognition of the power of interpersonal mean-
ings has enabled distinctions of genres according to whether they function as
macro-proposals (e.g. requesting or commanding) or macro-propositions
(giving information) (Martin, 1992). For example, in the academic domain,
persuasive genres functioning interpersonally as macro-propositions (i.e. per-
suading somebody that a position is valid) are privileged over those which func-
tion as macro-proposals (i.e. persuading somebody to take action). At the level
of discourse semantics, descriptions of interactional and evaluative systems allow
teachers to make explicit resources for negotiating unequal relationships such
as those between expert teacher/marker and apprentice student and for evalu-
ating field knowledge and research activity in ways appropriate to the discipline.
Such resources enable students to negotiate the tension between demonstrating
their mastery of knowledge to experts and developing a critical stance to that
knowledge, and enable teachers to track the emergence of that critical stance.
An understanding of academic discourse that includes metafunctionally ori-

ented descriptors at all strata is vital for teachers of literacy. The understandings
provided in the 3×3 framework represent a reservoir or bank of resources from
which resources can be selected and combined according to the demands
of particular discipline tasks. From this reservoir, selections need to be made
according to the demands of particular discipline tasks and students can be
supported to build an individual repertoire of resources according to their
learning pathway. It is here that the concept of genre, as an organizer of mean-
ings becomes essential for analysing texts and making their meanings explicit.
In the next section we will demonstrate how the descriptors included in the 3×3
can be used to analyse the resources which combine to form one genre, a bio-
logy report summary.
While space prevents each descriptor to be fully explicated here, the theore-
tical principles underpinning the framework will be demonstrated by tracing
a pathway of realization through strata from the perspective of each metafunc-
tion (i.e. along the horizontal axis of the 3×3 matrix). Teachers within the
SLATE project are encouraged to begin with a ‘top–down’ approach to the
analysis of student texts because problems addressed at higher strata may obvi-
ate the need to address lower level problems. However, a gaze which takes into
account all meanings at each stratum is also useful in some contexts (i.e. work-
ing down the vertical axis of the 3×3 matrix).
11.4 Resources of the 3×3
The text used to illustrate the trinocular and tri-stratal perspective of the 3×3
is typical of final drafts awarded an Excellent grade in a 1st-year biology sum-
mary writing assignment at City University Hong Kong (see Mahboob, Dreyfus,
Humphrey and Martin, this volume).
The biology summary shares with the source text (a published scientific
report) the social purpose of describing features of scientific phenomena and
can thus be classified as a descriptive report (Veel, 1997). However, the shift
towards recontextualizing and evaluating knowledge through summarizing
brings the text into relationship with the agnate genre of review, which is ori-
ented towards interpretation. As is typical in such texts, ideational and textual
meanings are foregrounded at each stratum; however, as will be illustrated
below, interpersonal meanings are also vital in allowing students to shift towards
producing multi-voiced or heteroglossic texts such as essays and literature
reviews (Hood, 2008, p. 1). The annotated text below includes both labelled
stages and the more delicate phases, which group the sequences of meanings
within the central stage. The key topics and subtopics within each phase are
bolded and topic sentences (Hyper-themes) are underlined.
Text 11.1 Report summary from Biology

The habitat and host diversities of symbionts in marine animals
Overview The article, Symbiotic diversity in marine animals: the art of harnessing
chemosynthesis, by Dubilier, Bergin and Lott (2008), describes the
findings of recent research into diverse chemosynthetic habitats and
hosts in marine environments.
Description Habitats of chemosynthetic symbioses in marine animals can be classi-
‘habitat diversity’ fied into deep sea and shallow water locations. In deep seas,
chemosynthetic primary production dominates the biomass at vents
‘deep sea habitats’ and cold seeps. Deep-sea hydrothermal vents are the most produc-
tive habitat. Both vents and seeps have high concentrations of
sulphide and oxygen. Deep-sea vent and seeps are also dependent of
photosynthesis because animals in chemosynthetic habitats need
oxygen to survive.
‘shallow water habitats’ Autotrophic communities have been found to dominate at deep-sea
sites but heterotrophic communities dominate in shallow waters. In
shallow waters, chemosynthetic symbioses occur only occasionally in
hydrothermal vents, cold seeps and whale and sunken wood falls, as
phototrophy supplies enough energy for animal communities. Some
shallow-water sediments are rich in sulphide concentration while
some are extremely low.
‘Host diversity’ In terms of host diversity, the number of chemosynthetic host species
identified is increasing with the help of colonization experiments.
Epibionts, endobionts and siboglinid worms are examples of mor-
phological diversity. They attach in extracellular or intracellular ways
to animals and the attached location is always consistent. Siboglinid
worms are in the trophosome and whale-bones. The size and the
origin of the trophosomal tissue might differ between tube worm
groups. Vent tube worms and bivalves in behavioural and physiologi-
cal strategies obtain reductants and oxidants by different methods.
Vent tube worms obtain sulphide and oxygen by gill-like branchial
plumes, anterior end, posterior end, extended roots and bridge
between the upper oxidized and lower reduced sediment layers.
11.4.1 Ideational meanings

Ideational meanings valued in the academic domain are construed at the level
of context through genres which are staged to build specialized field knowl-
edge relevant to discipline goals. A pathway of realization of ideational mean-
ings through the strata of Text 11.1 is shown in Figure 11.3.
From the perspective of genre, Text 11.1 realizes the social purpose of
classifying and describing in its entity-focused Description stage. Significantly,
however, the recontextualized nature of the summary text is made explicit with
an introductory Overview stage. In terms of register, the specialized field of
Text 11.1 is realized at the level of whole text through discipline-specific,
uncommonsense topics and subtopics (e.g. symbionts, hosts and habitats).
generalized
Figure 11.3 Pathway of realization of ideational meanings in Text 11.1
Figure 11.4 Classifying taxonomy within phase of Text 11.1
At the level of discourse semantics, the specialized field knowledge of Text

11.1 is built up through strings of elements across two phases within the Descrip-
tion stage which ensure that the text stays ‘on topic’. In the ‘habitat diversity’
phase, for example, ‘habitat’ is classified according to ‘types’ and these types
are further broken into sub-types as shown in Figure 11.4.
A large number of resources are implicated in realizing the classifying
function and the technical field at the level of grammar and expression. Of
particular importance are Classifier^Thing relationships in expanded noun
groups. In these examples from Text 11.1, the head noun (Thing) is in bold
and Classifiers are underlined.
1. deep-sea hydrothermal vents

2. chemosynthetic host species
As illustrated in these examples, nouns refer to classes of things in scientific

reports such as Text 11.1, allowing for the development of the extended taxo-
nomies described above (Halliday and Matthiessen, 1999, p. 615). At the word
level, this is achieved through plural forms of nouns without articles in Thing
position, and singular forms of nouns as Classifier. Further classification of fea-
tures according to habitat is achieved through circumstances of location in space
in this report. The use of discipline-specific, formal vocabulary ensures that the
delicate classifications construed in academic discourse are not confused and
also contributes to building high-status tenor relationships as discussed in the

following section.
11.4.2 Interpersonal meanings

Control of interpersonal resources is essential for convincing the expert reader
that the information selected from the source document is valid and significant.
Tenor relationships in the academic domain are often marked by absence of
interpersonal resources, such as personal pronouns; however, the 3×3 allows
for more delicate descriptions of the complex of resources which interact
prosodically across stages of the text and help to give a subjective slant to other-
wise objective meanings (Hood, 2006, p. 40). A pathway of realization of inter-
personal meanings through the strata of Text 11.1 is shown in Figure 11.5.
Taking an interpersonal perspective on genre (Martin, 1992; Iedema, 2004),
Text 11.1 convinces the reader through its construal as a macro-proposition,
giving rather than requesting or demanding information. This genre choice
and the focus on generalized classes of phenomena also help to construe an
authoritative, objective and impersonal tenor relationship.
At the discourse semantic level, authority is realized in the Description stage
through single voiced or monoglossic statements which assume that the infor-
mation is not contested (Martin and White, 2005, p. 100). The accumulation of
data, including lists of examples also contributes to the objectivity and authority
of the text. Significantly, however, in citing the original source document in the
Overview stage, the student writer has shifted overall responsibility for the
validity of the information to acknowledged experts in the field. By establishing
this authority at the beginning of the text (in a higher level Theme – see below),
a prosody is established which spreads the authority across both stages of the
text. The ability to create such prosodies and to control the voices of external
sources not only adds authority but also lays a foundation for an evaluative
stance to be developed.
At the level of grammar, both the monoglossic and attributed statements
are realized through indicative mood choices of Subject ^ Finite (Subject verb
Figure 11.5 Pathway of realization of interpersonal meanings in Text 11.1

agreement in traditional parlance; e.g. Siboglinid worms are . . .) which agree in

number. Control of agreement, like control of discipline-specific referencing
conventions, spelling and punctuation, also contributes to the authority of the
textual voice and thus to convincing the reader.
11.4.3 Textual meanings

Textual resources are used in the academic domain to organize ideational and
interpersonal meanings into coherent written text. One important resource
involved in realizing these meanings at all strata is that of theme, which refers
to the information in first position of a clause. Layers of Theme, which in the
3×3 are glossed as ‘topic sentences’, ‘topic phrases’ and ‘Text previews’, occur
in waves across different levels of text to signal or foreground particular experi-
ential and interpersonal meanings. A pathway of realization of the textual
resource of Theme through the strata of Text 11.1 is shown in Figure 11.6.
From a textual perspective, the first stage of Text 11.1, the Overview stage,
works as a Macro-theme to preview the information across the entire text. In
short report genres such as this, conclusions are not obligatory; however,
in longer texts, concluding stages serve an important textual function in
summarizing or reviewing main points. Theme also functions on the discourse
semantic level as topic sentences or Hyper-themes, which function to predict
information within phases of text. In Text 11.1, there are two layers of
Hyper-theme. The higher layer previews phases of ‘habitat diversity’ and ‘host
diversity’ while the more specific layer previews micro-phases of ‘deep sea’ and
‘shallow water’ habitats. While the term topic sentence is a useful way to gloss
Hyper-theme, it is important to note that more specific phases may be demar-
cated by a circumstance, functioning as Marked Theme. For example:
In deep seas, chemosynthetic primary production dominates the biomass at

vents and cold seeps.
Likewise, while the expression-level resource of paragraph is helpful for pre-

dicting the organization of the text in terms of phases of meaning and their
Figure 11.6 Pathway of realization of textual meanings in Text 11.1

predicting Hyper-themes, a paragraph may consist of more than one micro-

phase and thus contain more than one layer of discourse level Theme.
At sentence and clause level, Theme choices in reports typically sustain the
topic focus of a phase. Subsequent sentence Themes of the ‘deep sea habitat’
phase of Text 11.1, for example, consistently refer to the habitats within this
location (e.g. Deep sea hydrothermal vents . . .; both vents and seeps . . ..; Deep-sea vents
and seeps . . .). The choice of passive voice in academic texts is typically moti-
vated by this textual concern.
Drawing on resources from the three metafunctions across strata allows
literacy teachers to value the multifaceted meanings made by the student writer
of Text 11.1 and to address simultaneously, concerns of content, grammar,
structure, organization and evaluation. As the example above illustrates, genre
selects from and pulls together meanings from all strata and all metafunctions
in the interests of particular disciplinary tasks and thus can inform the selection
of linguistic resources for teaching and assessment.
11.5 Conclusion
The above illustration of the 3×3 demonstrates the concern of the SLATE proj-
ect to present a theoretically principled and coherent framework for literacy
teachers in supporting students to develop a powerful repertoire of linguistic
resources needed to access literacy and learning at tertiary level. The principles
of taking a metafunctional and multi-stratal perspective to language use in the
academic domain form the basis of ongoing work developing genre-specific
frameworks for teaching, assessment and feedback purposes. Further work is
also needed to explore the complex role of interpersonal resources and their
interactions with other meanings to produce convincing texts and to develop
the SFL metalanguage at the level of grammar and expression. With such
resources, teachers in the TESOL field will be well placed to support students
in the development of the literacies they need to access learning across curri-
culum areas.
Notes
1
Funding for this project comes from the City University of Hong Kong and the
University of Sydney.
2
While grammar and expression are modelled as two separate strata in SFL, fea-
tures of expression relevant to academic writing (e.g. paragraphing) have been
included in a combined grammar & expression column.
3
Strictly speaking, in SFL, each stratum affords the possibility of a constituency
hierarchy (i.e. rank); the 3×3 matrix is thus a pragmatically simplified model, in
a sense conflating the theoretically distinct dimensions of stratification and
constituency.
References
Bakhtin, M. M. (1935: Reprint in full 1981). The Dialogic Imagination: Four Essays
(C. Emerson and M. Holquist, trans.). Austin, TX: University of Texas Press.
Bonanno, H. and Jones, J. (2007). The MASUS Procedure: Measuring the Skills of Univer-
sity Students, a Diagnostic Assessment (3rd edn). Sydney: University of Sydney,
Learning Centre.
Chen, Y. and Foley, J. (2004). ‘Problems with the metaphorical reconstrual of
meaning in Chinese EFL learners’ expositions’, in L. Ravelli and R. Lewis (eds),
Analysing Academic Writing: Contextualized Frameworks. Continuum: London, UK,
pp. 190–209.
Coffin, C. and Hewings, A. (2004). ‘IELTS as preparation for tertiary writing:
distinctive interpersonal and textual strategies’, in L. Ravelli and R. Lewis (eds),
Analysing Academic Writing: Contextualized Frameworks. Continuum: London, UK,
pp. 153–71.
Derewianka, B. (2001). ‘Pedagogical grammars: their role in English language
teaching’, in A. Burns and C. Coffin (eds), Analysing English in a Global Context.
London and New York: Routledge, pp. 240–69.
Ellis, R. (2004). ‘Supporting genre-based literacy pedagogy with technology – The
implications for the framing and classification of the pedagogy’, in L. Ravelli and
R. Lewis (eds), Analysing Academic Writing: Contextualized Frameworks. Continuum:
London, UK, pp. 210–32.
Halliday, M. A. K. (1991). ‘The notion of context in language education’, in
T. Le and M. McCausland (eds), Language Education: Interaction and Development.
Proceedings of the International conference held in Ho Chi Minh, Vietnam, 30 March–
1 April. University of Tasmania, Launceston.
Halliday, M. A. K. and Hasan, R. (2006). ‘Retrospective on SFL and literacy’, in
R. Whittiker, M. O’Donnell, and A. McCabe (eds), Language and Literacy: Func-
tional Approaches. London: Continuum, pp. 15–44.
Halliday, M. A. K. (2007). Language and Education, J. J. Webster(ed.) (Collected
Works of M. A. K. Halliday, vol. 9). London: Continuum.
Halliday, M. A. K. and Mathiessen M. I. M. (1999). Construing Experience through
Meaning: A Language-Based Approach to Cognition. London and New York: Cassell.
Halliday, M. A. K. and Mathiessen M. I. M. (2004). An Introduction to Functional
Grammar (3ed edn). London: Arnold.
Hood, S. (2008). ‘Summary writing in academic contexts: implicating meaning in
processes of change’. Linguistics and Education 19 (4), 251–365.
Humphrey, S. (in press). ‘Modelling social affiliation and genre in the civic domain’,
in A. Mahboob and N. Knight (eds), Directions in Appliable Linguistics. London:
Continuum.
Iedema, R. (2004). Discourses of Command. Manuscripts, University of New South
Wales.
Jones, J. (2007). ‘Multiliteracies for academic purposes: multimodality in textbook
and computer-based learning materials in science at university’, in A. McCabe,
M. O’Donnell and R. Whittaker (eds), Advances in Language and Education.
London: Continuum, pp. 103–21.
Lee, S. (2009). ‘Attitude in undergraduate persuasive essays’. Prospect 23 (3), 43–58.

Macken-Horarik, M. (1996). ‘Literacy and learning across the curriculum: towards
a model of register from secondary school teachers’, in R. Hasan and G. Williams
(eds), Literacy in Society. London: Longman, pp. 232–78.
Mahboob, A., Dreyfus, S., Humphrey, S. and Martin, J. R. (this volume). ‘Appliable
linguistics and English language teaching: the Scaffolding Literacy in Adult and
Tertiary Environments (SLATE) project’, in A. Mahboob and N. Knight (eds),
Directions in Appliable Linguistics. London: Continuum.
Martin, J. R. (1992). ‘Macro-proposals: meaning by degree’, in W. C. Mann and
S. A. Thompson (eds), Discourse Descriptions: Diverse Linguistic Analyses of a Fund-
raising Text. Amsterdam/Philadelphia: John Benjamins, pp. 359–95.
Martin, J. R. (2000). ‘Factoring out exchange: types of structure’, in M. Coulthard,
J. Cotterill and F. Rock (eds),Working with Dialogue. Tubingen: Niemeyer, pp. 19–40.
Equinox.
Polias, J. and Dare, B. (2006). ‘Towards a pedagogical grammar’, in R. Whittiker,
M. O’Donnell, and A. McCabe (eds), Language and Literacy: Functional Approaches.
Rose, D. (2005). Learning to Read; Reading to Learn Teacher Resource Book. Sydney.
Ravelli, L. (2004). ‘Signalling the organization of written texts: Hyper-Themes in
management and history essays’, in L. Ravelli and R. Lewis (eds), Analysing Aca-
demic Writing: Contextualized Frameworks. Continuum: London, UK, pp. 105–30.
Stenglin, M. and Iedema, R. (2001). ‘How to analyse visual images: a guide for
TESOL teachers’, in A. Burns and C. Coffin (eds), Analysing English in a Global
Context. London and New York: Routledge, pp. 194–208.
Taylor, C. and Drury, H. (2007). ‘An integrated approach to teaching writing in the
sciences’, in A. Brew and J. Sachs (eds), Transforming a University: The Scholarship
of Teaching and Learning in Practice. Sydney: Sydney University Press, pp. 117–25.
Veel, R. (1997). ‘Learning how to mean – scientifically speaking: apprenticeship
into scientific discourse in the secondary school’, in F. Christie and J. R. Martin
(eds), Genre and Institutions: Social Processes in the Workplace and School. London:
Cassell, pp. 161–95.
Wignell, P. (2007). On the Discourse of Social Science. Darwin, NT. Charles Darwin Uni-
versity Press.
Woodward-Kron, R. (2005). ‘The role of genre and embedded genres in tertiary
students’ writing’. Prospect 20 (3), 24–41.
Chapter 12
Why Are Logical Connectives Sometimes

Detrimental to Coherence?
Alan Jones
12.1 Introduction
Applied linguists have long recognized that logical connectives are frequently
both overused and misused by second language learners of English (e.g. Crewe,
1990; Field and Yip, 1992; Tang and Ng, 1995; Flowerdew, 1998). Predictably,
there is considerable variation according to learners’ L1. Granger and Tyson, in
a study of ESL writers with French as their L1 found ‘no overall overuse’ but
‘strong evidence of overuse and underuse of individual connectors, as well
as semantic, stylistic and syntactic misuse’ (1996, p. 17). Misuse, however it is
defined, is often put down to the way logical connectives are taught in the class-
room and the way they are presented in textbooks: typically with oversimplified
definitions, minimal co-text and context, and often accompanied by examples
from made-up or simplified texts (Crewe, 1990, pp. 317–18; Milton and Tsang,
1993, pp. 231–2; Granger and Tyson, 1996, p. 25). However, this does not
explain the infelicities produced by educated English speakers in their aca-
demic writing. In short, it seems that the effective and skilful use of logical con-
nectives is something that must be acquired. In this chapter I summarize
research findings and juxtapose analyses from linguistics, pragmatics and psy-
chology with the aim of making these appliable to classroom teaching. Insights
gained from these sources can, it is thought, help teachers help novice writers
acquire more expertise in their uses of logical connectives and, indirectly, more
skills in the construction of causal and inferential coherence in texts.
Research in experimental psychology has shown that artificially enhanced
cohesion makes texts easier to comprehend and recall – but only for less skilled
readers and those with relatively little domain knowledge. For more skilled
readers with some domain knowledge, simplification or enhancement of texts
may be detrimental to learning. This led McNamara and his colleagues (1996)
to ask, ‘Are good texts always better?’ The implication is that an absence of explicit
connectives may benefit more practised and more knowledgeable readers.
Why Are Logical Connectives Detrimental? 201
Other research also suggests that a more sparing use of logical connectives has
other benefits for readers, e.g. ease of processing (Haberlandt, 1982; Taboada,
2006) and ease of memorization (Vivanco, 2005).
Allison (2002) has emphasized the interpretive role of the reader, maintain-
ing that ‘[s]ometimes, clear sequencing of ideas and material is best left to
speak for itself, as an underlying relationship between successive points can be
inferred’ (p. 80). Hoey (2001) has more systematically elaborated the crucial
role of reader expectations in explaining the interactive nature of all text. This
dialogical aspect of writing is discussed again below. I will first survey the sorts
of abstract categories that have been set up to account for logico-semantic rela-
tions. I indicate a preference for macro-categories over micro-categories, while
arguing that all the categories proposed are polysemous and hence potentially
problematic. Next I briefly review ‘subjective’ uses of logical connectives. Then
I examine some ineffective uses of logical connectives by language learners, fol-
lowing which I discuss their relative underuse in much English writing. I will
suggest that implicit conjunction (that is, underuse) plays an important role in
acknowledging the reader’s role in the construction of textual coherence. Next
I introduce grammatical metaphor as used in much specialized writing, show-
ing how it can render logical connectives unnecessary and superfluous. Then,
focusing on causal relations, I show that in English such relations tend to be
realized by means of verbs, verbal constructions, nominalizations, abstract
nouns, noun-modifiers and prepositions rather than logical connectives. I also
suggest that the most highly valued styles of writing in English today, under the
influence of new insights from science, emphasize correlation, constraint,
interdependence and emergence rather than cause-and-effect relations. I dis-
cuss the complex mixture of verb types to be found in such writing: intransitive
verbs with formative or inchoative meanings, and transitive verbs with causal,
quasi-causal and peri-causal meanings. Finally, the causal connective therefore is
used to illustrate some uses made of logical connectives by expert writers.
12.2 Some Definitions

Logical connectives are also referred to as conjunctive adjuncts or discourse
adjuncts (Halliday and Hasan, 1976), logical connectors (Celce-Murcia and
Larsen-Freeman, 1999), discourse markers (Fraser, 1999) and linking adver-
bials (Liu, 2008). They have long constituted a knotty problem for grammatical
description, the psychology of reading, text linguistics, pragmatics and dis-
course analysis. We distinguish here between:
a. Coordinating conjunctions (and, or, but);

b. Subordinating conjunctions (before, because, although); and
c. Conjunctive adjuncts (meanwhile, moreover, therefore, however).
Coordinating and subordinating conjunctions are sentence-level structural

items marking logico-semantic relations between clauses. Conjunctive adjuncts,
on the other hand, have a discourse-level, text-building function. In Rhetorical
Structure Theory (RST) they represent a subset of discourse markers, and are
said to mark Discourse Relations (see Taboada, 2006). Other kinds of discourse
markers are words like well, now, okay (Schiffrin, 1987) and evaluative terms like
frankly, unfortunately, etc. In this chapter I continue the older tradition (see
Crewe, 1990; Hyland, 2004; etc.) by referring to conjunctive adjuncts as logical
connectives.
In systemic-functional grammar, interclausal relations made explicit by con-
junctive adjuncts are described as the logico-semantic relations (Halliday, 1994,
pp. 193, 196, etc.). Relations between clauses are interpreted in terms of the
‘logical’ component of the linguistic system (Halliday, 1994, p. 193). However,
Halliday and Hasan (1976) had much earlier emphasized that the connections
between sentences (and clauses) ‘depend in the last resort on the meanings
that sentences express’ (p. 238), suggesting that these connections are to a
considerable extent semantic.
12.3 Under-specification, Polysemy and Ambiguity
Coordinating conjunctions regularly under-specify logico-semantic relations:

they are polysemous, and hence potentially ambiguous. For example, there are
three distinct meanings of but:
(i) adversative, as in they’re pretty, but I can’t grow them (‘on the other hand’);
(ii) replacive, as in don’t drown them, but give them just enough (‘instead’);
(iii) concessive, as in I don’t look after them, but they still grow (‘nevertheless’).
Halliday and Matthiessen (2004, p. 422) point out that ‘[o]nly the last embod-
ies a logical opposition between the two terms . . .’; thus (iii) can be paraphrased
using the logical connective although (e.g. although I don’t look after them, they still
grow). Meanwhile, Schleppegrell (1996) has detailed the polysemy and ambi-
guities of the subordinating conjunction because.
It should not surprise us, then, if conjunctive adjuncts – that is, logical con-
nectives – also under-specify meaning. The term thus for example is often
ambiguous as between a manner and a cause–consequence interpretation. This
may well be the reason it is frequently preferred by experienced writers over the
less ambiguous therefore. There have been numerous attempts to develop a
metalanguage capable of capturing all the semantic and pragmatic nuances of
logical connectives in a systematic and precise way. Four macro-level categories
were advanced in Halliday and Hasan (1976, pp. 238–67): additive, adversative,
causal and temporal. Subsequently, Halliday developed his powerful theory of
grammatical expansion (Halliday, 1994), with its three sub-types:
(a) elaboration, (b) extension and (c) enhancement.1
However, a crucial cross-cutting distinction is the one made between external

(objective) and internal (subjective or epistemic) conjunction (Halliday and
Hasan, 1976; Martin, 1992; Halliday, 1994; Martin and Rose, 2007). This is illus-
trated by so in the examples below:
a. The neighbours left for Melbourne last Friday. So they are not at home.
(external)
b. The lights are out in the neighbours’ apartment: So they must be away.
(internal)
Sweetser (1990) introduced a three-way distinction, contrasting the content

domain, exemplified in a) below, with the epistemic domain, illustrated in b), and
the speech-act domain, illustrated in c). Examples are from Degand and Maat
(2003):
a. John came back because he loves her.

b. John loves her, because he came back.
c. What are you doing tonight, because there’s a good movie on?
Macro-categories like these provide a useful over-arching framework, but in

applying them the need has been felt for finer distinctions or micro-categories.
Halliday and Matthiessen (2004) list 36 terminal nodes for the system of con-
junction, accounting for 87 lexical items. Their system can result in rather
unwieldy labels, such as:
z dismissive-adversative (‘anyway’)
z verifactive clarification (‘actually’)
In this way linguists have often advanced from ‘parsimonious’ to ‘profligate’

theories of coherence relations (Hovy and Maier, 1995). The most comprehen-
sive (and increasingly profligate) theory to date is undoubtedly RST. An early
version of the category set (from Mann and Thompson, 1988) is reproduced
below.
RST employs a plethora of labels for an ever-increasing number of functions.
Here is Taboada on a recent corpus-based study (2006, pp. 578–9):
The analyses followed the traditional RST system, with some modifications:
a larger number of relations, 78 in total, was used, in part because some of
the relations were further subclassified. For instance, Elaboration has the
following subclasses: elaboration-additional, elaboration- general-specific,
elaboration-object-attribute, elaboration-part-whole, elaboration-process-step,
and elaboration-set-member.
elaboration
circumstance volitional cause
solutionhood volitional result
cause cluster non-volitional cause
condition non-volitional result
subject matter otherwise purpose
interpretation
evaluation
restatement
summary
sequence motivation
contrast antithesis
background
enablement
evidence
presentational justify
concession
Figure 12.1 Mann and Thompson’s Discourse Relations (1988)
Knott et al. (2001) argue that Elaboration ‘is so diverse and difficult to define
that it should not be considered a proper relation at all’ (noted by Taboada,
2006, p. 579).2 Behrens (2004) refers to ‘the vague and difficult notion of
Elaboration’ and suggests that it is related to the ‘equally vague notion of
Consequence’ (p. 3). It seems generally clear that, because of the inherent
polysemy/ambiguity of categories proposed to account for logico-semantic
relations, these will always to some extent under-specify and to some extent
betray the exact nature of these relations, which always crucially rely upon the
content of the propositions involved (Halliday and Hasan, 1976; Hovy and
Maier, 1995).
12.4 Subjective Uses of Conjunctions and

Logical Connectives
A distinction between subjectively defined and objectively defined coherence
relations was perhaps first made by Halliday and Hasan (1976) in terms of
‘conjunction’. External conjunction links objective events in terms of temporal
succession, causation, addition or adversativity. Internal conjunction links
described events in terms of their logico-semantic function in inferencing
and argumentation. This distinction is developed in Martin (1992), Halliday
(1994), Martin and Rose (2007) and more especially Thompson (2005). A
similar distinction appeared in Van Dijk (1977), who wrote of semantic vs prag-
matic relations. Sanders et al. (1992; 1993) and Sanders (1997) also use these
terms. Redeker (1990) contrasted ideational and pragmatic, while Knott (2001)
opposes propositional with intentional. A similar dichotomy was proposed in RST,

where Mann and Thompson (1988) distinguished between the subject matter
(i.e. ‘external’) and presentational (i.e. ‘internal’) relations. Sweetser (1990, see
above) distinguished between the epistemic and content domains, however adding
the speech act domain. Jayez and Rossari (2001) discuss the effect of illocutionary
goals. While Pander Maat and Sanders (1997) distinguished volitional from
non-volitional content.
Accounts of the subjective functions of logical connectives in writing have
been influenced by research into spoken discourse. In spoken usage conjunc-
tions and connectives frequently indicate logico-semantic relations, but they
may also function as discourse markers, indexing a speaker’s attitude towards
what is being said (Schiffrin, 1987; Redeker, 1990; Fraser, 1999). Schiffrin
(1987) examined so and because in some detail; and Schleppegrell (1996)
focused on because, while Sanders (2005) notes that so often signals the switch
from a digressive move back to the main topic of conversation. Sweetser (1990)
suggested that some causal connectives become specialized for one domain;
thus English since and French puisque are used overwhelmingly to express
epistemic meanings, and German denn is only used in this domain. Regarding
causal connectives, Braunwald (1997) argued that their competent use involves
integrating language, thought and social understanding. And Painter (1999,
p. 146) ties the development of causal talk in child language development to
the interpersonal negotiation of action. In fact it seems possible that some ‘sub-
jective’ uses of connectives deemed inappropriate in academic writing may
originate in spoken usage (Schleppegrell, 1996; Hinkel, 2003).
It is clear that an understanding of the various types of subjective function
mentioned above is crucial for full comprehension and (ultimately) active con-
trol of logical connectives. The challenge inherent in this for language learners
and their teachers is obviously daunting.
12.5 Inappropriate Uses of Logical Connectives

Crewe’s examination of texts produced by second language learners suggested
that early-stage learners at least were using logical connectives as ‘surface-level
fillers’ (Crewe 1990, p. 321); it seemed they were trying to ‘impose surface logi-
cality on a piece of writing where no deep logicality exists’ (Crewe, 1990, p. 320).
The following example (from Chen’s 2006 study of advanced Taiwanese learn-
ers) shows that Crewe’s explanation is at first glance plausible.
In order to achieve the ultimate control of English, language learners are

encouraged to learn English as early as one can. Thus, there is a tendency that
Taiwan will turn to an ESL context in the near future. However, there must be
a severe impact on learner identity, and learners can never have ultimate
control of English (Belz, 2002). The above researches in different language
contexts all prove that learner’s identity is changing with language contexts;
moreover, learner would suffer from afar more dramatic struggle in a more
mainstream context. Therefore, if ESL context are hastily enacted in Taiwan,
where the mainstream language is still not English, then, it is for certain that
learners will never have ultimate achievement of English; rather, they will suf-
fer from not only a dramatic struggle, but also a severe self-identity problem.
Consequently, further researches are needed on this issue to suggest a better
language context for learners.
However application of finer-grained categories from RST shows that each

apparent ‘misuse’ represents a motivated selection, albeit inappropriate, from
a systematic set of meanings. Thus mislabels as Consequence a relation that
could be more precisely characterized as Elaboration (general > specific).
Other connectives are potentially suitable, but are not supported by other cohe-
sive devices which either point us away from the suggested discourse relation
or contradict it. However, in the third sentence, introduces two recognized
and potentially negative effects of second language learning, that is, its impact
on identity and the difficulty of attaining a useful level of proficiency in it;
however, the cues provided in the first two sentences do not make it clear that
these changes are to be seen as unambiguously beneficial, and those in the
third sentence fail to present the other possible side-effects as unambiguously
negative. Finally, poor word choices exacerbate the problem. By tendency, in
sentence 2, the writer clearly means possibility. The ‘misuse’ of logical connec-
tives is thus intimately linked to numerous other lexical and grammatical
choices made by the writer.
There is a strong tendency among some language learners to place logical
connectives at the beginning of the sentence, typically using them to introduce
New information. But as Green et al. point out (2000), when fronted in
this way such items often usurp the position of a more expected element of
information structure. To illustrate, the following came from by a tertiary-
educated native speaker:
Mentoring has several advantages as a professional training tool. For example,

mentoring can be effectively used to address issues of identity.
Notice how the connective phrase For example uses up the ‘thematic potential’
of the clause it precedes (Halliday, 1994), while adding no information that
cannot be deduced from the content of the clause. This incidentally would
account for the writer’s full repetition of the theme from the preceding sen-
tence (mentoring) instead of beginning his second sentence with the unmarked
theme: It . . . Repetition of this kind is a common concomitant of fronted con-
nectives, presenting Given information as though it were New, and thus detract-
ing from the reader’s focus on any New information contained in the rheme.
12.6 Underuse of Logical Connectives

in Native-speaker Writing
Corpus studies of American and British English reveal that the frequency with
which logical connectives are used varies between registers, genres and indi-
vidual authors (e.g. Biber et al., 1999; Liu, 2008). Heaviest use is made in con-
versation and academic writing, as compared with news reportage and fiction
(it seems so and then are particularly frequent in spoken English). According to
Taboada (2006), frequency of use was low in both task-oriented dialogues and
a large newspaper corpus: between 60 per cent and 70 per cent of relations
went unsignalled. Focusing on causality, Behrens (2004) shows that translations
into English from German or Norwegian exhibit far fewer causal-type connec-
tives than translations from English into German or Norwegian. In English
other devices (grammatical, semantic and pragmatic) are responsible for struc-
turing the events and the text. Tense, for instance, helps to realize temporal
relations, often suggesting causal ones. Prepositions like with, through and by
are also powerful signals of causation (Flowerdew, 1998). But above all it is the
verb that carries much of the burden of semantic coherence and that has fre-
quently also causal meaning. Causal verbs like lead to, produce and prevent play
a key role, but often the lexical verbs themselves realize causative meanings:
. . . two nicked double stranded DNA molecules were tethered between two
beads and were braided by twisting one of the beads by rotating the micropipette
to which it was attached . . . For a text to be coherent . . . it must deploy the
resources of cohesion in ways that are motivated by the register of which it is an
instance.
Hyland (2004) links the relative absence of explicit markers like conjunc-
tions and logical connectives to the nature of a writer’s audience, noting that
‘[w]here texts are intended for a specialist audience, . . . we find fewer textual
devices . . .’ (p. 116). Basing his claims on corpus studies, he states that
. . . writers often rely on their readers’ understanding of lexical relations to

see the implicit cohesion of the text, encoding connections in an underlying
semantic structure which draws on a web of social and cognitive expectations
about what the text is doing (Myers, 1991). Naïve readers lack this domain
knowledge, however, which means that textbook authors are unable to invoke
understandings of craft skills, interpretive practices and rhetorical structur-
ing. Because novices lack experience of the forms which give coherence
and life to those understandings, the author must attempt to construct this
experience, seeking to make the shared meanings of the discipline explicit,
indicating clear lines of thought through surface logicality.
Thus corpus evidence reinforces the view that logical connectives are for
some readers, in some sense, superfluous and detrimental to coherence (cf.
McNamara et al., 1996).
That the reader interacts purposefully with a text and, in some sense, with
its author, is widely accepted in current theories of reading. While mentally
processing a text, there is a ‘dialogue’ between the reader and the writer medi-
ated by the text and its content (see Nystrand, 1989; Hoey, 2001; Hyland, 2004;
Martin and White, 2005). Readers use shared knowledge (content schemata)
and shared expectations about textual structure (formal schemata, genre
knowledge) to reconstruct a writer’s intended meanings (Bamberg, 1983).
Interactive reading of this kind creates a powerful sense of intimacy between
reader and writer, and implicitness can function to increase this impression (cf.
Halliday and Hasan, 1976, p. 298 on unresolved cohesion).
12.7 When Domain Experts Write
Writers who are expert in a given domain generally work hard to find the right
language to communicate complex meanings clearly and (more to the point)
faithfully and precisely (Slotta et al., 1995). They rely to a large extent on the
lexical-cohesive potential of hierarchically organized taxonomies of interdepen-
dent ‘technical’ meanings (Martin, 1993) and the reader’s presumed knowledge
of the domain (Jones, 2005). Lexical and grammatical resources are supple-
mented by a range of other linguistic devices (Achugar and Schleppegrell,
2005): control of Given-New organization and thematic development, parallel
structure, genre knowledge and familiarity with recurrent text types (such as
the problem-solution pattern). Additionally, domain experts become very skilled
in the use of sophisticated rhetorical strategies since, as Hyland (2004, p.20)
puts it, the ‘cachet of acceptance’ is bestowed as the result of complicated social
interactions, involving careful persuasion and the negotiation of knowledge
claims with editors, reviewers and readers.
The writing of domain experts is generally marked by extensive use of
experiential and logical metaphor. One reason for this is that it allows writers to
package complex information into fewer words and to argue effectively about
complex topics. Experiential metaphor is the representation of processes, ordi-
narily realized by verbs, as things and qualities of things, realized by nouns
and adjectives. For instance, The temperature increased . . . becomes An increase in
temperature . . . Logical metaphor represents logico-semantic relations as pro-
cesses, by means of causal verbs (make, cause, produce, lead to, result in, ensure,
etc.); as things, by means of causal nouns (cause, reason, result, etc.); or as pro-
perties of things, realized by adjectives (resulting, etc.); instead of as logical
relations, using conjunctions (like because) or logical connectives (like hence,
therefore, consequently). Also, many abstract transitive verbs combine causation
with another meaning; increase when used transitively means cause to become larger
or more numerous; transitive shatter means cause to be split up into small pieces; and
so on. Using such resources, theories of causation are often embedded deep
within the clause. The sentence below from a current textbook (Seeley et al.,
2000) illustrates this; underneath the original sentence, grammatical meta-
phors are unpacked:
An increase in temperature [. . .] decreases the tendency for oxygen to remain bound to

hemoglobin.
Experiential metaphor Logical metaphor Experiential metaphor
‘The temperature increases’ ‘causes to ‘oxygen tends to remain bound to

decrease’ hemoglobin’
The catenative probability verb tend to is here nominalized as tendency, heading

the second nominal group and modalising it, another hallmark of scientific,
academic and much professional writing.
Domain experts often strain the resources of the language to capture the
kinds of processual and relational meanings that are central to specialized
understandings of the world. They tend to choose more abstract verbs and
more complex verbal groups to represent processes and relations, as well as
more abstract nominals to encode participants, than do novices (Slotta et al.,
1995). Along with grammatical metaphor, they use lexical metaphor, as in key
drivers, bridging the gap, etc. Novices tend to represent physical phenomena as
concrete material processes, using simple verbs of material process (The elec-
trons move . . .; electricity flows . . .); experts use more abstract terms and more
complex constructions to capture the non commonsensical nature of quantum
reality, representing processes with carefully selected verbs (The electrons pro-
pagate through the metal ), complex verbal groups (light energy is absorbed and
transformed, exert external forces to alter the fate of these reactions, and to reveal the rules
that govern the interconversion of mechanical and chemical energy in these reactions),
abstract and often very complex nominal groups (the interconversion of mechani-
cal and chemical energy; the formation of some type of strained, largely unstable, high-
energy transition state), complex postmodifiers (whose accessibility along the reaction
coordinate controls the rate of the reaction) and frequent deverbal action nouns
(interconversion, manifestation, decay, accessibility, compressibility, formation).
[Examples are from Slotta et al., 1995, and Bustamante, 2004.]
12.8 New Understandings, New Ways of Meaning
Over the past six hundred years or so, that is, since Chaucer’s day, the dominant
mode of expression for causal relations in English has changed markedly, evolv-
ing from one in which separate clauses representing separate processes were
linked by a causal conjunction, to one in which two nominalized processes are
linked by a causal-relational verb in a single clause (Halliday, 1993). Meanwhile,

over the last century or so, science and scholarship have been transformed.
There is increasing recognition, among sophisticated writers, not just scientists,
of what Bohm (1957) referred to as ‘the infinite richness of the relationships
that link natural phenomena’ (p. 32). There is a widespread expectation that
explanatory discourse will foreground the inherent complexity of natural,
social and mental phenomena which tend to be framed in terms of correlation,
dynamic interaction, indirect influence and conditioned processes rather than
cause-and-effect (Chi et al., 1994; Chi, 2001; 2005). While causation remains a
central concern in so far as science stills aims to produce explanations that are
operationalizable, that is, that can be tested and put to practical use, the kind of
necessity that was traditionally sought in isolated causes is now viewed as emerg-
ing from complex but coherent patterns of emergence and co-occurrence. This
assumption often stretches the linguistic resources of domain experts when
they have to put their thinking into words.
In general, domain experts rely heavily on four main types of verbs. Causal
verbs fall into two main groups. 1) The first represents enforced causation, in
which the causee is the passive recipient of an effect; we can identify are five
semantic sub-groups, the first two of which are essentially auxiliary verbs that
grammaticalize the causal operation: i) Verbs like make, cause, ensure; ii) Verbs
realizing more indirect types of causality like lead to, bring about, give rise to; iii)
Intrinsically causative verbs, realizing more concrete types of causation add to
these resources; examples are: place, attach, store, drive, sharpen, shatter, heat, etc.;
iv) Verbs and verbal constructions that realize more tenuous types of causal
process, such as play a role in, interfere with, influence, etc.; these can be described
as quasi-causal; v) verbs of creation in which the core meaning is ‘cause to come
into existence’ – examples are create, produce, generate. 2) The second main
group of causal verbs admits of some autonomy, or agentivity, on the part of
a causee. These verbs represent constraint-based causation, also known as
top-down causation, in which conditions are put in place that encourage an
effect rather than determining it. This group includes verbs like allow, permit,
encourage, foster, afford, etc. Such verbs might be referred to as peri-causal.
Two other important verb groups interact with accounts of causal verb
types. 3) The emergent nature of constraint-based processes is often realized
by formative or inchoative verbs, such as form, accumulate, separate, dissolve,
develop, arise, appear, etc. 4) But perhaps the most characteristic feature of
writing today, in all in the most highly valued registers of scientific, commercial,
political and bureaucratic discourse (Halliday and Matthiessen, 2004, pp. 234–5),
is represented by the prevalence of verbs encoding relations of signification
(Martin, 1992, p. 279). This group includes verbs like signify, realize, mean,
symbolize, reflect, suggest, etc. They encode a kind of semiotic relation that is
becoming increasingly important across unrelated disciplines. An increasing
call for the (re-)construal of information and ‘meaning’ reflects a concern with
the negotiation of meaning while adding a layer of semiotic complexity to much

(post-)modern prose. One effect, however, is to enrich textual coherence and
thus help make logical connectives less necessary than ever.
There is also a strong tendency in key registers to represent causal relations
as abstract nouns (cause, reason, and even problem, solution, explanation, design) or
as nominalized processes (attachment, filtration). Halliday (1993, p. 66) saw this
tendency as characteristic of much twentieth-century writing. It puts new
demands on writers who now need to control a wide range of verbs that realize
relational identifying processes, e.g. be and become; equal, add up to, come out as;
render, reformulate, translate; signify, encode, express, expound; indicate, show, demon-
strate suggest; mean, denote, define; represent, refer to, imply; act as, function as, per-
sonify; and many more (see Martin, 1992, pp. 281–3, for system networks). Many
of these verbs have transitive counterparts which are widely used in the same
registers: e.g. define, suggest and demonstrate, as well as identify (as), characterize as
and classify. Such verbs often hover in practice between cognitive and discoursal
meanings, taking on either signification according to context and purpose.
This ambivalence is enhanced by an ability in many cases to take propositional
objects:
The study shows an association between the increased apoptosis and reduced
ureteric bud branching . . . and a 40% decrease in nephron number at
birth.
The study shows that the increased apoptosis and reduced ureteric bud
branching . . . is associated with 40% decrease in nephron number at birth.
In the hard sciences we find not so much a discourse of causation as one

of conditioned, interdependent and emergent processes (Chi et al., 1994; Slotta
et al., 1995; Chi, 2005). The examples below will serve as brief illustrations (they
are from a paper on the physical stresses and strains generated during
chemical and biochemical reactions at the molecular level; Bustamante, 2004,
pp. 3061–2). Processes are in italics, participants are underlined; nouns that
metaphorically represent inchoative processes are in bold, and those that
represent causal processes in bold italics:
a. . . . reactions between molecular species follow pathways that involve the

formation of some type of strained, largely unstable, high-energy transition
state whose accessibility along the reaction coordinate controls the rate of
the reaction . . .
b. . . . stresses and strains develop in molecules as they move along a reaction
coordinate . . .
c. A site-specific nick in the duplex DNA is engineered adjacent to the rotor
attachment point; this design allows covalent bonds in the intact strand to serve
as free swivels, preventing torque from accumulating in the ‘lower’ DNA segment.
Thus, torque stored in the ‘upper’ segment can drive the rotation of a submi-
cron object on a low-friction molecular bearing.
Note the complex and abstract nature of participants in a) and b) – reactions,

pathways, transition states, reaction coordinates, etc. – as well as the relative absence
of semantically transitive verbs. Many participants have encoded processes, via
nominalization, but these are most often inchoative, interactive or emergent
processes. Coherence here is semantic rather than logical, and the represented
processes and relations entail or lead into rather than ‘cause’ one another.
In c), human intervention is described in a similarly circumspect manner
(a nick is engineered), shading into the peri-causal allows, which governs an
‘interaction’.
The nominalization of causal relations is linked to the nominalization of
causing and caused events:
A dramatic rise in mortgage delinquencies and foreclosures in the US was

the trigger for the subprime mortgage crisis.
Here the causal relation is realized by the lexical (non-grammatical) metaphor,

the trigger, which heads a nominal group representing the caused event. In much
scientific writing, however, the causing and caused participants are not events
but either states (often defective, and often inchoative) or agentless, contextu-
ally constrained processes. And these may be loosely rather than causally linked
(i.e. correlation rather than determination). Moreover, instead of ‘causes’ we
tend to find mention of ‘risks’ (or ‘risk factors’), that is, predisposing but non-
determining elements:
Low nephron number, inherited or acquired, has been linked to increased risk
of development of hypertension and renal failure.
To complicate this picture, causal relations (however realized) are routinely

attributed to some subjective source, specific or non-specific – while also other-
wise modalising the represented relations:
It is hypothesized that a decrease in the number of nephrons results in a

reduced filtration surface area, which in turn increases glomerular and sys-
temic blood pressure, possibly through limiting sodium excretion and causing
volume expansion.
This highlights the fact that internal causal relations operate on propositions
representing or implying external causal relations. Inferential relations are
thus overlaid on representations of relations between objects, etc., in the exter-
nal world, much like modality.
Finally, verbs realizing the semiotic relation mentioned above form an impor-
tant feature of much specialized writing today, realizing the interpretive
functions that are characteristic of highly valued registers in the twenty-first
century. They evaluate, interpret and construe a state, process, activity or event,
in the process pre-empting any competing interpretations or construals that
might be offered by a listener/reader:
The invasion of Grenada in 1983 can be seen as a result of the rivalry that
existed between the U.S. and Cuba during the Reagan years.
Physical migration control policies amount to an attempt to equilibrate the
labour market by reducing the excess supply.
The different verb types mentioned above often appear in the same passages,
and frequently in combination with causal verbs or connectives. This is symp-
tomatic of a contemporary type of meta-discourse ‘where the meanings that
are being construed are inherently symbolic ones’ (Halliday and Matthiessen,
2004, p. 234); that is, where competing accounts for events in the world are
interrelated, evaluated and competitively interpreted.
On causality, Bunge (2009) notes that modern science recognizes at least
eight other types of explanation for natural and human phenomena. In the
social sciences, too, a focus on process and contingency is widespread: structure
has become structuration, that is, a process rather than a state. Researchers
on the whole show a preference for more complex models of natural (dis-)
order, and human psychology and sociality, over simpler ones. Multi-factorial
analyses and accounts of contextual constraints are the norm. Naturally, these
ways of understanding influence the way domain experts write, and hence how
educated people write in socially valued registers.
12.9 Therefore in the Hands of Expert Writers
Therefore is a deceptively simply connective that can potentially index a wide

variety of logico-semantic relations: result, reason and purpose, for example
(Halliday, 1994). These are vague labels, and we may expect that (as with Elabo-
ration in RST) they be subject to finer distinctions when examining their actual
usage. Therefore can refer either to internal or external causation, though it is
most often used to express internal causation (or inference). Of course, as
Halliday and Hasan (1976) have noted, this distinction ‘tends to be a little less
clearcut in the context of causal relations than it is in other contexts, probably
because the notion of cause already involves some degree of interpretation by
the speaker’ (p. 257). The fact that low proficiency university students in Japan
have more trouble with causal relations, and specifically, the use of therefore,
than they do with adversative or illustrative ones (marked by however and for
example) may be due to this ambivalence (Ozono and Ito, 2003).
Both the meaning and rhetorical effect of therefore are sensitive to its position
in the sentence. In sentence-initial position it can have an incongruous trium-
phant or remonstrative resonance (try reading some examples aloud). Yet this
is precisely where many language learners (and novice writers more generally)
tend to place it (Flowerdew, 1998; Green et al., 2000). Therefore (like the other
connectives) thematizes textual structure and indeed the reasoning process
itself. One effect of this is to implicate the writer as the one ‘performing’ the
reasoning, and to present the reasoning itself as epideictic display rather than
available for interpretation. Expert writers most often use thematized therefore to
introduce a conclusion, decision or suggestion (Flowerdew, 1998). The strong
presupposition is that the preceding text justifies the conclusion, decision or
strong suggestion. Three examples, with commentaries in brackets, follow.
According to the authors, many Swedish learners are not aware of the fact
that these less formal conjuncts are not always appropriate in academic
writing. Therefore, a teaching suggestion is that teachers should work more
on raising students’ sensitivity to register distinctions in the target language.
[Therefore prefaces a suggestion supposedly warranted by a preceding state-
ment of fact.]
. . ., their framework still forms the base of many recent investigations on
cohesive ties (e.g. Chang, 1997). Therefore, a modified version of Halliday and
Hasan’s framework is used in this study.
[Therefore prefaces an action-decision supposedly warranted by a preceding
generalization.]
These five areas of responsibility are seen as inseparable. Therefore, it is the
duty of management continuously to assess the priorities and discharge its
responsibilities as best it can on the basis of that assessment.
[Therefore prefaces a value-based action-decision supposedly warranted by
a reported generalization.]
Writers often avoid the potentially strident effect of a thematized adjunct by

(a) positioning it just after the grammatical subject or (b) embedding it even
deeper in the clause, after a copula or modal auxiliary verb. Each of these posi-
tions has distinct strategic (i.e. rhetorical) effects, giving special emphasis either
to the preceding subject (a) or the immediately following segments of text (b).
The latter effect is typically achieved when the subject is semantically or rhetori-
cally too slight to carry special emphasis. I illustrate the two possibilities in
examples 1–3 and 4–8 below, respectively; italics indicate themes that have
become marked due to the marked position of therefore; underlining indicates
New information embedded in the theme (normally Given).
1. Candidates’ talk, therefore, must be persuasive so that they come across as

credible, trustworthy and adaptable.
2. The challenge, therefore, is to find a way of working together in a dialogic

mode . . .
3. Failure, therefore, can be interpreted as a clear judgement of unsuitability, . . .
4. A particular incentive to accept the invitation to respond to Halliday’s article, there-
fore, was that it would involve a systematic exploration of the extent of the
similarity that I believed to exist.
Sometimes a logical connective appears after a copula or modal auxiliary,3 often

in a projecting clause whose main function is to predicate a theme. The predi-
cated theme, when followed by a logical connective, is ‘doubly marked’. What
is typically foregrounded in this way is some attitude or assessment of the writer,
which is thereby represented factively. I illustrate this below, underlining the
text that becomes (as it were coincidentally) thematized.
5. It is therefore with regret that I note that ALAA was not formally notified of this
review or the Discussion Paper, . . .
6. It is, therefore, important that ethnic categorisations and equal opportunities
practices should reflect the changes in immigration patterns of ethnic
groups, rather than their visibility in terms of colour.
7. It is therefore problematic for the researcher to impose categories such as
‘Afro-Caribbean’ on their informants.
8. Determination of the elastic properties of DNA is, therefore, essential to understand-
ing DNA: . . .
When therefore is positioned after the grammatical subject (the topical theme, as
in 1–4 above), the reader has to process the subject nominal group before the role
of that sentence in the writer’s line of reasoning has been made explicit. The subject
nominal group in scientific and academic writing is often complex, frequently
containing new information (though the bulk is given); this means the reader
has to work even harder at interpreting the discourse relation. Many of the
above uses of therefore are strategic, that is, rhetorical, and frequently serve to
introduce not the effect of some cause or an evidence-based inference, but to
support a subjective decision based on unspecified grounds, or support a par-
ticular interpretation or evaluation of a situation (as illustrated further above;
see comments in square brackets).
Therefore, when used to link external events, is deeply ambiguous, mixing
cause with reason and even purpose. Therefore is most often used by expert
writers to encode internal causal relations, or inferences. It often prefaces and
‘warrants’ a suggestion, strong recommendation or decision. However, all its
uses are coloured by subjective interpretation and contingent meanings in
which the nature of the process and the participants in question needs to be
taken into account.
12.10 Discussion
This chapter suggests that perceptions of overuse and misuse of logical connec-
tives on the part of language learners can be attributed to a combination of
factors: a) the complexity and ambivalence of the logical relations that connec-
tives realize in much native-speaker writing, often seamlessly melding internal
and external meanings; b) the tendency of language learners and novice writ-
ers generally to position logical connectives sentence-initially, disrupting more
effective patterns of thematic development and information structure; c) a
preference by English-speaking readers for implicit conjunction; d) a prefer-
ence by English speakers generally for embedded causation, as realized by tran-
sitive verbs, abstract nouns, nominalized transitive verbs and causal-relational
verbs, noun-modifiers and prepositions; e) the complex nature of textual coher-
ence, realized in highly valued registers by the delicate interweaving of inchoa-
tive, causal, quasi-causal and peri-causal meanings; and f) the growing complexity
of the meanings and understandings that skilled writers typically try to encode
in contemporary texts under the indirect influence of new insights from
science.
The preference for implicit conjunction (or at least a more sparing use of
logical connectives) in much English writing plays an important role in con-
structing a relationship with the reader that recognizes his or her role in the
construction of textual coherence. Readers of specialized texts with some
grounding in the discourse of the field expect writers to hedge knowledge
claims and inferences in ways that acknowledge an audience of informed and
critical readers. Writers who are expert in specific domains often rely more
heavily on readers’ presumed familiarity with domain knowledge than they do
on overtly performed reasoning processes. Lexicosemantic associations based
on the interdependence of technical meanings, taxonomically organized, assist
such readers to construct coherent textual representations corresponding,
approximately, with the mental representations of the writers.
The task that confronts second language learners who wish to acquire
expertise in writing and reading specialized texts in English is clearly enormous.
But the nature of the task needs to be explicated clearly before any real progress
can be made. Second language pedagogy, especially in the area of academic
writing, needs to adopt a more holistic approach to the representation of causal
relations in texts. Most fundamentally, internal causation needs to be distin-
guished from external causation. Degrees and types of external causation
expressed by abstract verbs need to be thoroughly explored, and peri-causal
relations distinguished from more deterministic ones. Students need to become
aware of the ways in which inchoative verbs, in particular, interact with causative
ones to construct an overall web of logical and semantic coherence that carries
meaning forward without the need for explicit connectives. In short we need to
take analytic insights from the pages of the research literatures, and from a
variety of theoretical paradigms, and apply them to classroom teaching.
Notes
1
To these he later added projection.
2
Kaldor et al. (1998) identified 38 pedagogically rhetorical functions in a student
corpus and found that these constituted a useful tool for teaching purposes.
3
I will not discuss clause final position here, as this is more characteristic of sponta-
neous spoken discourse.
References
Achugar, M. and Schleppegrell, M. J. (2005). ‘Beyond connectors: the construction
of cause in history textbooks’. Linguistics and Education 16 (3), 298–318.
Allison, D. (2002). Approaching English Language Research. Singapore: Singapore
University Press.
Bamberg, B. (1983). ‘What makes a text coherent?’ College Composition and Commu-
nication 34 (4), 417–29.
Behrens, B. (2004). ‘Cohesive ties in translation: a contrastive study of the Norwegian
connective dermed’. Languages in Contrast 5 (1), 3–32.
Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (1999). Longman
Grammar of Spoken and Written English. London: Pearson Education Ltd.
Bohm, D. (1957). Causality and Chance in Modern Physics. London: Routledge.
Braunwald, S. R. (1997). ‘The development of because and so: connecting lan-
guage, thought, and social understanding’, in J. Costermans and M. Fayol (eds),
Processing Interclausal Relationships: Studies in the Production and Comprehension of
Text. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 121–37.
Bunge, M. (2009). Causality and Modern Science (4th revised edn). New Brunswick,
NJ: Transaction Publishers.
Bustamante, C. (2004). ‘Of torques, forces, and protein machines’. Protein Science
13, 3061–5.
Celce-Murcia, M. and Larsen-Freeman, D. (1999: Reprinted in full 1983). The Grammar
Book: An ESL/EFL Teacher’s Course. Boston, MA: Heinle and Heinle Publishing
Company.
Chen, C. W.-Y. (2006). ‘The use of conjunctive adverbials in the academic papers
of advanced Taiwanese EFL learners’. International Journal of Corpus Linguistics 11
(1), 113–30.
Chi, M. T. H. (2001). ‘Why do students fail to understand complex dynamic
concepts?’ Paper presented at AERA 2001 (American Educational Research
Association annual meeting), Seattle, April 1–14.
Chi, M. T. H. (2005). ‘Commonsense conceptions of emergent process: why some
misconceptions are robust’. The Journal of the Learning Sciences, 14 (2), 161–99.
Chi, M. T. H., Slotta, J. D. and de Leeuw, N. (1994). ‘From things to processes: a
theory of conceptual change for learning science concepts’. Learning and Instruc-
tion 4, 27–43.
Crewe, William J. (1990). ‘The illogic of logical connectives’. ELT Journal 44, 316–25.
Degand, L. and Pander Maat, H. (2003). ‘A contrastive study of Dutch and French
causal connectives on the Speaker Involvement Scale’, in A. Verhagen and J. van
de Weijer (eds), Usage Based Approaches to Dutch. Utrecht: LOT, pp. 175–99.
Field, Y. and Yip, L. M. O. (1992). ‘A comparison of internal conjunctive cohesion

in the English essay writing of cantonese speakers and native speakers of
English’. RELC Journal 23, 15–28.
Flowerdew, L. (1998). ‘Integrating “expert” and “interlanguage” computer corpora
findings on causality: Discoveries for teachers and students’. English for Specific
Purposes 17 (4), 329–45.
Fraser, B. (1999). ‘What are discourse markers?’ Journal of Pragmatics 31 (7), 931–52.
Granger, S. and Tyson, S. (1996). ‘Connector usage in the English essay writing of
native and non-native EFL speakers of English’. World Englishes 15 (1), 17–27.
Green, C. F., Christopher, E. R. and Mei, J. L. K. (2000). ‘The incidence and effects
on coherence of marked themes in interlanguage texts: a corpus-based enquiry’.
English for Specific Purposes 19 (22), 99–113.
Haberlandt, K. (1982). ‘Reader expectations in text comprehension’, in J. F. Le Ny
and W. Kintsch (eds), Language and Comprehension. North-Holland: Amsterdam,
pp. 239–49.
Halliday, M. A. K. (1993). ‘On the language of physical science’, in M. A. K. Halliday
and J. R. Martin (eds),Writing Science: Literacy and Discursive Power. London:
Falmer Press, pp. 54–68.
Arnold.
Halliday, M. A. K., and Hasan, R. (1976). Cohesion in English. London: Longman.
Hinkel, E. (2003). ‘Adverbial markers and tone in L1 and L2 students’ writing’.
Journal of Pragmatics 35, 1049–68.
Hoey, M. (2001). Textual Interaction: An Introduction to Written Discourse Analysis.
London: Routledge.
Hovy, E. and Maier, E. (1995). ‘Parsimonious or profligate: how many and which
discourse structure relations?’ Retrieved 28 February 2009 from http://www.isi.
edu/natural-language/people/hovy/papers/93discproc.pdf.
Hyland, K. (2004). Disciplinary Discourses: Social Interactions in Academic Writing
(Michigan Classics Edition). Michigan: Michigan University Press.
Jayez, J. and Rossari, C. (2001). ‘The discourse-level sensitivity of consequence
discourse markers in French’. Cognitive Linguistics 12 (3), 275–90.
Jones, A. (2005). ‘Conceptual development in technical and textbook writing:
A challenge for L1 and L2 student readers’. In Proceedings of the International
Professional Communication Conference. Limerick, CD-ROM.
Kaldor, S., Herriman, M. and Rochecouste, J. (1998). Tertiary Student Writing (Series:
Framing student literacy: Cross-cultural aspects of communication skills in
Australian university settings). N.S.W.: National Centre for English Language
Teaching and Research, Macquarie University, in association with the Graduate
School of Education, University of W A.
Knott, A. (2001). ‘Semantic and pragmatic relations and their intended effects’, in
T. Sanders, J. Schilperoord and W. Spooren (eds), Text Representation: Linguistic
and Psycholinguistic Aspects. Amsterdam: Benjamins, pp. 127–52.
Knott, A., Oberlander, J., O’Donnell, M. and Mellish, C. (2001). ‘Beyond elabora-
tion: the interaction of relations and focus in coherent text’, in T. Sanders,
J. Schilperoord and W. Spooren (eds), Text Representation: Linguistic and Psycholin-

guistic Aspects. Amsterdam and Philadelphia: John Benjamins, pp. 181–96.
Liu, D. (2008). ‘Linking adverbials: an across-register corpus study and its Implica-
tions’. International Journal of Corpus Linguistics 13 (4), 491–518.
Mann, W. C. and Thompson, S. A. (1988). ‘Rhetorical structure theory: toward
a functional theory of text organization’. Text 8 (3), 243–81.
Martin, J. R. (1992). English Text: System and Structure. Amsterdam: Benjamins.
Martin, J. R. (1993). ‘Life as a noun: arresting the universe in science and humani-
ties’, in M. A. K. Halliday and J. R. Martin (eds), Writing Science: Literacy and
Discursive Power. London: The Falmer Press, pp. 221–67.
English. New York: Palgrave.
(2nd edn). London: Continuum.
McNamara, D. S., Kintsch, E., Songer, N. B. and Kintsch, W. (1996). ‘Are good texts
always better? Text coherence, background knowledge, and levels of understand-
ing in learning from text’. Cognition and Instruction 14, 1–43.
Milton, J. and Tsang, E. S. C. (1993). ‘A corpus-based study of logical connectors
in EFL students’ writing: directions for future research’, in R. Perbertom and
E. S.G. Tsang (eds), Lexis in Studies. Hong Kong: Hong Kong University Press,
pp. 215–46.
Myers, G. (1991). ‘Lexical cohesion and specialized knowledge in science and
popular science texts’. Discourse Processes 14 (1), 1–26.
Nystrand, M. (1989). ‘A social interactive model of writing’. Written Communication
6, 66–85.
Ozono, S., and Ito, H. (2003). ‘Logical connectives as catalysts for interactive L2
reading’. System 31 (2), 283–97.
Painter, C. (1999). Learning through Language in Early Childhood. London:
Continuum.
Pander Maat, H., and Sanders, T. (1997). ‘Subjectivity in causal connectives: an
empirical study of language in use’. Cognitive Linguistics 12 (3), 247–73.
Redeker, G. (1990). ‘Ideational and pragmatic markers of discourse structure’.
Journal of Pragmatics 14, 367–81.
Sanders, T. (1997). ‘Semantic and pragmatic sources of coherence: on the categori-
zation of coherence relations in context’. Discourse Processes 24, 119–47.
Sanders ,T. (2005). ‘Coherence, causality and cognitive complexity in discourse’.
Retrieved 14 February 2009 from http://w3.erss.univ-tlse2.fr:8080/index.
jsp?perso=bras&subURL=sem05/proceedings-final/03-Sanders.pdf (24-11-08).
Sanders, T., Spooren, W. and Noordman, L. (1992). ‘Toward a taxonomy of coher-
ence relations’. Discourse Processes 15, 1–35.
Sanders, T., Spooren, W. and Noordman, L. (1993). ‘Coherence relations in a cog-
nitive theory of discourse representation’. Cognitive Linguistics 4 (2), 93–133.
Schiffrin, D. (1987). Discourse Markers. Cambridge University Press, Cambridge.
Schleppegrell, M. J. (1996). ‘Conjunction in spoken English and ESL writing’.
Applied Linguistics 17, 271–85.
Seeley, R. R., Stephens, T. D. and Philip, T. (2000). Anatomy & Physiology (5th edn).
New York: McGraw-Hill.
Slotta, J. D., Chi, M. T. H. and Joram, E. (1995). ‘Assessing students’ misclassifica-

tions of physics concepts: an ontological basis for conceptual change’. Cognition
and Instruction 13, 373–400.
Sweetser, E. (1990). From Etymology to Pragmatics: Metaphorical and cultural Aspects of
Semantic Structure. Cambridge: Cambridge University Press.
Taboada, M. (2006). ‘Discourse markers as signals (or not) of rhetorical relations’.
Journal of Pragmatics 38 (4), 567–92.
Tang, E. and Ng, C. (1995). ‘A Study of the Use of Connectives in ESL Students’
Writing’. Perspectives (City University of Hong Kong), 7, 105–22. Retrieved 12
February 2009 from http://sunzi1.lib.hku.hk/hkjo/view/10/1000093.pdf.
Thompson, G. (2005). ‘But me some buts: a multidimensional view of conjunction’.
Text, 25 (6), 763–91.
Van Dijk, T. A. (1977). Text and Context: Explorations in the Semantics and Pragmatics of
Discourse. London: Longman.
Vivanco, V. (2005). ‘The absence of connectives and the maintenance of coherence
in publicity texts’. Journal of Pragmatics 37, 1233–49.
Chapter 13
Contestable Reality: A Multi-level View on

Modality in Multimodal Pedagogic Context
Yumin Chen
13.1 Introduction
The school of linguistic thought underpinning the present study is Systemic

Functional Linguistics (henceforth SFL), which in recent years has been
referred to as an ‘appliable’, problem-solving theory that ‘can be put to use in
different operational contexts’ (Halliday, 2006, p. 19). This chapter illustrates
the appliability of SFL in the research areas of multimodal discourse analysis
and textbook research.
The link between linguistics, social semiotics and pedagogy forms the nexus
of the present study, with the linguistic concept ‘modality’ as the focal point.
Van Leeuwen (2005, p. 165) suggests that research on modality began with the
‘absolute, context-independent truth of assertions’ in the philosophy of lan-
guage before it moved to linguistics. This perspective on modality was adopted
and further developed by linguists, logicians and social semioticians. In what
follows a brief retrospective on how modality has been traditionally treated in
linguistics and how it is recognized as part of interpersonal meaning will be
provided. The central concern of this chapter is to explore ‘coding orienta-
tion’1 (Kress and Van Leeuwen, 2006), a concept strongly associated with modal-
ity in social semiotics, in multimodal texts for different groups of learners. The
data drawn upon are selected texts for teaching English as a Foreign Language
(henceforth EFL) to primary and secondary students in China.
13.2 Social Semiotic Approach to Modality
There is a long tradition of scholarship in linguistic studies on modality. Modality

concerning the question of ‘what counts as real’ has warranted scholarly atten-
tion among social semioticians. Due to the focus of the current research, rele-
vant linguistic approaches to this concept will be sketched, followed by detailed
explanations of modality in visual communication.
13.2.1 Linguistic modality

Traditionally, the term ‘modality’ is defined as ‘the manner in which the mean-
ing of a clause is qualified so as to reflect the speakers’ judgement of the likeli-
hood of the proposition it expresses being true’ (Quirk et al., 1985, p. 219).
Much of the research on modality seems to focus on modal auxiliary verbs
such as can, could, must, may, might, will, and would, etc. In traditional grammar,
the meanings of modality are divided into two categories, i.e. intrinsic modality
covering permission, obligation and volition, and extrinsic modality that involves
possibility, ability, necessity and prediction (Quirk et al., 1985, pp. 219–21). The
distinction between epistemic modality and deontic modality in pragmatics
(Coates, 1983; Palmer, 1990) again reflects the two different uses of the modals.
It is believed that most of the modals are used in both senses, and modal verbs
can be considered in term of kinds (e.g. epistemic, deontic and dynamic) and
degrees of modality (e.g. possibility and necessity) (Palmer, 1990, pp. 8, 36).
The theoretical rationale underpinning this research owes most to the lin-
guistic theory developed by M. A. K. Halliday and his colleagues. According to
Halliday (1994, pp. 88–9, 356–7), modality refers to the intermediate degrees of
meaning between positive polarity and negative polarity. In English the choices
of ‘yes’ and ‘no’ are expressed in the Finite element of the mood structure or
Mood Adjuncts like no, while various kinds of interminacy in between are con-
veyed through modality. It is further categorized into modalization (e.g. prob-
ability, usuality) and modulation (e.g. obligation, inclination) based on whether
it is concerned with a proposition or proposal. Halliday (1994, pp. 355–63) also
identifies three variables in describing modality: type, orientation and value.
Types of modality include probability, usuality, obligation and inclination, which
represent different ways of construing the semantic space between positive and
negative poles. The second variable ‘orientation’ is used to discuss how each
type of modality is realized. The distinction between subjective modality and
objective modality, combined with explicit and implicit variants, gives rise to
four ways (i.e. explicitly subjective, implicitly subjective, explicitly objective and
implicitly objective) through which each of the aforementioned types can be
realized. The third variable is concerned with the value attached to modality.
Take probability for instance. It can be of high (e.g. certainly), median (e.g. prob-
ably) or low (e.g. possibly) value.
As Hodge and Kress (1988, p. 124) articulate, one of Halliday’s fundamental
contributions to the study of modality lies in his extension of the use of modal-
ity beyond auxiliary verbs to cover all the elements with the same function,
including nouns (e.g. It is a matter of fact that . . .), verbs (e.g. I doubt that . . .),
adjectives (e.g. It is foolish to deny that . . .), and other expressions such as kind
of, hardly and the like. Furthermore, Halliday suggests that modality is part of
the interpersonal component of his functional grammar, rather than placing it
in the ideational component as most logicians contend. To quote Hodge and
Kress,
Contestable Reality 223
Halliday’s theory recognizes that modality is a matter of the relation of the

participants in a verbal interaction, hence squarely in the domain of the
social, and that modal forms are the traces of the activity of speakers acting in
a social context. (Hodge and Kress, 1988, p. 124)
Instead of assuming modality as expressing the actual objective truth that

exists between propositions and the real world, social semioticians examine
it in relation to the semiotic resources employed to express ‘how true or . . . how
real a given representation should be taken’ (Van Leeuwen, 2005, p. 281;
emphasis in the original). The social semiotic approach to modality in visual
communication will be explicated below, before a multi-level view on modality
in texts for different educational contexts is developed.
13.2.2 Modality in visual communication

Social semioticians (e.g. Hodge and Kress, 1988; Hodge and Tripp, 1986; Kress
and Hodge, 1979; Van Leeuwen, 1999; 2005) adapt the term ‘modality’ from
linguistics to the studies of other semiotic phenomena, including visual images
(Hodge and Kress, 1988; Kress and Van Leeuwen, 2006), sound and music (Van
Leeuwen, 1999), and three-dimensional objects (Kress and Van Leeuwen, 2006,
pp. 252–5). In the eyes of social semioticians, modality is used to describe ‘the
stance of participants in the semiotic process towards the state and the status
of the system of classification of the mimetic plane’ (Hodge and Kress, 1988,
p. 122). Therefore, when exploring modality we need to examine the semiotic
resources available to a message producer in negotiating with the message
receiver about how true a given representation should be taken.
Adopting a social semiotic approach to modality, the current research pro-
poses that ‘truth’ is the shared truth produced by members of a group or com-
munity with the same values, and it is closely related to both the ‘affirmation of
solidarity’ and the ‘assertion of power’ (Hodge and Kress, 1988). As Kress and
Van Leeuwen (2006) point out,
. . . visual modality rests on culturally and historically determined standards

of what is real and what is not, and not on the objective correspondence of
the visual image to a reality defined in some ways independently of it. (Kress
and Van Leeuwen, 2006, p. 163)
Modality always involves at least two parties, and the degree of affinity involved
posits the social relationship on a continuum from affirmation (i.e. high affinity/
modality) to negation (i.e. weak or zero affinity/ modality) (Hodge and Kress,
1988, pp. 122–3, 164). To take linguistic phenomena for example, some genres
(e.g. encyclopaedia) are regarded as more factual than others (e.g. fiction).
Nevertheless, what is regarded as credible for a certain social group may not be
reliable in the eyes of another social group (Hodge and Kress, 1988, p. 121;
Kress and Van Leeuwen, 2006, p. 171). Whoever controls modality controls
which version of reality is treated as the valid version, and the accepted repre-
sentation of reality may serve as ‘the basis of judgement and action’ (Hodge
and Kress, 1988, p. 147). In a nutshell, truth can be constantly challenged and
tested in every social or semiotic exchange, and thus modality can be inter-
preted as a social interpersonal concept along these lines.
Modality in visual communication, which is the focus of the current research,
has been theorized in the development of social semiotics. According to Hodge
and Kress (1988, p. 142), the modality value of a visual display is not fixed, but
depends on receiver position and orientation, because receivers can be posi-
tioned in different ways in relation to mimetic content as well as to texts and
producers. Kress and Van Leeuwen (2006, pp. 160–3) further provide a detailed
framework for analysing visual modality, which involves eight modality markers:
colour saturation, colour differentiation, colour modulation, contextualization,
representation, depth, illumination and brightness. Each of the eight aspects is
treated as a continuum running from the maximum potential to the lowest
degree of articulation, with various degrees in between. It should be noted that
modality does not always increase with the amplification of articulation. Along
each of the continuums there is one point representing the highest modality
value. In terms of photographic naturalism, the point of highest value does
not rest upon either extreme of the continuum, but lies in a certain point in
between. For example, in the case of colour there are three dimensions of colou-
ration to be considered: colour saturation, differentiation and modulation. The
cline of colour saturation ranges from the use of maximally saturated colour to
the absence of saturation (i.e. black and white). The naturalistic modality scale
for colour saturation is represented in Figure 13.1.
The second dimension, colour differentiation, refers to the scale running
from a maximally diversified range of colours to monochrome, while the scale
of colour modulation ranges from the representation of all subtle nuances and
full modulation of a given colour (e.g. the uses of different shades of green) to
the use of plain, unmodulated colour. When it comes to the articulation of
details in an image, both background and foreground need to be taken into
consideration. Kress and Van Leeuwen (2006, p. 161) use ‘contextualization’ to
Figure 13.1 Modality scale for colour saturation (Kress and Van Leeuwen, 2006,
p. 160)
describe the continuum running from the most fully articulated and detailed
background to the absence of background. As for the depiction of detail in the
foreground, the term ‘representation’ is used to account for the pictorial details
of represented participants (Kress and Van Leeuwen, 2006, p. 161). The scale
of representation also runs from maximum depiction of detail to maximum
abstraction.
The modality marker ‘depth’ is concerned with perspective, whose scale
ranges from maximally deep perspective to the absence of depth. According to
the naturalistic standard, the highest modality rests on the point along the scale
that represents a central perspective, whereby objects far from the lens tend to
reduce in size and the real-life parallel lines converge at a vanishing point within
or outside the picture frame. Another modality marker that can be scaled is
‘illumination’, which refers to the play of light and shade. The various degrees
on this continuum range from the fullest representation of light and shade to
its absence. Lastly, the modality marker ‘brightness’ is a scale that runs from a
maximum number of degrees of brightness, to only the dark and light versions
of a given colour.
Photographic naturalism, as mentioned above, is the most commonly found,
everyday standard as to what we see with the naked eye. It is generally deter-
mined by the resolution of the standard 35mm photographic technology and
thus termed as ‘naturalistic coding orientation’ (Kress and Van Leeuwen, 2006,
p. 165) in social semiotic theory. However, naturalism is by no means the only
coding orientation. There are various contexts in which different kinds of truth
are favoured. For instance, scientific visuals often adopt an ‘abstract coding
orientation’, which aims to represent the essence or generality of things. In this
context, images with low articulation in colour, perspective, light and shadow
(e.g. diagrams, figures) are regarded as visual displays with high modality value.
In contrast, in advertising and commercial context a ‘sensory coding orienta-
tion’ is frequently employed. Images in this context are designed to arouse
sensations, and thus they are generally entertaining, ‘more than real’ in many
aspects so as to produce the illusion of the benefits that the advertised products
claim to bring. Still another coding orientation is the ‘technological’ one, for
which the reality principle lies in the ‘effectiveness’ or ‘usefulness’ of a given
image as a ‘blue print’ (e.g. maps, architectural layouts).
13.3 A Comparative Study of Coding Orientation

in Texts for Different Pedagogic Contexts
So far we have explained relevant concepts for exploring modality in visual
communication, with special reference to eight modality markers, whose con-
figurations express the modality of a given visual event, and four types of coding
orientation, which provide criteria for modality judgement in different com-
municative contexts. We are now in a position to conduct a comparative study
on selected texts taken from primary, junior and senior secondary EFL text-
books, published by People’s Education Press (henceforth PEP) between 2002
and 2006. The purpose of the study is to account for the similarities and differ-
ences between these texts in terms of the choice of coding orientation, and
shed light on the underlying reasons for different choices in various pedagogic
contexts. The data under examination are three teaching units concerning
animals, which are comparable in the sense that they share the same field, that
is, the same ‘topic or focus of the activity’ in which communicative parties are
engaged, or more broadly speaking, the same ‘institutional focus or social activ-
ity type’ (Eggins, 2004, pp. 9, 103; see also Martin, 1984, p. 23; 1992, p. 536).
13.3.1 Sensory coding orientation in We Love Animals

The unit concerning animals in primary textbooks is Unit 4 We Love Animals
in PEP Primary English Students’ Book I for Year 3 (2003, pp. 38–47). Cartoon is the
dominant image style in primary EFL textbooks. Among 857 visual images in
the whole series, 839 of them are cartoons, accounting for 97.9 per cent. We
take the teaching section Culture (see Figure 13.2) as an example to examine
the coding orientation in primary textbooks.
Four images are involved in the text, each of which depicts one animal con-
sidered typical in a given country (i.e. panda for China, beaver for Canada, eagle
for USA, and kangaroo for Australia), with the contour of the country as the
background. The visual realization in each modality marker varies in the degree
of articulation. In terms of colouration, a high degree of colour saturation as
Figure 13.2 Sensory coding orientation (Excerpted from PEP Primary English
Students’ Book I for Year 3, 2003, p. 47). Reproduced with permission
well as a medium degree of colour differentiation and modulation are employed.

The foregrounded animals are portrayed as emotive, ‘more than real’, in that
their emotions and temperament are added into the personified visual displays
via facial expressions and gestures. In real life normally we cannot tell how an
animal feels without certain expertise, but the personification adopted here
overtly reveals to us something of the animals’ emotions. To be specific, the
panda is represented as outgoing with waving gestures and a wide smile; the
beaver seems happy but somewhat cocky with arms akimbo; the red-eyed eagle
appears a bit fierce with wings spread; and the kangaroo looks funny but caring,
with a superman’s cloak on its shoulder and a baby kangaroo in its pouch. In
addition, the animals are represented as disproportionately big as compared
with the country contours in the monochrome background, which further
enhances the entertaining effect of the represented participants in the fore-
ground. As the pictorial techniques used in these images follow the pleasure
principle whose purpose is mainly for fun, the criterion of truth adopted in the
text can be recognized as sensory coding orientation.
Textbook editors are by no means school-aged children, and normally they
hold a naturalistic coding orientation rather than the ‘exaggerated’ and ‘sensa-
tional’ style as a criterion of truth. Obviously the image producers of the pri-
mary textbook have chosen to adjust their own truth criterion to a sensory
criterion in designing visual displays for primary students. As Hodge and Kress
(1988, pp. 151–3) point out, when there is instability between the ‘world of the
producer (WP)’ and ‘world of the reader (WR)’, effort must be made to resolve
the difference between WP and WR or that difference will be the basis for
antagonism. These efforts may come from the message producer or receiver or
both. In the case of the primary textbook under discussion, the effort to resolve
the difference is made in the WP. The sensory visual style is presumed to attract
the attention and excite the imagination of the primary English learners. In a
nutshell, sensory coding orientation in the primary textbook creates a virtual
world in which the truth criterion is the pleasure principle. This visual arrange-
ment is assumed to be of high modality value to primary school students. In
doing so, textbook editors arguably establish a strong solidarity with the school-
aged textbook readers.
13.3.2 Abstract-sensory coding orientation in Why Do You Like Koalas

There is a strikingly similar instance in the junior secondary textbook exercise
under attention, an exercise entitled ‘Match the animals with the countries’ in
Unit 3 Why Do You Like Koalas in Go for it Students’ Book II for Year 7 (2005, p. 15;
see Figure 13.3). Both the primary school text and the junior secondary school
text include images depicting animals and the corresponding countries. In
the junior secondary textbook, however, the images of animals and countries
are divided into two sets. The lower set represents three animals (lion, panda
Figure 13.3 Abstract-sensory coding orientation (Excerpted from Go for it

Students’ Book II for Year 7, 2005, p. 15). Reproduced with permission
and koala), while the upper set depicts three countries with names as captions
(China, Australia and South Africa). Here junior secondary students are no
longer provided with the correct match between animals and countries, but
required to do the matching by themselves. The repeated use of similar teaching
materials complies with the requirement of a ‘high reappearance rate for lan-
guage teaching materials’, as stipulated in the Curriculum Standards for English.
Furthermore, the transition from simple to complex instructional designs
reflects the principle of ‘proceeding in an orderly way and advancing step by
step’ (Ministry of Education of the People’s Republic of China, 2001, p. 48).
In terms of visual treatments, reduced articulation is employed in representa-
tion, contextualization, illumination and depth. Simple line drawing is adopted
to capture the essence of the animals, i.e. what makes them identifiable as
lion, panda and koala. For example, the long heavy mane around the top and
sides of the neck, which is the symbol of a male lion, allows the viewer to easily
recognize this animal. The shapes of the three animals and the textures of their
fur are brought down to the essential quality. In addition, these animals are
pictured in their actual living environment. For instance, the koala is depicted
as clinging to a eucalyptus tree. Nonetheless, the background has a low level of
articulation and merely represents the general scenario of the environment,
leaving out all the details such as the texture of tree bark which would otherwise
be indicated in naturalistic photographs. The abstract style is also reflected
in the low articulation of illumination and depth. There is no play of light
and shade or perspectival foreshortening in the images. The maps of the three
countries, in which most of the modality markers are greatly reduced in articula-
tion, further enhance the sense of abstractness. Although colour differentiation
can be found, the purpose of using various colours in the maps is to distinguish
the countries under attention from the adjacent countries and oceans. Under-
lying the abstract modality is the truth criterion which holds that the more an
image arrests the essential quality or the general truth, the higher the modality
value will be.
There is still some tension between the abstract modality and the emotive
orientation in the images. Unlike the typical abstract images such as scientific
diagrams, the three animal images draw on the pleasure principle (Kress and
Van Leeuwen, 2006) to some degree. A high degree of colour saturation is
applied, and subtle nuances of a given colour are identified. Take the green
colour for instance. The variations of greenness include the bright green for
twigs and new leaves, the brownish green for meadow and dark green for leaves
on the eucalyptus tree. The use of colour here is not merely for distinguishing
different components, nor does it aim at faithfully reproducing the real colours
in natural environment. The colouration here is ‘a source of pleasure and affec-
tive meanings’ (Kress and Van Leeuwen, 2006, p. 165), whose purpose is to
make the images pleasant to the eyes and to produce an affective effect that
helps attract attention and maintain interest.
In short, the coding orientation in the junior secondary textbook is both
abstract and sensory. The adoption of the abstract-sensory coding orientation
in junior secondary educational context has to do with the role of abstract
modality in education and the pleasure principle that continues into adolescent
education. Abstract modality goes beyond the surface to capture the underly-
ing essence (cf. realism in art and literature as discussed by Van Leeuwen, 2005,
p. 168). It is used among sociocultural elites (Kress and Van Leeuwen, 2006,
p. 165), and an understanding of producing and reading images with an
abstract coding orientation is part of the education of ‘cultivated’ people. An
abstract coding orientation is more evident in textbooks on natural science
where diagrams and figures are widely applied. As we can see, the tendency
towards a ‘conceptual’, ‘abstract’ coding orientation gains momentum in the
junior secondary context, while a sensory style is retained. The tension between
the educational purpose and pleasure principle accounts for the coding orien-
tation in the junior secondary textbook, which is ‘abstract’ and ‘conceptual’
with respect to representation and contextualization, but pleasure-eliciting in
terms of colouration.
13.3.3 Naturalistic coding orientation in Wildlife Protection

The teaching unit concerning animals in senior secondary textbooks deals with
a more serious topic, ‘Wildlife Protection’, which is the fourth unit in New Senior
English for China Student’s Book 2 (2004, pp. 25–32) for Year 10 students (see
Figure 13.4). We concentrate our discussion on the ‘Warming Up’ section,
which describes three animals (i.e. panda, milu deer and South China tiger) that
Figure 13.4 Naturalistic coding orientation (Excerpted from New Senior English
for China Student’s Book 2, 2004, p. 25). Reproduced with permission.
were once endangered in China due to lack of food supply or over-hunding.

The corresponding verbal texts also introduce some measures taken to protect
these species from extinction and the resulting achievements.
Images of the three animals are presented along with the information on
the causes of the near extinction, the animals’ habitats, as well as the contrast
between the previous worsening condition and the current improved situation.
These images are photographs with naturalistic coding orientation, represent-
ing the real-life situation of the wild animals in natural environment, for
example, two pandas eating in a bamboo grove, a number of milu deer drink-
ing by a river and a South China tiger running on meadow. The relevant truth
criterion here is the naturalistic standard.
As noted above, the naturalistic coding orientation is the shared, dominant
truth criterion for the general public. Images with naturalistic coding orienta-
tion represent what is photographed the way we see it with the naked eye.
When it comes to public concerns such as pressing environmental problems,
the adoption of cartoons might downplay the seriousness of the issue, while the
application of schematic drawings that represent animals as symbols or speci-
mens may result in distancing the viewer from the endangered species urgently
in need of help. Naturalistic photographs, on the other hand, represent the
endangered wildlife animals in the way that is in accordance with the cultural
norm. They faithfully reproduce the living conditions of these endangered
animals, hence creating the sense of being personally on the scene. In other
words, these photographs follow the coding orientation that all members of a
given culture share, regardless of their educational levels or special training
(Kress and Van Leeuwen, 2006, pp. 165–6). The naturalistic coding orientation
here positions the viewer as involved in the environmental protection campaign
in which all members in the shared culture are supposed to participate.
It may be inferred from the above multi-level, comparative analysis that the
pictorial techniques employed in the primary textbook follow the pleasure
principle and hence sensory coding orientation is adopted. The sense of
abstractness mounts upwardly as students move through the school years from
primary to junior secondary education. On the one hand, the vibrant, exciting
colouration still acts as a source of pleasure; on the other hand, the abstract
representation and contextualization capture the underlying essence of what is
represented. The senior secondary textbook adopts a naturalistic coding orien-
tation, which positions the viewers as sharing the dominant cultural norms and
involved in the ongoing environmental protection campaign.
13.4 Conclusion
In this chapter we have considered the ‘interdependence’ (Hodge and Kress,

1988, p. 161) between modality and social relations from a multi-level perspec-
tive, through a comparative study of coding orientation in multimodal texts on
a similar topic for different pedagogic contexts. It was found that what counts
as real is socially defined and specific to a given communicative context. The
way in which multimodal resources are deployed to convey modality in peda-
gogic setting is conditioned by as well as constitutive of the solidarity between
textbook editors and different groups of learners.
Up to this point, we have been concerned with an in-depth analysis of a rela-
tively small number of texts. If we look at the panorama of the whole textbook,
it would be necessary to consider whether the choice of a certain coding orien-
tation is appropriate, and what implications it has for the given pedagogic con-
text. Table 13.1 demonstrates the numbers and proportion of various visual
styles (cartoon, portrait and photograph) in the primary, junior and senior sec-
ondary EFL textbooks under examination. It can be inferred that sensory cod-
ing orientation has a pervasive influence on the textbooks for children and
teens (cartoons accounting for 97.9 per cent and 89.6 per cent in primary and
junior secondary textbooks). The image producers seem to surrender their
normal naturalistic criterion for truth to cater for what is assumed to be real to
children. Whether this practice is justifiable or excessive demands scholarly
attention in ensuing research. Among possible candidates, a sociological stance
towards pedagogic discourse and nature of knowledge (e.g. Bernstein, 1990;
2000; Christie and Martin, 2007) may provide a remedy for the conceivable risk
of ‘childist ideology’ (Martin, 1985).
Another dimension that may be worth exploring in further studies is the way
that members of a social group that normally share the same coding orienta-
tion do not have access to just one style of visual display. For instance, it is found
in our data that there can be various choices of coding orientation in different
Table 13.1 The distribution of visual styles in EFL textbooks for different levels of
education
cartoon portrait photograph
primary textbooks 839 (97.9%) 16 (1.9%) 2 (0.2%)

junior secondary textbooks 403 (89.6%) 5 (1.1%) 42 (9.3%)
senior secondary textbooks 27 (29.7%) 15 (16.5%) 49 (53.8%)
constituent genres within the same teaching unit (technically a macrogenre,

see Martin, 1994; Martin and Rose, 2008). To account for this complex phe-
nomenon, we may need to consider the influence of the contextual variable
field (Eggins, 2004; Halliday, 1994; Martin, 1984; 1992) in order to investigate
how certain modality markers may be adjusted to some extent due to the varia-
tions in the topic or focus of the social activity.
The social definition of truth, as we have been arguing in this chapter, is
constructed through the affinity and antagonism between social groups, and it
reflects the shared value and alignment among members of a given community.
Social relations are ‘constructed and mediated through semiotic activity’
(Hodge and Kress, 1988, p. 161). By making clear the affordances (including
possibilities and limitations) of multiple semiotic systems from a social semiotic
perspective, we may hopefully take a step towards a comprehensive and critical
understanding of the multimodal features in the present-day pedagogic
context.
Note
1
As Kress and Van Leeuwen (2006) explain, the term ‘coding orientation’ comes
from Bernstein’s (1981) concept and refers to sets of principles which inform the
way in which texts are coded by specific social groups or within specific institu-
tional contexts.
References
Bernstein, B. (1981). ‘Codes, modalities and the process of cultural reproduction:
a model’. Language and Society 10, 327–63.
Bernstein, B. (1990). Class, Codes and Control, Volume 4: The Structuring of Pedagogical
Discourse. London: Routledge and Kegan Paul.
(revised edn). Lanham: Rowman and Littlefield.
Christie, F. and Martin, J. R. (eds) (2007). Language, Knowledge and Pedagogy:
Functional Linguistic and Sociological Perspectives. London: Continuum.
Coates, J. (1983). The Semantics of the Modal Auxiliaries. London: Croom Helm.
Eggins, S. (2004). An Introduction to Systemic Functional Linguistics (2nd edn).
London: Continuum.
Arnold.
Halliday, M. A. K. (2006). ‘Some theoretical considerations underlying the teaching
of English in China’. The Journal of English Studies (Sichuan International Studies
University) 4, 7–20.
Hodge, R. and Kress, G. (1988). Social Semiotics. Cambridge: Polity Press.
Hodge, R. and Tripp, D. (1986). Children and Television. Cambridge: Polity Press.
Kress, G. and Hodge, R. (1979). Language as Ideology. London: Routledge and Kegan
Paul.
Kress, G. and Van Leeuwen, T. (2006). Reading Images: The Grammar of Visual Design.
London: Routledge.
Martin, J. R. (1984). ‘Language, register and genre’, in Christie, F. (ed.), Language
Studies: Children Writing Course Reader. Geelong, Victoria: Deakin University Press,
pp. 21–9.
Martin, J. R. (1985). Factual Writing: Exploring and Challenging Social Reality. Geelong,
Victoria: Deakin University Press.
Martin, J. R. (1994). ‘Modelling big texts: a systemic functional approach to multi-
genericity’. Network 21, 29–52.
Equinox.
Ministry of Education of the People’s Republic of China (2001), Curriculum
Standards for English. Beijing: Beijing Normal University Press.
Palmer, F. R. (1990). Modality and the English Modals (2nd edn). London:
Longman.
Quirk, R., Greenbaum, S., Leech,G. and Svartvik, J. (1985). A Comprehensive
Grammar of the English Language. London: Longman.
Van Leeuwen, T. (2005). Introducing Social Semiotics. London: Routledge.
Textbooks cited:
Go for it Students’ Book II for Year 7 (2005). Beijing: People’s Education Press.
New Senior English for China Student’s Book 2 (2004). Beijing: People’s Education
Press.
PEP Primary English Students’ Book I for Year 3 (2003). Beijing: People’s Education
Press.
Chapter 14
Making Many Meanings in Popular Rap Music
David Caldwell
14.1 Introduction
In 1995, musicologist Robert Walser put forward the following challenge to

scholars attempting to describe the rhetorical properties of rap music:
If we are to understand why rap is so important to millions of people and why

it stands at the center of debates over culture and affects struggles over
resources, analysing lyrics is not enough – any more than is formalist musical
analysis, or sociolinguistic analysis . . . We need to begin to hear not only what
these rappers are saying, but also what these musicians are composing – how
they are using rhythm, rhyme, and rhetoric to enact survival and celebration,
clamour and community. (Walser, 1995, p. 212)
Since then, a substantial body of research has investigated rap music and
the hip-hop culture from which it derives (e.g. Krims, 2000; Keyes, 2002;
Maxwell, 2003; Richardson, 2006; Pennycook, 2007). With the exception of some
fine ethnomusicological studies (e.g. Keyes, 2002; Maxwell, 2003), and Krims’
(2000) musical poetics, most research has focused on the language practices of
rap music and its connections with African-American ideologies, hip-hop cul-
ture, popular music, literacy education and global language practices. As such,
this research is exclusively interested in the musical practices of rap musicians;
what they are composing, and what their compositions ‘mean’.
In recent times, the Systemic Functional Linguistics (SFL)(e.g. Halliday,
1978) model of language has been successfully applied to modes of meaning
other than language, including architecture (e.g. O’Toole, 1994), visual images
(e.g. Kress and Van Leeuwen, 1996), movement (e.g. Martinec, 2000) and
music (e.g. Van Leeuwen, 1999; Caldwell, 2010). SFL is now widely considered
a theory of social semiotics, as well as linguistics. A key feature of SFL and social
semiotics is the description of language and other modes of meaning in the
form of system networks:
Making Many Meanings in Popular Rap Music 235
Systemic theory gets its name from the fact that the grammar of a language is
represented in the form of system networks, not as an inventory of structures.
Of course, structure is an essential part of the description; but it is inter-
preted as the outward form taken by systemic choices, not as the defining
characteristic of language. A language is a resource for making meaning, and
meaning resides in systemic patterns of choice. (Halliday and Matthiessen,
2004, p. 23)
Following this tradition, Van Leeuwen (1999) has described the meaning
potential of sound in both language and music as a set of detailed system net-
works for perspective, time, melody and timbre. These system networks are
especially ‘appliable’ to the rap music context analysed here. Traditional music
analysis has tended to focus on the structural organization of musical sounds
(e.g. Lerdahl and Jackendoff, 1983), whereas Van Leeuwen is interested in par-
adigmatic organization, or ‘choice’. This is important because the process of
composing rap music relies heavily on ‘sampling’ from a set of pre-determined
musical sounds (see Rose, 1994). And as such, much of the focus is on paradig-
matic organization; selecting a particular musical sound from a larger system of
contrasting, meaningful choices.
This chapter will apply Van Leeuwen’s (1999) system networks to a selection
of vocalists sampled by ‘popular’ rapper and producer Kanye West. In short, the
chapter will examine the meaning potentials that can be assigned to West’s
vocalists according to the particular choices that characterize those voices. The
chapter then draws on that analysis to construct a preliminary semiotic profile
or ‘repertoire’ of West. Essentially, the chapter aims to answer the following
kinds of questions: what choices does West make in terms of his selection of
vocalists when constructing his rap songs? Does West sample widely, or is there
a consistent meaning potential associated with the ‘voices’ he samples? And
ultimately, what is his overall meaning potential; the character of Kanye West’s
semiotic repertoire?
In the tradition of social semiotics, the chapter will conclude by speculating
on Kanye West’s immense popularity. Drawing on Bernstein’s sociology (e.g.
1996) and his adaption of Durkheim’s distinction between mechanical and
organic solidarity, the chapter will put forward a hypothesis as to why West is so
widely consumed; why it is that he appeals to an audience outside the typical
hip-hop community.
14.2 Data
The data comprise a corpus of 30 rap songs from contemporary, ‘popular’,

African-American rap artist Kanye West (The College Dropout, 2004; Late Registra-
tion, 2005 and Graduation, 2007). The term ‘popular’ music is a particularly
loaded term and has been the subject of much debate within musicology and
related disciplines (Hesmondhalgh and Negus, 2002, pp. 1–11). Many musi-
cologists would argue for example that all rap music is ‘popular’ and that any
analysis of rap music constitutes ‘popular’ music studies. For this chapter how-
ever, the meaning of ‘popular’ is simply defined as the extent to which a rap
song is ‘widely experienced and/or enjoyed’ (Hesmondhalgh and Negus, 2002,
p. 2). While this is a difficult variable to measure, ‘accreditation’ is a means by
which one can at least get a sense of whether an artist’s songs are widely con-
sumed or not. In terms of Kanye West, Table 14.1 shows that since his inaugural
album in 2004, West has won over 50 awards and been nominated for more
than 100.
Importantly, these statistics are comparable to other successful North American
rap artists such as Eminem, Jay-Z, 50 Cent and Snoop Dogg. Table 14.1 not only
Table 14.1 List of Kanye West’s awards (from Aceshowbiz:

Kanye West’s Awards, n.d.)
Award Wins Nominations1
American Music Awards 0 7

BET Awards 6 3
BET Hip-Hop Awards 2 14
Billboard 4 4
Billboard R &B/Hip-Hop Awards 2 8
BMI Urban Awards 2 0
BRIT Awards 2 2
ECHO Awards, Germany 0 2
Grammy Awards 10 14
International Dance Music Awards 1 0
Metero Ireland Music Awards 1 0
MOBO Awards 5 4
MTV Asia Awards 0 2
MTV Australia Video Music Awards 0 3
MTV Europe Music Awards 1 6
MTV Philipinas Video Music Awards 0 1
MTV Video Music Awards 1 12
MTV Video Music Awards Japan 0 4
MuchMusic Video Awards 0 5
NAACP Image Awards 8 1
People’s Choice Awards, USA 0 1
ShockWaves NME Awards 1 0
Soul Train Music Awards 1 2
Source Awards 3 5
Teen Choice Awards 0 3
TRL Awards 1 1
Vibe Awards 1 3
World Music Awards 1 1
Total 53 108
highlights West’s popularity, but it also suggests that West’s consumers are from
a range of communities. West has received several nominations and awards
from associations that are specifically affiliated with the African-American com-
munity, for example, the Music of Black Origin Awards (MOBO), the Black Enter-
tainment Television Awards (BET) and the National Association for the Advancement
of Coloured People Image Awards (NAACP). However, West has also received many
nominations and awards from mainstream American music associations such
as the Grammy Awards and Billboard Awards, as well as a range of international
awards such as the British Record Industry Trust Awards (BRIT), MTV Asia,
Australia and Europe Awards, as well as the World Music Awards. Now these asso-
ciations do not exclude the hip-hop community. In fact, many have award
categories specific to that music culture. However, at the very least, they are
associations that represent communities that are not exclusively affiliated with
hip-hop culture.
West’s popularity, and particularly his appeal to those outside of the typical
hip-hop community, is not a coincidence. As an artist, West has made a con-
scious and very deliberate attempt to engage with a socially diverse consumer
base. The following extract from Time magazine (Tyrangiel, 2005), shows that
West is not concerned with aligning his music with one particular community.
Rather, West quite explicitly seeks to affiliate with as many consumers as possible:
One night while making the record, Brion says, he and West got in a state of
giddy exhaustion unique to people who spends hours a day for months on
end in a windowless recording studio. ‘We had just been talking about some-
thing, and there was one of those weird, intense lulls,’ says Brion. ‘Kanye
looks at me, and he goes, “You know that saying: you can’t be all things to all
people? Well seriously, why not? I want to be all things to all people.”’ Brion
waited for a moment, then burst into laughter. ‘I knew he wasn’t kidding, and
he’s smart enough to know that wanting to be loved by everybody is probably
really bad for your mental health, but at the same time his point was, you
know, why not try?’ You never know. He just might succeed. (Tyrangiel, 2005,
pp. 7–8)
14.3 Method of Analysis
Van Leeuwen’s (1999) book-length publication Speech Music Sound describes

the communicative resources of sound as they are used in music, speech and
other multi-modal contexts such as film and advertising. For Van Leeuwen
(1999), the meaning ‘potential’ of sound is realized in two main ways: the
experiential2 and provenance. While provenance is clearly important in the
rap-music genre, especially given the practice of ‘sampling’, this chapter will
focus exclusively on the experiential meaning potential: ‘the idea that our
experience of what we physically have to do to produce a particular sound
creates a meaning potential for that sound’ (Van Leeuwen, 1999, p. 205).
The analysis will specifically apply Van Leeuwen’s (1999) systems of timing or
rhythm (Figure 14.1) and voice quality (Figure 14.2).
Figure 14.1 System network for timing (after Van Leeuwen, 1999, p. 61)
Figure 14.2 System network for sound quality (Van Leeuwen, 1999, p. 151)
The application of the entire systems of rhythm and voice quality is beyond
the scope of this chapter. As such, only the sub-systems of tempo: slow and fast,
pitch register: high and low, as well as roughness: rough and smooth sounds,
will be analysed (circled in Figures 14.1 and 14.2). It must be noted that these
‘choices’ are not discrete either/or choices. Rather, they represent two poles
along a cline in which choice is a matter of degree. So for example, while a
tempo might be fast or slow, it can also be moderate, moderate to fast, moder-
ate to slow or anywhere along a cline from fast to slow.
The system of tempo is generally characterized according to the speed of that
tempo; is it fast or slow? More technically, tempo is realized as the ‘space’ or
duration of time between a pulse or ‘beat’. When there is a long amount of time
between pulses, the tempo is slow. When the space between a pulse is short, the
tempo is fast, or at least comparatively faster than the tempo which has a longer
space of time between pulses. For this chapter however, tempo will be consid-
ered in a slightly different way. Essentially, the analysis will measure the speed
of the rappers ‘flow’; how quickly they articulate their syllables. In other words,
what is the space between each of the rappers’ syllables? To do this, the analysis
will take a short extract of rapping (around 10 seconds), count the number of
syllables expressed and then produce an average figure of syllables expressed
per minute (spm).
From Van Leeuwen’s (1999) experiential perspective, the meanings associ-
ated with a fast and slow tempo are determined by what one does physically to
produce those tempos. So, for example, a fast tempo can be associated with
high energy levels, and perhaps even emotional attributes such as energetic,
excited, intense and nervous. A slow tempo on the other hand can be associ-
ated with low energy levels and attributes such as relaxed, calm, deliberate,
measured, and so on.
High and low pitch register is the scale from very low sounds to very high
sounds, usually associated with gender. This analysis is not interested in the
movement of a pitch, but rather the basic level or pitch ‘register’ of a voice. For
Van Leeuwen (1999), the high to low pitch register scale has the potential to
mean many things including emotional states, size and gender. Much like a fast
tempo, a high pitch register generally indicates an increased level of energy,
and as such can be associated with attributes such as excited, energetic, tense
and nervous. Van Leeuwen also associates a high pitch register with size and
gender, in the case of high pitch register, meaning ‘small’ and feminine. In
those examples, the experiential meaning potential extends beyond what one
does to produce a particular sound to consider the inherent physiology of
things that create those sounds.
While there can be exceptions, high pitch sounds tend to be expressed by
things, both human and non-human, that are small in size. And as such, a high
pitch register has come to be associated with meaning ‘small’. In addition,
women tend to have a higher pitch level than men, so a high pitch register has
also come to be associated with the ‘feminine’. In contrast, a low pitch register
has similar meaning potentials to a slow tempo; it implies a reduction in energy

or physical intensity. In terms of emotional states then, a low pitch register can
be associated with meanings such as relaxed, calm, deliberate and measured.
For size and gender, a low pitch register is generally associated with meaning
‘big’ and ‘masculine’.
The cline from rough to smooth sounds has to do with the perception of
friction in a sound, or whether there is no friction and the sound is perceived
as clean and smooth. With respect to rough and smooth sounds, Van Leeuwen
(1999) suggests that their meaning derives from precisely what they are: rough
sounds mean ‘roughness’, and smooth sounds mean ‘smoothness’. Again, fol-
lowing Van Leeuwen’s experiential rationale, one might add that rough sounds
have the potential to mean other similar attributes such as tough, hard and
strong. In contrast, a smooth sound may be associated with near-synonyms such
as soft, silky and pleasant.
While the tempo of the rhythm can be easily measured as syllables per min-
ute (spm), the pitch register and roughness of the vocal quality is more diffi-
cult to quantify. As such, the analysis of the voice quality is based on aural
perception and not any kind of computerized acoustic analysis. Because this
chapter is interested in the meaningful choices a voice can construe, as well as
the choices West has made in sampling those voices, the analysis involved
searching the West corpus, finding those songs that included other vocalists,
and then analysing those voices according to the three systems. Where possi-
ble, three voices were chosen for each of the systems, representing both poles
on the cline, as well as a median value. In this way, the analysis is similar to a
‘minimal pair’ test in phonology; identifying perceivable, meaningful contrasts
in the data.
14.4 Findings
The findings will be presented along a cline (Figures 14.3–14.8). Each sampled
vocalist that represents a distinct, meaningful choice is then plotted along that
cline in relation to the sampled voices they contrast with. Each vocalist is named
and referenced according to the song name, year of album publication, track
number (t) and time segment in the song (s).
The analysis includes two distinct data sets: the rapping voice and the singing
voice. Both voice types will be analysed according to rhythm, pitch register and
roughness. However, the fact that West samples both the rapping and singing
voice is itself a significant finding. While this chapter will not describe the char-
acter of these voice types, it is clear that each vocal performance is unique, and
affords different experiential meaning potentials. For this chapter, both voice
types are included so as to highlight the extent of West’s sampling.
It is also worth reiterating that the meaningful choices illustrated below are
not necessarily the only instances of that choice sampled by West. For each
cline, the most extreme realization of a particular choice is selected. This is not
to say that the West corpus comprises only one rapper with a rough voice qual-
ity, or one singer with a high pitch register. The aim of these illustrations is
to review the meaningful ‘choices’ available, and then highlight the extent to
which West samples from those systems. Also, as mentioned, these systems
of choice are gradable. So although the median or ‘in-between’ choice for each
variable is not discussed, they were identified. And this is noteworthy. First, not
only does it help reinforce one’s perception of the distinction between particu-
lar variables, but it also highlights the extent of West’s sampling.
14.5 Rhythm
The Figures below illustrate the slow to fast tempos sampled by West for the
rapping voice (Figure 14.3) and the singing voice (Figure 14.4).
As mentioned, the analysis of tempo can be quite easily quantified (as spm)
and then plotted along a cline from slow to fast rapping or singing. For the
Figure 14.3 Slow to fast rapping (as syllables per minute [spm]) in the Kanye
West Corpus
Figure 14.4 Slow to fast singing (as syllables per minute [spm]) in the Kanye West
Corpus
rapping voice (Figure 14.3), there is an immense difference between the ‘slow’
rapping of Mos Def (168 spm) and the fast rapping of Twista (516 spm). In fact,
one could argue that both rappers are very close to the threshold of slow and
fast rapping; the point at which vocal performance can no longer be defined
as rapping. Mos Def for example could express fewer syllables per minute.
However, that would mean only one syllable per pulse, in which case, the vocal
performance sounds more like chanting than rapping. Similarly, Twista’s per-
formance is at the very, very fast end of the rapping scale. In fact, most of his
lyrics are difficult, if not impossible to comprehend.
The contrast between slow and fast tempos in the singing voice (Figure 14.4)
is not as marked as it is in the rapping voice. Nevertheless, the contrasts are still
clearly meaningful when we compare for example the slow, melodic vocals of
Albert Brumley, with the fast, up-tempo vocals of Connie Mitchell. What is most
important here however is not so much the contrast between the singing voices,
but rather, how these findings relate to West’s overall corpus. As mentioned,
the rapping voice, by its very nature, must express a minimal number of sylla-
bles per minute, otherwise the performer is no longer rapping. The singing
voice however enables (or restricts) a vocalist to far fewer syllables per minute.
Brumley’s vocals equate to only one syllable per second. The point is this: by
sampling the singing voice, West is able to introduce a much slower tempo of
vocal performance than could ever be achieved by exclusively collaborating
with rappers.
14.6 Pitch Register
The Figures below illustrate the low to high pitch registers sampled by West for
the rapping voice (Figure 14.5) and the singing voice (Figure14. 6).
The pitch register analysis, like the roughness analysis, has not been quanti-
fied or formally measured. However, an acoustic analysis is not necessary to
reveal the fundamental contrasts between these voices. There is a clear, perceiv-
able distinction between the low pitch registers of rapper GLC and singer Bill
Figure 14.5 Low to high rapping pitch registers in the Kanye West Corpus
Figure 14.6 Low to high singing pitch registers in the Kanye West Corpus
Withers when compared with the high pitch registers of Strings and Laura
Nyro. Of course this is in part attributable to the differences in gender. One
might argue that it is unfair to analyse for pitch register when gender is not
a controlled variable. However, the fundamental concern of this chapter is
‘choice’. And that means these contrasts are important. West has chosen female
vocalists, both rappers and singers. Accordingly, West has sampled both low
pitch registers, as well as high pitch registers. Even more importantly, the differ-
ence between these low and high pitch registers is extreme. Rapper GLC is
renowned for his low pitch register. It is arguably the most salient feature of
his voice quality. In contrast, Laura Nyro’s pitch register is not simply ‘high’; it
is extremely high. It may have even been electronically manipulated as it sounds
similar to those high-pitched, ‘chipmunk’ voices.
14.7 Roughness
The Figures below illustrate the smooth to rough voices sampled by West for
the rapping voice (Figure 14.7) and the singing voice (Figure 14.8).
The perception of smoothness and roughness in the vocalists’ voice quality is
found in both the rapping voice and the singing voice. Much like the rhythm
and pitch register analysis, the smoothness and roughness illustrated in the
Figures 14.7 and 14.8 are easily perceivable examples of those choices. The
smooth rapping voice of Lupe Fiasco is in stark contrast to the rough, gritty,
obstructed voice quality of Lil’ Wayne. In a similar way to rapper GLC and his
low pitch register, these artists are renowned for their smoothness and rough-
ness respectively. Lil’ Wayne in particular has a very creaky, rough sounding
voice quality.
The contrast between smoothness and roughness is also easily perceivable in
the singing voice. In a similar way to the high pitch register of Laura Nyro, West
and his producers have manipulated the voice quality of Daft Punk to make it
especially rough. It could be argued that the resulting sound is more vibrato
than it is rough. Regardless, there is clearly a distinction between the smooth,
Figure 14.7 Smooth to rough rapping in the Kanye West Corpus
Figure 14.8 Smooth to rough singing in the Kanye West Corpus
tender singing voice of Labi Siffri, and the synthesized, vibrating and ultimately
‘obstructed’ vocals of Daft Punk.
14.8 The West Repertoire

The findings reveal that Kanye West samples widely. In terms of the tempo,
pitch register and roughness of his vocalists, West engages with the full range of
meaning potentials available for each of these systems. Put simply, West’s eclec-
tic sampling enables him to increase his meaning potential, and as such, he has
an immense repertoire; he makes many meanings. Following Van Leeuwen’s
(1999) experiential semiotics, the West repertoire can be further characterized
in terms of specific attributes, mostly to do with some kind of emotional or
physical disposition, as well as attributes to do with gender and size. Figure 14.9
is a crude representation of the West repertoire according to some of the main
contrasting attributes identified in the analysis.
The West repertoire can also be represented from a genre-oriented, topologi-
cal perspective (Martin and Rose, 2008). Following the reference text The All
Figure 14.9 Some attributes (or meaning potentials) in the Kanye West
Repertoire
Figure 14.10 The Kanye West repertoire from a topological perspective
Music Guide to Hip-Hop (2003) and its list of rap styles, Figure 14.10 compares
the meaning potential of West’s vocalists to the vocals of other rap styles. In
short, Party Rap has tended to be characterized according to high pitch regis-
ters and smooth voice quality, Political Rap according to fast tempos, Hardcore
Rap according to low pitch registers and rough voice quality and Old School
Rap according to slow tempos. Figure 14.10 shows that West, through his vast
sampling of vocalists, is not limited to one particular style, and is therefore able
to expand his semiotic repertoire.
While the discussion and illustrations thus far are useful in terms of generally
characterizing West’s semiotic repertoire, it does raise two important questions:
to what extent can one relate the experiential meaning potentials of the vocals
to the actual artist who produces those vocals, and to what extent can one
characterize the meaning potential of Kanye West according to the vocalists he

samples from.
As illustrated in Figure 14.9, it seems reasonable to assign the musical attri-
butes or meanings associated with an artist’s performance to that artist and
his or her semiotic repertoire. For example, it is common for a music critic or
fan to refer to an artist as ‘brash’ or ‘strong’ or ‘nasty’ with respect to the sounds
they produce. But can a ‘rough’ sounding voice simply be equated with a ‘rough’
artist? In its new formation, the attribute takes on a slightly different meaning.
From an Appraisal perspective (Martin and White, 2005), the term ‘rough’ is
no longer an Appreciation of a thing, but rather a Judgement of someone; per-
haps as a negative Judgement of propriety, or a positive Judgement of tenacity.
According to Cléirigh (personal communication, 17 October 2008), one way
of resolving this issue may be to reconsider voice quality in Halliday’s (1985,
pp. 30–1) terms; as indexical of an individual’s character.
The other vexed issue is the extent to which West’s vocalists can be consid-
ered part of his semiotic repertoire. In Figure 14.10, for example, West’s own
voice quality does not ‘cover’ the same amount of semiotic territory as it does
when his collaborators are included. In fact, West’s vocals would at best only
‘cover’ Party Rap, and maybe Old School Rap. He does not have the fast tempo
or the low, rough pitch register of the Political rappers and Hardcore rappers.
So is it then fair to characterize West’s repertoire as huge and immense if it
includes other vocalists? The point here is that West is not just a rapper, he is
also a producer. And the role of producer in the rap music context is highly
valued. West has to select vocalists as well as other musical sounds that are
then integrated into an entire musical composition. So, in the role of producer,
Kanye West’s repertoire is immense.
14.9 Some Thoughts
In September 2007, rap artists Kanye West and 50 Cent engaged in a well-
publicized ‘feud’ to see which rapper would sell the most copies of their upcom-
ing albums Graduation (2007) and Curtis (2007). Within a week of the albums’
release, Kanye West was declared the victor. The Billboard charts (Mayfield,
2007) showed that in its first six days of release, West’s Graduation had accumu-
lated a total album sale figure of 957,000, which ranks it fifteenth for ‘all-time’
sales weeks. In contrast, 50 Cent’s Curtis sold a total number of 691,000 albums;
a solid result, but significantly less than West’s Graduation.
A primary concern for social semiotics is to not only describe semiotic systems,
but to investigate how these systems are actually used in real-life social contexts.
In a performance context such as this, where there is an agenda to sell records
and make money, the extent to which an artist affiliates with their consumers is
particularly important. If we simply define ‘affiliation’ as the extent to which we
can ‘share’ in another’s feelings (Martin, 2004), then a semiotic profile which
describes the kinds of feelings construed by an artist should provide some

important insights. In this particular case, is there something different about
the meanings Kanye West construes compared with 50 Cent? And if so, why do
these enable him to affiliate with more consumers than 50 Cent?
A complete semiotic profile of both West and 50 Cent is an immense, if not
impossible task. Not only does one have to consider an artist’s musical sounds,
but there is also the meaning potential of their rap lyrics, video clips, clothing,
merchandise and the like. For heuristic purposes though, music websites such
as Allmusic (Birchmeier, n.d.) provide a useful overview of an artists’ semiotic
repertoire. Table 14.2 illustrates the kinds of ‘moods’ that people generally
associate with the music of Kanye West and 50 Cent.
The West profile outlined above helps support the semiotic analysis described
in this chapter: that West has an immense semiotic repertoire. More impor-
tantly though, these profiles allow analysts to generally compare and contrast
the repertoires of specific artists. In this case, the profiles show a clear contrast
between the repertoire of Kanye West and 50 Cent. Not only does West’s list
have a greater number of ‘moods’, there is also a more varied range of meanings
or ‘moods’ associated with West. 50 Cent’s list tends to be similar kinds of terms,
basically to do with aggression and toughness. West on the other hand ranges
Table 14.2 The ‘moods’ of Kanye West and 50 Cent

(from Birchmeier, n.d.)
Kanye West 50 Cent
Humorous Street-Smart
Brash Thuggish
Energetic Menacing
Confident Confident
Swaggering Brash
Reflective Provocative
Street-Smart Aggressive
Snide Confrontational
Party/Celebratory Rebellious
Confrontational Volatile
Boisterous Angry
Fun Fiery
Gleeful Hostile
Summery Harsh
Giddy Visceral
Relaxed Witty
Hungry
Intense
Passionate
Laid-back/Mellow
Amiable/Good Natured
from extremely positive moods such as ‘giddy’ and ‘excited’ to more aggressive
moods such as ‘brash’ and ‘confrontational’.
So how do these profiles inform this discussion? Can one simply argue that
West is able to affiliate with many consumers, and in this case, more consumers
than 50 Cent, because he offers his audience a smorgasbord of meanings from
which they can ‘pick and choose’ according to the meanings they share in
and identify with? In this way, the process of affiliation is simple: the greater the
meaning potential, the greater the number of consumers. And this may in part
explain why West is more ‘popular’ than 50 Cent. However, the sociology of
Bernstein (1996) offers an even more convincing explanation. In particular,
Durkheim’s distinction between mechanical and organic solidarity seems par-
ticularly relevant here. The following extract provides a basic definition of both
mechanical and organic solidarity:
Mechanical solidarity is said to characterize a society where there is a high

degree of uniformity and consensus. Values and sentiments are shared
and adhered to . . . Organic solidarity is the outcome of diversity . . . there is
progressively greater scope for the development of individual differentiation
among persons and institutions. Individuation entails a corresponding
decrease in consensual value systems and sentiments. (Atkinson, 1985, p. 24)
50 Cent’s repertoire is limited in its meaning potential. The kinds of mean-

ings, ‘moods’ or values that 50 Cent expresses are not diverse. They are limited
to attributes such as ‘aggression’ and ‘toughness’. In this way, 50 Cent repre-
sents mechanical social integration; his consumers value uniformity and do not
seek to share in diversity. Kanye West, in contrast, is especially representative of
organic social integration. His immense repertoire of meanings facilitates a
solidarity built on diversity, not uniformity. His consumers value individual dif-
ferentiation. They do not only ‘pick and choose’ from his repertoire, they
engage with the entire smorgasbord of meanings.
There is much more to be said here in terms of social class and how that
might relate to West’s ‘individuation’ and his capacity to affiliate with a large,
socially diverse consumer base. Does 50 Cent align with working-class consum-
ers, and Kanye with middle-class consumers? Probably an oversimplification,
but certainly worthy of further investigation. If nothing else, West seems to have
‘tapped into’ the values of Generation X and Y, who, like West, want to be ‘all
things to all people’.
Notes
1
The number of nominations does not include Wins.
2
The term ‘experiential’ is not to be confused with the experiential component of
the ideational metafunction (Halliday and Matthiessen, 2004).
References
50 Cent. (2007). Curtis [CD]. Santa Monica, CA: Aftermath, Interscope, Shady.
Aceshowbiz: Kanye West’s Awards (n.d.). Retrieved 12 November 2008 from http://
www.aceshowbiz.com/celebrity/kanye_west/awards.html.
Atkinson, P. (1985). Language, Structure and Reproduction: An Introduction to the Socio-
logy of Basil Bernstein. London: Methuen.
Bernstein, B. (1996). Pedagogy, Symbolic Control and Identity: Theory, Research and
Critique. London: Taylor & Francis.
Birchmeier, J. (n.d). ‘50 Cent Biography’. Retrieved 12 November 2008 from http://
www.allmusic.com/cg/amg.dll.
Birchmeier, J. (n.d). ‘Kanye West Biography’. Retrieved 12 November 2008 from
http://www.allmusic.com/cg/amg.dl.
Caldwell, D. (2010). ‘Making metre mean: rhythm and identity in the music of Kanye
West’, in M. Bednarek and J. R. Martin (eds), New Discourse on Language: Func-
tional Perspectives on Multimodality, Identity and Affiliation. London: Continuum,
pp. 59–80.
Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of
Language and Meaning. London: Edward Arnold.
Halliday, M. A. K. (1985). Spoken and Written Language. Victoria: Deakin University.
Halliday, M. A. K., and Matthiessen, C. M. I. M. (2004). Introduction to Functional
Grammar (3rd edn). London: Edward Arnold.
Hesmondhalgh, D. and Negus, K. (2002). ‘Introduction’, in D. Hesmondalgh and
K. Negus (eds), Popular Music Studies. London: Arnold, pp. 1–11.
Keyes, C. (2002). Rap Music and Street Consciousness. Illinois: University of Illinois
Press.
Kress, G. and Van Leeuwen, T. (1996). Reading Images: The Grammar of Visual Design.
London: Routledge.
Krims, A. (2000). Rap Music and the Poetics of Identity. Cambridge: Cambridge Univer-
sity Press.
Lerdahl, F. and Jackendoff, R. (1983). A Generative Theory of Tonal Music. Cambridge,
MA: MIT Press.
Martin, J. R. (2004). ‘Mourning: how we get aligned’. Discourse and Society 15(2–3),
321–44.
Equinox.
Martinec, R. (2000). ‘Types of process in action’. Semiotica 130 (3/4), 243–68.
Maxwell, I. (2003). Phat Beats, Dope Rhymes: Hip-hop Down under Comin’ Upper. Middle-
town: Wesleyan University Press.
Mayfield, G. (2007). Kanye Crushes 50 Cent in Huge Album Sales Week. Billboard.
Retrieved 12 November 2008 from http://www.billboard.com/bbcom/news/
article_display.jsp?vnu_content_id=1003642725.
O’Toole, M. (1994). The Language of Displayed Art. London: Leicester University
Press.
Pennycook, A. (2007). Global Englishes and Transcultural Flows. London: Routledge.
Richardson, E. (2006). Hip-hop Literacies. London: Routledge.
Rose, T. (1994). Black Noise: Rap Music and Black Culture in Contemporary America.
London: Wesleyan University Press.
Tyrangiel, J. (2005). Why you can’t ignore Kanye. Time. Retrieved 15 May 2007 from
http://www.time.com/time/prinout/0,8816,1096429,00.html.
Walser, R. (1995). ‘Rhythm, rhyme, and rhetoric in the music of public enemy’.
Ethnomusicology 39 (2), 193–218.
West, K. (2004). The College Dropout. New York, NY: Roc-A-Fella/Island Def Jam.
West, K. (2005). Late Registration. New York, NY: Roc-A-Fella/Island Def Jam.
West, K. (2007). Graduation. New York, NY: Roc-A-Fella/Island Def Jam.
Chapter 15
Rank in Visual Grammar: Some Implications

for Multimodal Discourse Analysis1
Sumin Zhao
15.1 Introduction
Since the groundbreaking publication of Reading Images: The Grammar of Visual

Design (Kress and Van Leeuwen, 2006 [1996]) and The Language of Displayed Art
(O’Toole, 1994), multimodal discourse analysis (MDA) has established itself as
one of the key research areas within the discourse analysis community. One of
the key pursuits of MDA has always been the understanding of human semiotic
practices in various social contexts. As a result, recent years have witnessed the
increasing impact of MDA studies among semiotic practitioners (e.g. Bateman,
2008; Martinec and Van Leeuwen, 2009) and language educators (e.g. Unsworth,
2007; 2008). MDA research, in this sense, showcases the new approach – appli-
able linguistics – to linguistic and semiotic research advocated in this volume
(see Mahboob and Knight, this volume).
Like the two MDA classics, many contemporary MDA studies have close theo-
retical connection to the general theory of Systemic Functional Linguistics
(SFL). These studies, on the one hand, draw heavily on the SFL description of
language and its model of text–context relations. On the other hand, the explo-
rations of modalities other than language (MOLs) (e.g. colour: Kress and Van
Leeuwen, 2002; Painter, 2008; gesture: Martinec, 2001; 2004; Zappavigna, 2010;
laughter: Knight, forthcoming; mathematical symbolism: O’Halloran, 2005; music:
Van Leeuwen, 1999; Caldwell, in press; space: Stenglin, 2004; Ravelli, 2006; topog-
raphy: Van Leeuwen 2006), and more recently the investigations of intersemi-
otic relations (e.g. Martinec and Salway, 2005; Matthiessen, 2007a; Norris, 2004;
O’Halloran, 2008; Royce, 2007; Zhao, 2010) have encouraged the researchers
to reflect back on some much debated issues in SFL: rank, genre, instantiation
and dynamic modelling, to name just a few.
In this chapter, I will focus on one of these issues: rank. Rank is essentially
a theory of compositional hierarchy. Lying at the centre of the theory is the
Figure 15.1 Rank scale in English
notion of constituency. In the SFL model of English (see Figure 15.1), for
instance, there are four scales of constituency at the lexicogrammar level,
known as ranks (from the highest to the lowest): clause, phrase/group, word
and morpheme. Each rank consists of one or more units of the rank immedi-
ately below. Through grammatical recourses such as logico-semantic relations
(e.g. elaborating, extending, projection, etc), units at every rank can form com-
plexes, for example, clause complexes, group complexes, etc. A unit at higher
rank may also down-rank to function in the structure of a unit of its own rank
or of a lower rank (cf. Halliday and Matthiessen, 2004).
The purpose of this chapter is NOT to argue for or against the use of rank in
multimodal descriptions. Instead I am interested in the ways in which the re-
conceptualization of rank has shaped the systemic functional descriptions of
MOLs and multimodal texts in general. Simply put, the chapter focuses on
examining the practice of theorization in MDA rather than any theory per se. I
will begin my discussion with a short review of O’Toole’s (1994) rank-based and
Kress and Van Leeuwen’s (2006) rank-free grammars of images. I will then trace
the origins of the concept of rank in SFL and illustrate the ways in which the
theoretical borrowing of the concept may promote or impede our understand-
ings of multimodality. Finally, I will use the rank ‘debate’ as a springboard for
addressing some of the urgent issues in the development of MDA theories and
practices.
15.2 To Rank or Not to Rank?
Within the MDA community, there has been a long-established disagreement

over the adaptation of rank (cf. Martinec, 2005) in the theoretical description
of MOLs and multimodal texts in general. The division has its origins in the
different treatments of visual structures in the two MDA classics: the rank-free
visual grammar of Kress and Van Leeuwen (2006) and the framework for multi-
rank analysis of images proposed by O’Toole (1994). If, therefore, we are to
Rank in Visual Grammar 253
Figure 15.2 A Queensland hero
account for the ideational structure2 (i.e. the representation of events, the
objects and participants involved and the circumstances in which they occur) of
the photographic image in Figure 15.2, we are left with two options. Applying
Kress and Van Leeuwen’s visual grammar, we could analyse the visual structure
as a Reactional Process, since the image construes the process of ‘looking’. The
girls are the Reactors (the observers), and are connected through eyeline
vectors (see Figure 15.2) to the Phenomenon (the observed) – the man in the
middle. Further, the signing action of the man can be treated as an embedded
Transactional Action Process, the man being the Actor (the doer) and the note-
book being the Goal (the participant impacted by the doing).
Alternatively, applying O’Toole’s rank grammar, we could describe the
image as having three ranks (from the highest to the lowest): Episode – Figure –
Member. At the rank of Episode, a man is signing an autograph for a group of
girls in front of a house. Moving a rank down to the behaviour of individual
Figures, the man is signing on a notebook, while the girls are watching his
action attentively. At the rank of Member, the man is dark-skinned and his facial
expression is rather serious.
Both analyses seem plausible. As discourse analysts, we are now facing a
dilemma: which analysis shall we choose? To provide a solution for the prob-
lem, I suggest that we could draw a parallel between the current dispute and the
debate between Halliday (2002 [1966]) and Matthews (1966, cited in Halliday
2002[1966]) over the merit of a rank constituency grammar of language in the
1960s. In essence, Halliday’s and Matthew’s stances reflected two different
hypotheses on the nature of language. Both approaches were difficult to falsify
theoretically – Halliday’s rank grammar, nevertheless, has been favoured among
the SFL linguists and many others for its descriptive power. Similarly, if we need
to argue for/against a rank grammar of image and other MOLs, we need to
raise at least two questions: whether the theory can be falsified3; and whether it
provides more descriptive power.
In the development of MDA theories, however, these questions are not always
explicitly addressed. The common assumption holds that the choice of includ-
ing or excluding rank in the theoretical description is determined by factors
such as the complexity and the nature of the semiotic phenomenon under
examination, the purpose of the analysis, etc. (cf. Martinec, 2005). This affor-
dance of relative freedom in including/excluding rank in MDA description has
encouraged exploration of a diverse range of MOLs and multimodal texts. As a
result, a fruitful body of works has been produced in the last decade (cf. Baldry
and Thibault, 2006; Bateman, 2008; Jewitt, 2009; Bednarek and Martin, 2010;
Norris, 2004; O’Halloran, 2004; Royce and Bowcher, 2007; Unsworth, 2008;
Ventola et al., 2004).
The flexibility in theoretical conceptualization, on the other hand, has led to
confusion and low consistency in analytical practices. If we are to extend our
analysis to Figure 15.2, we will immediately encounter such problems. Employ-
ing Kress and Van Leeuwen’s visual grammar, for instance, we would find it
difficult to determine the main process (realized by a vector) in the image.
There are at least five vectors in the image (see Figure 15.3): 1) an action vector
extending from the batsman towards the top-right of the frame; 2) another
action vector linking the participant on the bottom-left (the fielder) to the
ground; 3) a third action vector leading the running figure on the bottom-right
(the fielder) to the cricket ball on the ground; 4) the eyeline vector linking
the participant on the left (the wicketkeeper) to either the falling fielder or
the cricket ball; 5) the eyeline vectors directed from two participants on the top
of the frame (the fielders) to the batsman. We have all together five processes:
two Non-Transactional Actions (the batsman and the falling fielder), one Trans-
actional Action (the running fielder) and two Transactional Reactions (the
gazes). One possible solution here would be to choose the Non-Transactional
Action of the batsman as the main process since the textual salience is given to
her. The other processes then could be treated as ‘embedded’ minor processes
(Kress and Van Leeuwen, 2006, p. 107). This analysis does seem to address the
issue of multiple processes in a complex image. However, simply grouping pro-
cesses as major or minor ones fails to capture the nuances of the semantic rela-
tions between the processes such as ‘causality’ (the falling of the fielder is
possibly triggered by the previous actions of the batsman) or ‘simultaneity’ (the
fielder at bottom-left runs at the same time as the batsman). More importantly,
it leaves one central question of SF-MDA research unaddressed: how do (static)
images construe/represent a series of actions differently from language?
To capture the complexity, it would seem reasonable to introduce the
concept of rank. If we apply O’Toole’s (1994) four-scale visual grammar to the
same image in Figure 15.2, we would be able to recognize the following units: 1)
Figure 15.3 Women cricketers fielding
Work (the whole fielding scene); 2) Episodes (the three gazing fielders, the
batsman, the falling fielder, the running fielder); 3) Figures (each individual
participants) and 4) Members. Next, we could analyse each unit as well as the
relations between the units at a given rank (e.g. the relations between the bats-
man Episode and the falling fielder Episode). While addressing the complexity
of the image, the assigning of the rank scale nevertheless gives rise to a different
set of problems. For instance, the four Episodes of the fielding scene do not in
fact make up Work – the rank above. Instead they are linked by certain semantic
relations and hence form a complex – the Episode complex. This leads to two
further questions: 1) is complexing allowed at all ranks? For example, if mem-
ber complexes can be generated by the same semantic resource; and 2) should
the image be analysed as involving two rank-shifts? If this is the case, should
these rank-shifts be treated as up-ranking movement (a Figure upgrading to
function as an Episode) or down-ranking one (a Figure embedding an Episode)?
For the analyst, one foreseeable difficulty would be to decide on the direction
of the analysis: from Member (the lowest rank) to Work (the highest rank)
or vice versa? If we proceed from the higher rank to the lower ones, as most
rank-based analytical practices do, it would be relatively problematic to decide
on the depth of the analysis. Furthermore, in Figure 15.3, the facial expres-
sion/make-up/gesture of each participant can hardly be recognized. In this
case, should the analysis stop at the rank of Figure (i.e. the act of individual
participant)? If in a similar photo such details are present, should the des-
cription then include the rank of Member (i.e. the expression/gesture of the
individual)?
Another problem in rank-based MDA model is that it has unintentionally
created a discrepancy between the theoretical conceptualization of rank and its
application in the MDA practices. In Halliday’s functional grammar, rank serves
as the starting point for generalizing systems and structures of the language.
In most MDA studies, in contrast, rank is usually the ending point of the theo-
retical description. A rank-based research typically involves assigning rank scale
and proposing potential meaning-making systems (e.g. gaze, expression). The
paradigmatic choices and their syntagmatic realizations within each system,
however, are seldom specified. As a result, it is difficult to ensure consistency
across analyses since there is no generalized principle guiding the application
of the systems in analysis.
So far, I have compared Kress and Van Leeuwen’s and O’Toole’s treatments
of rank in the description of image, showing both their advantages and
disadvantages. What needs to be emphasized here is that the purpose of my
analytical exercise is not to criticize either theory. The inability of Kress and Van
Leeuwen’s visual grammar to account for large-scale images seems to be the
consequence of a deliberate theoretical choice (see Van Leeuwen in Martinec,
2005). Many aspects of O’Toole’s model require further theorization for few
system networks have being offered so far. We have not yet been able to test the
accountability of constituents at any rank4.
What we need now is to further test the descriptive advantages of including
or excluding rank in the descriptions of MOLs and multimodal texts. However,
in my opinion, the prevailing trend in current MDA research does not seem to
encourage such theoretical development. In the following section, I will pres-
ent the reasons why we are unable to provide solutions to the current rank
debate in MDA by re-interpreting Halliday’s original conception of rank in SFL
theory.
15.3 Rank: Re-reading Halliday
In this section, I will provide a re-interpretation of Halliday’s rank theory.

The purpose of the re-reading is to reveal the missing elements in the current
debate on rank in MDA. For me, there are at least three interrelated aspects in
Halliday’s rank theory need to be emphasized here:
1. Rank is not an isolated SFL parameter. Theoretically, it is closely linked with

other SFL parameters, such as system/structure, metafunctions, etc.
2. Rank theory is essentially a hypothesis about the nature of language.
3. Rank grammar is ultimately a constituency grammar.
15.3.1 Rank is not an isolated parameter in the theoretical

descriptions of SFL
First, the MDA borrowing of SFL theory tends to treat linguistic parameters
as isolated categories. In SFL, however, the theoretical description of one
parameter is closely associated with the descriptions of the others. Rank, for
instance, ‘defines a point of origin for structures and systems’. In Halliday’s
words:
. . . to show that a system operates on a given rank is the first step in stating its
relationship to other systems; likewise assigning an item to a given rank is the
first step in stating the systemic and structural relations into which it may
enter and those which it may embody within itself. On the structure axis, rank
is a form of generalization about bracketing, and makes it easier to avoid the
imposition of unnecessary structures . . . (Halliday, 2002 [1966], p. 120)
Systemic functional linguistic theory centres essentially around the mapping

of paradigmatic systems – transitivity, mood and theme, and their syntag-
matic realizations at the rank of clause.
To argue for a rank-based grammar of any semiotic recourse, therefore, we
need first provide some consistent descriptions of the systemic (paradigmatic)
relations at a given rank and then prove the accountability of constituents at
other ranks. However, as previously mentioned, although most rank-based stud-
ies have been able to assign rank scales and propose potential meaning-making
systems at each rank, they fail to map out the paradigmatic choices. In most
rank-free models where systems are offered, on the other hand, the main chal-
lenge is to substantiate structural (syntagmatic) realization for each systemic
(paradigmatic) choice.
15.3.2 Rank grammar is a constituency theory of language

The second point I want to put forward is that rank grammar is essentially a
constituency grammar of language. The concept of constituency is best explained
in Pike’s theory of tagememics (cf. Pike, 1982). Taking the following clause for
example – Eddie Gilbert is signing autographs, we would observe that the clause is
made up of parts or particles (Eddie – Gilbert – is – signing – autographs), and each
of these parts is in turn made up of smaller particles (auto-graph – s). A particle
is called a constituent. Constituency theory, largely influenced by the studies of
written language, has played a dominant role in the development of modern
linguistic theories. However, if we read the clause aloud in a natural manner, we
will notice that the boundaries between the particles become fuzzy. We might
not, for instance, be able to separate Gilbert from is. To capture this phenome-
non, Pike proposed a new structure: the wave structure. Wave structure is not an
alternative interpretation of the nature of language but a complementary one.
To put it in another way, language is both particulate-like and wave-like.
Influenced by Pike’s theory of tagememics, the SFL theorists (Halliday, 2002
[1979]; Martin, 1996; 2000a) recognize the three types of meanings, known
Figure 15.4 Types of structures and types of meanings
as metafunctions, as complementary in their structural manifestations (see

Figure 15.4):
Ideational Meaning (construing human experience): particulate

Interpersonal Meaning (enacting interpersonal relations): prosodic
Textual Meaning (organizing information): periodic
The concept of complementary structures, in my opinion, remains essen-

tially a conceptual one. In SFL, the clause is considered the central meaning-
processing unit (cf. Halliday and Matthiessen, 2004) both in theory and in
practice. The fact that the prosodic and periodic structures map onto the par-
ticulate structure in a clause seem to have obscured the need for disentangling
these different abstractions of meaning at the clause level. Once we move
beyond the clause, the particulate-based model of meaning starts to reveal its
limitation. Both Cléirigh’s (1998) work on phonology and Martin and White’s
(2005) appraisal theory (discourse-semantics), for example, have advocated
the importance of prosodic structure as a complementary perspective.
Now, if we are to apply these principles to the description of images, we need
to ask at least two basic questions: is visual structure particulate-like; and is
an image comparable to a clause? To answer the first question, we need to
re-examine the images in Figures 15.2 and 15.3. In both cases, there seem to be
some strong arguments for constituency since the participants (i.e. the human
figures) are easily recognizable as units. However, if we zoom into each indi-
vidual participant, it becomes increasingly difficult to recognize the boundaries
between particulates. Taking facial expression as an example, it is questionable
to describe the expression of ‘happiness’ as made up of eyes, teeth and blushes
on the cheeks. Part–whole (constituency) relations may also be problematic
if we are to further include visual genres where the boundaries between units
are indistinguishable, such as landscapes or abstract paintings.
The point here is that language construes experience in terms of part–whole
relations, which are in turn manifested in linguistic structures. The semantic
Figure 15.5 The construal of membership in language and image
elements of processes and participants, for instance, are realized respectively by

the grammatical units of verbal group and nominal group. Images, in contrast,
do not seem to organize experience in the same fashion. Visual processes, for
instance, are sometimes embedded in/embodied in the participants, especially
the ones categorized by Kress and Van Leeuwen (2006) as Non-Transactional
Reaction and Symbolic structures (see Figure 15.5). If we are not able to recog-
nize constituency in the first place, the introduction of rank in visual analysis
will always be open to discussion.
Turning to the second question – whether an image is comparable to a clause,
the popular consensus in MDA seems to be that it is, whether the analyst
includes or excludes rank in the description. Such belief is in fact a reflection
of our understanding of text, as I will illustrate in the following section.
15.3.3 Rank grammar is a theory of the clause

Although in theory, the constituency model should be able to apply to semantic
units as well as grammatical ones, in practice, it is far from clear that a single
rank scale can be generalized across various types of text. Existing literature
suggests that texts of different registers operate with different compositional
scales (see Cloran, 1994; Sinclair and Coutlhard, 1975 cited in Halliday and
Matthiessen, 2004; see also Eggins and Slade, 2005; Van Leeuwen, 1993). Images
in terms of rank scale seem to resemble a text rather than a clause. That is, dif-
ferent types of images tend to operate with different rank scales. A portrait for
instance does not usually involve hierarchical structure, while a Chinese hori-
zontal scroll painting will typically consist of more than four compositional
scales. If we are to propose a rank grammar of image, a lot more research needs
to be carried out on images of various genres.
However, it is unclear at this stage that constituency-based research would be
more productive for MDA. A unified, generalized rank scale such as the one for
lexicogrammar offers great theoretical advantage, for it ensures the stability of
the systemic descriptions. In other words, it allows us to map out the potential
of a whole language in a relatively manageable number of systems. However, if
rank scale itself is subjective to registerial variations as in the case of semantics
(discourse-semantics) or images, it will be even more difficult to predict the

potential systemic variants.
The underlining issue here is whether the constituency model is the prefer-
able option for MDA and discourse analysis in general. Although in theory, SFL
acknowledges both the product and process nature of text, in the analytical
practices adopted, a text is largely treated as being made up of smaller parts,
just like a clause. Treating text as a collection of clauses undoubtedly has
its theoretical advantage for it allows us to generate language patterns across
unprecedented amounts of texts (e.g. Butt et al., 2004) and assists in testing and
modifying the system networks that were originally built on manual analysis.
However, the nature of the method necessitates the sacrifice of the discursive
features that are key to the meaning-making process, such as sequencing
(Zappavigna, 2010) and non-linear coupling (i.e. the ‘snowballing’ of mean-
ing) (Zhao, 2010). In short, it prompts unintentionally a synoptic/product view
of text at the cost of a dynamic/process one.
The limitation of the synoptic perspective is further accentuated in the analy-
sis of multimodal texts. To illustrate this point, I will again use the image in
Figure 15.2 as an example; only this time I will put the image back in its original
text environment. In the original layout, there are three different sets of verbal
text accompanying the image, the master title, the caption and the main text.
The textual elements (e.g. the placement of the visual element: top-down, left-right,
visual salience: font, size, etc.) naturalize a possible unfolding path of the text as
the following (see Figure 15.6): A Queensland hero – ‘the image’ – Eddie Gilbert
signing autographs – Though in the end, the cricket officials succeeded in destroying
Gilbert’s career because of his skin. Australian fans loved him. He was very popular and
Australians took national pride in his career.
What we immediately notice is the subtle shift in our description of the
visual structure as well as the text–image relations during the unfolding of
the multimodal text. The verbal texts establish respectively three different
sets of relations with the image: 1) the master title ‘classifies’ the central partici-
pant in the image as ‘a hero’; 2) the caption ‘names’ the visually represented
participant, while the image provides the circumstantial details for the verbal
Figure 15.6 The unfolding path of the multimodal text

text; for example, where the signing took place; 3) the main text provides extra
ideational information that is not represented in the image such as the destroy-
ing of Gilbert’s career. At same time, it attributes interpersonal meanings to the
event in the image, that is, positive affect (love, pride) and judgement (popular)
(cf. Martin and White, 2005). The girls’ act of watching, for example, can be
seen as the token for ‘love’ and ‘pride’. In other words, the ideational meanings
of the image are now coupled (Martin, 2000b; Zhao, 2010) with the interper-
sonal meanings construed in the text.
The example here shows the limitation of a constituency model, for a multi-
modal text is not simply made up of smaller units of texts and images, nor is its
meaning an accumulation of separate parts. Rather, it is a meaning-making
process, in which choice from one semiotic system is constantly coupling with
or decoupling from the choice made in another system (Zhao, 2010).
15.4 The Implications of Rank Debate for MDA and SFL
So far, I have reviewed the advantages and disadvantages of using rank in MDA
and traced the origin of the rank theory in SFL. Through the discussion, I have
shown the reasons why the argument for or against a rank-based MDA approach
does not necessarily hold the key to an understanding of the complexity of
MOLs and multimodal texts. In this section, I will summarize these arguments
and map out the implications of the rank debate for MDA and SFL research.
The first issue emerging from the rank debate concerns the validity of theo-
retical borrowings, that is, how and to what extent the linguistic categories can
be used in the description of MOLs and multimodal texts. In its development,
MDA theory has utilized a wide selection of SFL parameters, such as metafunc-
tions, stratification, systemic (paradigmatic) relations and rank; and occasion-
ally structure (e.g. Martin and Stenglin, 2007; O’Halloran, 2008). The borrowing
of systemic categories, however, has seldom been systematic. The presence or
absence of a parameter tends to be justified purely with its applicability to the
description of one particular type of MOL or multimodal text. The theoretical
consequences of the choice on the overall coherence of the theory are seldom
acknowledged. Linguistic categories, as Halliday (2002 [1992], p. 202) pointed
out, are ‘not defined individually but in relation to one another’. In this sense,
arguing for rank requires the total accountability of systems across metafunc-
tions at all ranks, or at least at the rank central to the meaning processing.
Arguing against rank, on the other hand, requires addressing the potential of
complex structures such as those involving embedding. The purpose-oriented
theoretical borrowing of SFL has so far been proven to be constructive for the
development of MDA theory, for it has enabled us to engage with a wide variety
of MOLs and multimodal texts. However, the merit owes itself to the extrava-
gance of SFL theory rather than the applications of the linguistic categories.
It seems to me that the flexibility in the theoretical adaptation might have ham-
pered the development of a coherent MDA approach that ensures objectivity
and high consistency both within the same analysis or across analyses. The ques-
tion that needs to be addressed is ultimately a methodological one – how valid
it is to borrow a set of theoretical categories without introducing the underlin-
ing principles?
The rank debate also forces us to re-examine the dynamic relations between
theory and data. In the history of MDA, its relationship with SFL theory remains
monolateral instead of dialogic. That is to say, to capture the complex nature of
multimodality, our approach is to exhaust the rich analytical tools SFL theory
offers. While the attempt fails, however, there is little feedback on the theory
itself. As biologists Goldenfeld and Woese (2007, p. 369) have eloquently
expressed, ‘the most fundamental patterns of scientific discovery is the revolu-
tion in thought that accompanies a new body of data’. In this sense, what our
explorations in MDA ultimately offer us is a chance to re-examine and renew
our understanding of human semiotic processes. As I have demonstrated ear-
lier, the issues raised in multimodal research are hardly new. Our attempts to
introduce the concept of rank into MDA reflect deeply our perception of text
structure as one based on constituency. To move beyond constituency, one solu-
tion would be to resume the unfinished SFL dialogue on dynamic modelling
initiated in the mid-1980s (e.g. Bateman, 1989; Berry, 1981; Martin, 1985;
Matthiessen, 1993; O’Donnell, 1990; Ravelli, 1995; Ventola, 1987, Zeng, 1996).
The final issue the rank debate brings up is in fact an even older one: the
issue of using language as metalanguage for describing language itself. The
dilemma of ‘language turned back on itself’ is accepted as inevitable in linguis-
tics. Dealing with MOLs, however, has highlighted the tension. There are at
least two issues that need to be urgently addressed: if the meaning of modalities
other than language can be translated into language, if non-constituency
based/dynamic nature of MOLs and multimodal texts can be captured by a
metalanguage which is essentially synoptic and constituency-based? If we accept
that language is a model for construing experience, which is manifested through
its structures; we should anticipate the possibilities that MOLs construe experi-
ence differently from language. In other words, the differences between lan-
guage, image and other MOLs lie both in their structures and meaning patterns.
In this sense, using language as a metalanguage puts more issues at risk than we
could have anticipated. We would, on the one hand, over-emphasize one type
of structure (meaning) – particulate – over the others. On the other hand, we
cannot avoid a verbal-centric interpretation of meaning patterns in MOLs and
multimodal texts. It seems to me that the biggest challenge lies in the search for
a new metalanguage, one that can capture the non-linear and dynamic nature
of multimodal discourse and human semiosis in general.
15.5 Data
The first golden age of cricket Curriculum Cooperation, 2005. The Le@rning
Federation leaning object L684.
Notes
1
I would like to express my appreciation to Dr Emilia Djonov and Dr Michele
Zappavigna for their valuable comments on an earlier draft of the paper. I am
also grateful to David Caldwell for proofreading the final draft and sharing his
extensive knowledge of cricket.
2
Ideational structure is referred to as representational structure in Kress and Van
Leeuwen (2006[1996]) and O’Toole (1994). It is also referred to as experiential
structure in O’Halloran (2004). For the purpose of this chapter, I label it as ide-
ational structure/metafunction; following Halliday and Matthiessen (1999; 2004),
since it concerns the construal of experiences.
3
In Halliday’s original paper, he proposed that one criterion for examining the
validity of a linguistic theory is to see ‘whether the theory can be falsified’, another
being if ‘it is worth making in the first place’ (Halliday, 2002[1966], p. 119). It is
not clear from this paper or his subsequent publications that if Halliday uses the
term ‘falsify’ in a strict Popperian sense (see Popper, 1934 [2002]). Here, I have
chosen to follow the original arguments of Halliday, since the concept of ‘falsifi-
ability’ is beyond the scope of this chapter and will not be relevant to my central
arguments. Whether Popper’s model is adequate for the type of linguistic research
discussed in this chapter is nevertheless open to debate.
4
There is the possibility that O’Toole has a different conceptualization of rank,
though it is not explicitly defined. However, the use of rank with other SFL param-
eters (metafunction) leads to the natural conclusion that it is comparable to
Halliday’s concept of rank.
References
Baldry, A. P. and Thibault, P. J. (2006). Multimodal Transcription and Text. London:
Equinox.
Bateman, J. A. (1989). ‘Dynamic systemic-functional grammar: a new frontier’. Word
40 (1–2), 263–87.
Bateman, J. A. (2008). Multimodality and Genre: A Foundation for the Systematic Analysis
of Multimodal Documents. Hampshire: Palgrave Macmillan.
Bednarek, M. and Martin, J. R. (eds) (2010). New Discourse on Language: Functional
Perspectives on Multimodality, Identity, and Affiliation. London: Continuum.
Berry, M. (1981). ‘Systemic linguistics and discourse analysis: a multi-layered approach
to exchange structure’, in M. Coulthard and M. Montgomery (eds), Studies in
Discourse Analysis. London: Routhledge and Kegan Paul, pp. 120–46.
Butt, D., Lukin, A. and Matthiessen, C. M. I. M. (2004). ‘Grammar – the first covert
operation of war’. Discourse and Society 5 (2–3), 267–90.
Caldwell, D. (in press). ‘Making metre mean: identity and affiliation in the rap
music of Kanye West’, in M. Bednarek and J. R. Martin (eds), New Discourse on
Language: Functional Perspectives on Multimodality, Identity, and Affiliation. London:
Continuum.
Cléirigh, C. (1998). ‘A selectionist model of the genesis of phonic texture: systemic
phonology & universal Darwinism’. Unpublished PhD thesis, University of Sydney,
Sydney.
Cloran, C. (1994). Rhetorical Units and Decontextualisation: An Enquiry into Some

Eggins, S. and Slade, D. (2005). Analyzing Casual Conversation (2nd edn). London;
New York: Cassel.
Goldenfeld, N. and Woese, C. (2007). ‘Biology’s next revolution’. Nature 445, 369.
Halliday, M. A. K. (2002: Reprinted in full 1966). ‘The concept of rank: a reply’,
in J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 1: On Grammar.
Halliday, M. A. K. (2002: Reprinted in full 1979). ‘Modes of meaning and modes of
expression: types of grammatical structure and their determination by different
semantic functions’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday:
Vol 1: On Grammar. London: Continuum, pp. 196–218.
Halliday, M. A. K. (2002: Reprinted in full 1992). ‘Systemic grammar and the con-
cept of a “science of language”’, in J. J. Webster (ed.), The collected works of
M.A.K. Halliday, Vol 2: Language and Linguistics, Volume 2. London: Continuum,
pp. 199–212.
Halliday, M. A. K. and Matthiessen, C. M. I. M. (1999). Construing Experience
Through Meaning: A Language-based Approach to Cognition. London: Cassell.
Grammar (3ed edn). London: Arnold.
Jewitt, J. (ed.) (2009). Handbook of Multimodal Analysis. London: Routledge.
Knight, N.K. (forthcoming). ‘The interpersonal semiotics of having a laugh’, in
S. Dreyfus, S. Hood & M. Stenglin (eds), Semiotic Margins. London: Continuum.
Kress, G. and Van Leeuwen, T. (2002). ‘Colour as a semiotic mode: notes for a
grammar of colour’. Visual Communication 1 (3), 343–68.
Kress, G. and Van Leeuwen, T. (2006). Reading Images: The Grammar of Visual Design
(2nd edn). London: Routledge.
Martin, J. R. (1985). ‘Process and text: two aspects of human semiosis’, in J. D. Benson
and W. S. Greaves (eds), Systemic Perspectives on Discourse: Selected Theoretical Papers
from the 9th International Systemic Workshop. Norwood, NJ: Ablex (Advances in Dis-
course processes 15), pp. 248–74.
Martin, J. R. (1996). ‘Types of structure: deconstructing notions of constituency
in clause and text’, in. E. H. Hovy and D. R. Scott (eds), Computational and
Conversational Discourse: Burning Issues – An Interdisciplinary Account. Heidelberg:
Springer, pp. 39–66.
Martin, J. R. (2000a). ‘Factoring out exchange: types of structure’, in M. Coulthard,
J. Cotterill and F. Rock (eds), Working with Dialogue. Tubingen: Niemeyer, pp. 19–40.
Martin, J. R. (2000b). ‘Beyond exchange: appraisal systems in English’, in S. Hunston
Discourse. Oxford: Oxford University Press, pp. 142–75.
Martin, J. R. and Stenglin, M. (2007). ‘Materialising reconciliation: negotiating
difference in a post-colonial exhibition’, in T. Royce and W. Bowcher (eds), New
Directions in the Analysis of Multimodal Discourse. Mahwah, NJ: Lawrence Erlbaum
Associates, pp. 215–38.
Martinec, R. (2001). ‘Interpersonal resources in action’. Semiotica 135 (1/4), 117–45.
Martinec, R. (2004). ‘Gestures which co-occur with speech as a systematic resource:

the realisation of experiential meaning in indexes’. Social Semiotics 14 (2),
193–213.
Martinec, R. (2005). ‘Topics in multimodality’, in R. Hasan, C. M. I. M. Matthiessen
and J. Webster (eds), Continuing Discourse on Language: A Functional Perspective.
London: Equinox, pp. 157–81.
Martinec, R. and Salway, A. (2005). ‘A system for image-text relations in new (and
old) media’. Visual Communication 4 (3), 337–71.
Martinec, R. and Van Leeuwen, T. (2009). The Language of New Media Design: Theory
and Practice. London and New York: Routledge.
Matthiessen, C. M. I. M. (1993). ‘Instantial systems and logogenesis’. Paper pre-
sented at the Third Chinese Systemic-Functional symposium, Hangzhou, June
17–20 1993. Unpublished ms, draft 15/ix/93.
Matthiessen, C. M. I. M. (2007a). ‘The multimodal page: a systemic functional
exploration’, in T. Royce and W. Bowcher (eds), New Directions in the Analysis of
Multimodal Discourse. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 1–62.
Norris, S. (2004). Analyzing Multimodal Interaction: A Methodological Framework.
London: Routledge.
O’ Donnell, M. (1990). ‘A dynamic model of exchange’. Word 41 (3), 293–327.
O’Halloran, K. (ed.) (2004). Multimodal Discourse Analysis: Systemic Functional
Perspectives. London: Continuum (Open Linguistics Series).
O’ Halloran, K. (2005). Mathematical Discourse: Language, Symbolism and Visual Images.
London and New York: Continuum.
O’Halloran, K. (2008). ‘Systemic functional-multimodal discourse analysis (SF-MDA):
constructing ideational meaning using language and visual imagery’. Visual
Communication 7 (4), 443–75.
O’Toole, M. (1994). The Language of Displayed Art. London: Leicester University Press.
Painter, C. (2008). ‘The role of colour in children’s picture books: choices in
ambience’, in L. Unsworth (ed.), Multimodal Semiotics and Multiliteracies Educa-
tion: Transdisciplinary Approaches to Research and Professional Practice. London:
Pike, K. L. (1982). Linguistics Concepts: An Introduction to Tagmemics. Lincoln: Univer-
sity of Nebraska Press.
Popper, K. (1934: Reprinted in full 2002). The Logic of Scientific Discovery.
London and New York: Routledge.
Ravelli, L. (1995). ‘A dynamic perspective: implications for metafunction interaction
and an understanding of theme’, in R. Hasan and P. Fries (eds), On Subject and
Theme: A Discourse Functional Perspective. Benjamins: Amsterdam, pp. 187–234.
Ravelli, L. (2006). Museum Texts: Communication Frameworks. London and New York:
Routledge.
Royce, T. (2007). ‘Intersemiotic complementarity: a framework for multimodal
discourse analysis’, in T. Royce and W. Bowcher (eds), New Directions in the
Analysis of Multimodal Discourse. Mahwah, NJ: Lawrence Erlbaum Associates,
pp. 63–109.
Royce, T. and Bowcher, W. (eds) (2007). New Directions in the Analysis of Multimodal
Discourse. Mahwah, NJ: Lawrence Erlbaum Associates.
Stenglin, M. K. (2004). ‘Packaging curiosities: towards a grammar of three-
dimensional space’. Unpublished PhD thesis, University of Sydney, Sydney.
Unsworth, L. (2007). ‘Multiliteracies and multimodal text analysis in classroom

work with children’s literature’, in T. Royce and W. Bowcher (eds), New Directions
in the Analysis of Multimodal Discourse. Mahwah, NJ: Lawrence Erlbaum Associates,
pp. 331–59.
Unsworth, L. (ed.) (2008). Multimodal Semiotics and Multiliteracies Education: Transdis-
ciplinary Approaches to Research and Professional Practice. London: Continuum.
Van Leeuwen, T. (1993). ‘Language and Representation: The Recontextualisation
of Participants, Activities and Reactions’. Unpublished PhD thesis, University of
Sydney, Sydney.
Van Leeuwen, T. (2006). ‘Typographic meaning’. Visual Communication 4 (2), 137–43.
Ventola, E. (1987). The Structure of Social Interaction: A Systemic Approach to the Semiotics
of Service Encounters. London: Frances Pinter.
Ventola, E., Charles, C., and Kaltenbacher, M. (eds) (2004). Perspectives on Multi-
modality. Amsterdam: John Benjamins.
Zappavigna, M. (2008). ‘Visualising instantiation: text visualisation techniques for
preserving logogenesis’, in S. Dreyfus, S. Hood and M. Stenglin (eds), Proceedings
of the Semiotic Margins Conference, University of Sydney, 2007. Australian Systemic
Functional Linguistics Association.
Zappavigna, M. (2010). ‘The coupling of gesture and phonology’, in M. Bednarek
and J. R. Martin (eds), New Discourse on Language: Functional Perspectives on
Multimodality, Identity, and Affiliation. London: Continuum.
Zeng, L. (1996). ‘Planning text in an integrated multilingual meaning space:
Theory and implementation’. Unpublished PhD thesis, University of Sydney,
Sydney.
Zhao, S. M. (2010). ‘Intersemiotic relations as logogenetic patterns: the time
factor in hypertext description’, in M. Bednarek and J. R. Martin (eds), New Dis-
course on Language: Functional Perspectives on Multimodality, Identity, and Affiliation.
London: Continuum.
Chapter 16
The Meaning of ‘Not’ Is Not in ‘Not’

Ruqaiya Hasan
16.1 Introduction
For human beings, the world certainly turns on the exchange of meaning.
Even a casual observation of daily life reveals the active participation of lan-
guage in a wide range of human social practices – which means that much of
our practical life is beholden to acts of meaning. As the discourse here1 deve-
lops (Section 16.2.2), it will become obvious that this foregrounding of mean-
ing is not designed to underplay the importance of ‘syntax’ or ‘expression’,2
but in a very real sense the raison d’être for syntax and expression is their role in
bringing about the ‘sens-ible’3 manifestation of meaning. We might argue quite
rightly that without them meaning could not be accessed, but by the same
token, without meaning the being of these phenomena – if they had any –
would have no bearing on human life – and certainly not on language: mean-
ing, literally, is the essence of language, on which more later.
Meaning in the popular sense of the word – that is, not necessarily linguistic
but by any means – has been central to all major aspects of human life. This was
true in the early stages of the evolution of the human species, and it is true
today when thanks to woefully limited ideas about progress we might be
approaching the end of that long drawn drama. Since the claim about the
centrality of meaning to human life may sound exaggerated, I will briefly intro-
duce some evolutionary and developmental perspectives based on research in
such branches of the human sciences as archaeological anthropology, neuro-
science, psychology, sociology and last and most pertinent, one particular
approach to the study of language, which treats linguistic meaning as the
essence of humanity, viz., systemic functional linguistics.
16.1.1 Meaning in social life: archaeological anthropology

The traditional view in narrating ‘The Human Ancestor’s Tale’ (with apologies
to Dawkins, 2005) had presented tool making as one of the most basic steps in
the evolution of the human species. However, today research in archaeological
anthropology (e.g. Noble and Davidson, 1996; Davidson, 2007) places social
interaction at the centre of this process, arguing that the production of tools
logically presupposes the existence of some form of communication for Homo
sapiens. With hindsight, the image of a group of early ‘ancestors’ crafting their
tools, each in his own ‘bubble’ of silence does strike one as somewhat absurd,
especially when evidence suggests that the tools were used collectively and
became, in fact, a commodity for exchange fairly early in our history. What was
being exchanged in the interactions was significant in two senses: it was signifi-
cant for the community’s survival through a direct management of the material
environment, and it was significant as a step in facilitating the development of
acts of meaning as an evolving mode of human existence. Whatever its modal-
ity, social exchange consists in message-ing, and the point of a message is its
meaning. The importance of meaning thus predates those anatomical changes
which, by facilitating auditory modes of meaning became a milestone in human
cultural evolution (Marwick, 2005) and which, as the precursor to the ‘meaning-
sound’ conjunction, have been long celebrated in linguistics as a crucial pro-
perty of human language. In the next few subsections, however, the concern
will be with the popular sense of meaning in communication irrespective of
the means by which it is effected: in the game of evolution language arrived
later than many other means.
16.1.2 Meaning making mind: neuroscience

Archaeological anthropology also provides cogent ground for discarding the
Cartesian mental myths as relics of the past; it suggests, instead, that the human
brain has evolved to its present state in response to a complex set of evolution-
ary pressures (Noble and Davidson, 1996; Davidson, 2003; 2005), in the midst
of which there had continued the typically human motif of action and interac-
tion, so that material action and communication moved in parallel; which
implies in turn that both the physical and the mental resources were simultane-
ously involved in the process of evolution. The emergence of symbolic behav-
iour and verbal language are milestones not only in the evolution of language,
but also in that of the human brain. This phylogenetic perspective on the
making of mind foregrounds the dialectic that holds social practices, acts of
meaning, and the physical and functional evolution of the brain in a mutually
evolving relation – a dialectic whereby the history of any one is part of the
history of the other two.
Recent research in neuroscience appears to support this evolutionary view.
The findings it offers concerning the trajectory of the brain’s structural and
functional development apply to the entire human species. As Homo sapiens,
we are inordinately proud of the faculties of our brain, treating them as
independent of its ‘physique’: the clear distinction made between the two has
thus given rise to the widespread duality of brain and mind, as two distinct
phenomena. Modern neuroscience explodes this myth: it describes mind, though
The Meaning of ‘Not’ Is Not in ‘Not’ 269
cautiously, as a ‘personalized brain’ (Greenfield, 1997; 2008). Attempting to

explain why human brains are not ‘robotic’ – why we do not act as clones of
each other – the researchers suggest that each individual brain is shaped by the
experiences it encounters in the living of life. For a variety of reasons, the latter
are not – indeed, cannot be – identical for every member, and the ways in which
different individuals encounter and ‘deal with’ the different experiences in
different ways creates what Jean Lave (1997) has called ‘habits of the mind’.
Thus each individual brain develops specific ways of being, doing and saying – a
mental disposition towards recognizing aspects of experience as relevant infor-
mation: it is these processes that we call ‘mind’. And a large part of the experi-
ence of living is the experience of ‘language-ing’: meaning thus spans the social
universe extending from a community’s lifestyle to its members’ ability to act as
social agents, a significant part of their identity.
16.1.3 Meaning, mind and culture: semiotic mediation

To ask how mind is shaped by experience is to ask with Bernstein (1990) how
the ‘outside’ becomes the ‘inside’. Gregory Bateson once humorously remarked
(1972) that inside the brain there are no monkeys or coconuts: experience,
per se, does not get transplanted in the brain/mind. On the basis of modern
research on the brain, and of our first-person experience of experiencing, we
may claim with some confidence that to be internalized all experience must be
both sens-ible and ‘intellig-ible’ (Russell, 1940); it must be sensuous, impinging
on (some of) the five senses, and, at the same time, the brain must somehow
‘make sense’ of it, must ‘absorb’ its essence, that which gives it a ‘point’ in the
life of the experiencing being. This is where we turn to Vygotsky’s concept of
‘semiotic mediation’. This is best seen as a ‘hinge concept’ capable of connect-
ing nature to nurture, phylogeny to ontogeny, community to individual and the
material to the semiotic.
To present a highly simplified account: archaeological anthropology is an
enquiry into the communal history of Homo sapiens, and neuroscience is
an exploration of the structure and functioning of the brain in the members
of this species; in both cases the orientation is phylogenetic; neither research is
concerned with the developmental trajectory of single members of the com-
munity, which must always display some variation. But the riddle of the transfor-
mation of some outside experience into the habits of particular ‘personalized
brains’ can be resolved only by a close look into the relationship of single
human organisms to the community to which they belong: in other words
the perspective has to be ontogenetic. Vygotsky (1978; Wertch, 1985a; 1985b)
solved this riddle with his concept of semiotic mediation. He postulated a con-
tinuity between the biogenetic foundation of the neonate’s activities and the
socio-genetic foundation of its activities as it moves through life becoming an
individual. Initially culture has little hand in creating or shaping these bioge-
netic activities. However, many of these activities are assigned some significance
by the caregiver who is always already acculturated: this is where sociogenesis

sets in. It is this assigning of significance by an already socialized individual that
forms the basis of continuity between the bio- and the sociogenetic, for indeed
the baby grows with the growing habit of meaning (Halliday, 1975; Hasan,
2001): little wonder none of us can remember when meaning first became
meaningful to us. In this way, Vygotsky in fact re-wrote the prefix ‘onto-’ as
‘socio-’ in the context of human development; there is very little that is ontoge-
netic in human behaviour, that did not begin as a communally recognized
social activity, whether material or semiotic. For each mind, the internalized
significance of these experiences is individuated because each history of experi-
ence is in some ways unique: what is being experienced is unique; the experi-
encing, internalizing brain is unique, thus each new internalization of sociogenetic
experience by an individual is at once irrevocably social, and indisputably individual.
A distinctive feature of Vygotsky’s notion of mediation was its open-ness to
various classes of experience. For Vygotsky, developmental psychology was not
about how nature designed the brain to function once for all, but rather about
how the naturally designed brain of a human being has the potential for extend-
ing itself in absorbing the relevant aspects of the universe presented socially.
Thus, for example, the infant absorbs the significance of the caregiver’s uptake
of a gesture, which links a material event to meaning; with the introduction of
a material tool, the worker’s earlier relationship to his work changes in grasping
the significance of the tool in the performance of that task; it is however with
the arrival of linguistic meanings that the character of semiotic mediation changes quali-
tatively: talk between participants of some activity does more than act as an
instrument – it encourages analysis; the newcomer to the school learns con-
cepts, their interrelations, forms of reasoning and other strategies, all mostly
through the abstract tool of language. This acts as the basis for the formation
of higher mental functions. Though there is much more to semiotic mediation,4
the above narrative indicates clearly the central place of meanings in creating
minds – and human minds5 are the true force behind the social agents whose
actions create, maintain and change human societies.
16.1.4 Bernstein and social variation: from interaction to ideology

There is a great deal of convergence between Vygotsky, Bernstein and Halliday
(Hasan, 2005a). Both Vygotsky and Bernstein take the social environment as
critical in shaping human development, assigning a central importance to
social interaction, and within that, centrally to language. However, in Vygotsky’s
work, society appears somewhat monochromatic, as if all types of social inter-
action were equally within reach of all social agents (Hasan, 2005a). With a
keen awareness of the systematic relationship between social structure, agents’
social positioning and forms of interaction, Bernstein’s research (1990; 2000)
attempted to show that experience of interaction for social agents was varied
from one social group to another; and that it was not simply a source of infor-
mation, it also defined the participant’s orders of relevance: different forms
of interaction produced different forms of consciousness. This was not a statement
about mental ‘development’ per se: it pointed to the social genesis of that men-
tal disposition which acts as the principle for selecting some particular sets of
values and beliefs and the principles of their legitimation in their immediate
community, which obviously vary across the various groups in a society. Thus
for Bernstein, orientations to meaning are ideological orientations which act
for and against cultural continuity/change, maintaining and fighting the per-
sistence of social hierarchy: this I take to be the essence of Bernstein’s ‘coding
orientation’.
16.1.5 Beyond the ‘things in heaven and earth’:

from protolanguage to linguistic meaning
Although in the work of both Vygotsky and Bernstein, linguistic meaning is
assigned an important role, it is Halliday who in his research (1975) focused on
a sociogenetic theory of not only the ontogenesis of language in individuals but
also what linguistic meaning does for the growing child6. Simplifying a good
deal, I present here the gist of his enquiry under two headings: (i) continuities
from material action to linguistic interaction (2004, p. 9) and (ii) the signifi-
cance of becoming a language user (1975; 1980). That the exchange of mean-
ings in the wider sense of the word predates linguistic meaning has been implied
throughout this discussion. Figure 16.1 represents Halliday’s view of the signifi-
cant steps in this journey from the material to the semiotic where very early
‘bodily actions (including . . . movements of the limbs and . . . movements
of the vocal organs) . . . get co-opted . . . as symbolic expressions’ (Halliday,
2004, p. 9).
The view presented by Halliday categorizes communication by reference to
their modes of meaning: (i) pre-semiotic move expressed as some bodily act,
for example, a yell – if the yell is seen as significant it is thanks to the receiver;
(ii) isolated symbol manifested as bodily action, but unmistakably7 vested with
significance; (iii) a primary semiotic system – Halliday (1973a; 1975) refers to
this as protolanguage8 – manifested as vocal symbols, each with a specific import;
Figure 16.1 Moving and meaning (source: Halliday, 2004, p. 9)

contextual evidence about the baby’s communicative intent is palpable, though

the expression, albeit vocal, is still ‘alien’ to the acculturated adult; there is
no syntax in the early stages, though the function of the message in context
is obvious to caregivers; and, finally, (iv) the onset of the higher order semiotic,
that is, language, where expression and meanings in the utterance come
increasingly to bear systematic relationship to the caregiver’s language. This
classification of the different kinds of acts of ‘meaning exchange’ foregrounds
on the one hand the enormously varied nature of that to which we refer as
‘meaning’ in popular usage, and on the other, the compatibility between the
evolutionary and the developmental trajectories.
It is not easy to describe quickly the significance of a child becoming a lan-
guage user: Halliday captures the enormity of this enterprise in ‘Three aspects
of children’s language learning: learning language, learning through language,
learning about language’ (1980). The appreciation of the multiple domains of
human experience captured within these three descriptive nominal groups
depends on how one conceptualizes language. With Halliday’s functional
approach the logical implication is that what language is and what language can
do for its speakers are not unrelated phenomena. So far as the potential of
language for meaning is concerned, there is no exaggeration in the claim that
it is limitless: it can go beyond what Hamlet referred to as the ‘things in heaven
and earth’, covering past, present and future, concrete and abstract, real and
imaginary: the limit of meaning in language is the limit of its speakers’ imagina-
tion. Not surprisingly, its contribution is central in shaping individuals from
their infancy – the nature of the universe they internalize, their ideological
orientations, their ways of being, doing and saying, the habits of their minds,
the points of departure for their imagination. Halliday’s functional approach
has enabled researchers to shed some light on how linguistic interaction is
transformed into ideological orientation through habitual exchanges of meaning
(Cloran, 1994; Williams, 1995; Hasan, 2009a). Vygotsky and Bernstein give us
a sociogenetic theory of the development of mind and mental disposition;
Halliday’s linguistics has the means of revealing the potential that, at least in
theory, is available to every language user. As the model of language as meaning
potential, it has the power to deconstruct the essence of any ideological orienta-
tion whatever – both as content and as process. We need to ask: What gives this
powerful potential to linguistic meanings for shaping human minds, and thus
human lives?
16.2 Meaning: Here, There and Everywhere
The word ‘meaning’ in its popular sense (especially as used in Sections 16.1.1–
16.1.3) may be paraphrased as ‘making sense’ or ‘deriving significance’ – in
short, ‘interpreting’: obviously, in this perspective, any phenomenon ‘has’ mean-
ing so long as it is open to interpretation. This open-ended view of meaning may
appear chaotic to those who like their data neat and tidy; nonetheless, in light
of first-person experience, it has to be granted that human beings do make
sense of an immense range of phenomena – for example, the house one lives
in, the garden that adds to its status – or not, as the case may be – the clothes
one wears, the fabric, the design; one’s comportment, aloof, reserved or out-
going; and a myriad of such phenomena – all ‘say something’ to those who
encounter them. Making sense seems to be an imperative of the brain: it is
impossible to state with certainty what, if anything, can not be interpreted; what
seems certain is that the brain is a naturally designed interpreter: so long as it is not
dozing, not drunk, not deranged, not dead, it must make a ‘reading’. I am
tempted to suggest that the habit in human beings to (un-/sub-)consciously
interpret the elements of their environment is biogenetic – perhaps a naturally
evolved disposition: the survival value of such a capacity to the species can
hardly be disputed.9
16.2.1 Semiology, multimodality and meaning

That meaning in the above sense is so ubiquitous is primarily an outcome of
how we define sign/symbol: if the sign/symbol ‘is’ simply ‘anything that, by
custom or convention, stands for something else’ (Noble and Davidson, 1996,
p. 5)10, then inevitably the entire world turns into a set of symbols, for the fact
is that things, persons, processes and what not, can all ‘stand for something
else’, and typically do. Meaning in this all embracing sense is a meaningful con-
cept, as argued above. But from the point of view of devising an integrated
framework for its description, it poses some interesting problems. To put it
mildly, it is a challenge to Saussure’s ‘science of semiology’,11 and by the same
token to the burgeoning trade of multimodality: with this conception of sign,
semiology as a ‘science’ (cf. Saussure, 1966, p. 16) would need to study all those
elements of the universe that have the potentiality of ‘standing for something
else’. To the best of my knowledge, the principles governing the identification
of such signs and the goal(s) of their study have so far remained rather unclear.
This line of argument opens a huge debate; and in the interest of ‘science’ it
must be conducted sometime. A careful pursuit of this kind would be a signifi-
cant contribution to the study of communication in culture. However, I pro-
pose to side-step these issues by restricting myself to linguistic meaning, in
order to throw some light on the source of its power.
As a preface to what follows, note that underlying the description and
identification of linguistic meaning is the weight of an entire theory of lan-
guage, but the various linguistic theories are not in agreement on what it
means to say ‘linguistic meaning’. This is true of even varieties of what is seen
as ‘the same’ theoretical framework, viz., Systemic Functional Linguistics. The
approach presented here is grounded in my interpretation of Halliday’s SFL
(henceforth, SFL).12
16.2.2 Concepts for meaning in language: (a) stratification

and realization
This is not the place for presenting details about the theory as a whole. How-
ever, we do need to introduce a few fundamental theoretical concepts, particu-
larly relevant to the explication of ‘meaning in language’. The first of these
concerns stratification, which logically leads us to realization. Figure 16.2 shows
how the two are related.
The inner ellipse in Figure 16.2, which is divided by a prominent horizontal
line into two halves, represents a language-internal view of stratification. Fol-
lowing Hjelmslev (1961), the top half of the ellipse is called content, the lower
half, expression: this is primary stratification with ‘minimal coding’; it presents
a picture of language as a symbolic system, as the term is used above (Section
16.2.1). Halliday (1961) and before him Firth (1957) see language as a multiple
coding system; as Figure 16.2 shows the primary strata are divided13 further
into two strata each14: content is reconceptualized as semantics and lexicogram-
mar –that is, (linguistic) meaning and wording, respectively – while the expres-
sion level is re-stratified as phonology and phonetics. It is important to note
here the abstractness of content, and the materiality of expression: both
meaning and wording per se are ‘invisible’ in the sense that they cannot be
‘accessed’ by any of the senses and are therefore not available for any mental
process without the intervention of phonology and phonetics which (like
orthography) are ‘visible’ in the sense that they impinge on the human senses.
The outer surround of the ellipse represents the communal environment of
Figure 16.2 Strata in a functional theory of language

language, namely context, which, despite being language-external, bears an

inherent relationship to language. For this reason it must form an integral part
of any adequate theory of language. In SFL it is taken to represent the fifth theoreti-
cal stratum. Figure 16.2 is iconic of the fact that context of culture bears a
stronger relationship to the two strata of meaning and wording, and a consider-
ably weaker one to phonology and phonetics.
Stratification separates out each aspect of the overall organization of the lin-
guistic theory, individuating each by virtue of the kind of abstractions each is
capable of describing: to give very simple examples, the application of true/
false or appropriate/inappropriate are relations valid only at the level of con-
text – there are no false or inappropriate meanings or wordings, much less
phonological or phonetic units; synonymy/antonymy are semantic relations;
the other strata have little to say about them; declarative/interrogative or hypo-
taxis/parataxis are categories at the level of lexicogrammar – the units at the
other strata do not contract such relations, despite the fact that we might talk
of someone as having a ‘declarative style’ – an idiomatic usage, not to be con-
fused with technical terminology; homophony is applicable to phonological
objects, not to wording; bi-labial/dental describe qualities of phonetic units.
Stratification, thus, represents a principle for the division of labour in the realm
of linguistic description. This is not to claim that there are no categories based
on inter-stratal relations, but simply to insist that each stratum is concerned
with a distinct order of abstraction.
Obviously, the separation of strata is an artefact of analysis. In actuality, lan-
guage always occurs in some social context, presenting itself as a seamless flow
of meaning-wording-sound, which typically functions in harmony with the social
occasion of talk. Realization (Halliday, 1992; Hasan, 1995; 1996; 2009b; 2009c;
Matthiessen, 2007; Butt, 2001; 2008) is that theoretical concept which specifies
the inherent connections among the linguistic strata whereby the perception of
their seamless unity is maintained: it is the most fundamental concept for exam-
ining inter-stratal relations. I have sometimes remarked (e.g. Hasan, 1996;
2009b) that realization is a hard worked concept: it has been made to do many
different kinds of thing, two of which are particularly relevant to the present
discussion. First, consider realization as it works in primary stratification: in
Figure 16.2, the bold horizontal line, dividing the ellipse into two halves, repre-
sents the Saussurean line of arbitrariness. Hjelmslev’s content and expression,
in fact, closely correspond to Saussure’s signified and signifier, respectively.
Saussure’s sign, a two-part entity, is a conjunction of the two – a conjunction
without any logical basis: the signified, an abstract phenomenon, accessible to
mental activity alone; the signifier, an acoustic shape with materiality, imping-
ing on the senses, makes the meaning it signals accessible to mental processing.
Saussure described the basis of their conjunction as arbitrary, rooted in conven-
tion. This is the relation that holds between the primary strata of content and
expression: units of expression ‘stand for’ units of content, or ‘signal’ them; to
sum up, the realization relationship of phonetics-phonology as two strata of

expression to semantics-lexicogrammar as two strata of content is characteristi-
cally arbitrary.
However, so far as the three strata above the stratum of phonology are con-
cerned, their mutual relationship is different. Halliday (1988) has referred to
the realization relation between lexicogrammar and semantics as natural. What
this means is not that nature has granted us a biological facility for the percep-
tion of specific pairings of semantic and lexicogrammatical units, but rather
that linguistic meaning is un-knowable without lexicogrammar; and one might
add: the process of meaning is impossible without context. This points to a
qualitatively different kind of work done by the realization relation across
context, meaning and wording – it is not arbitrary but dialectical, so that the
relationship logically locks the three together in such a way that the absence of
one from this triad would leave the other two impotent: the recognition of all
three working together is necessary for doing adequate linguistics. Based entirely
on communal convention, the arbitrary realization relation of phonology to
‘worded-meaning-in-context’ is quite unlike the dialectical realization relation
which is based in necessity – a necessity that arises from the internal logic of
language not by communal convention. If language is to function as system and
process ‘in the living of life’ (Firth, 1957), this internal logic must hold. Note
the double headed arrows linking the three upper strata of context, semantics
and lexicogrammar: they point to the bi-directionality of this dialectical realiza-
tion, which is motivated by the need to recognize two reciprocal perspectives
essential to the maintenance and evolution of language. The ‘downward’ move
from context to semantics to lexicogrammar is the encoding view, the ‘upward’
move ‘is’ the decoding view: the two indicate respectively the speaker and
addressee position vis-á-vis the ‘speech event’ (Firth, 1957). To put it more
simply, in producing an utterance, the issue is the speaker’s sense of the
relevant context which identifies the relevant semantic domains. This enables
the selection of meanings to be meant in this instance, and that leads to word-
ing choices15. No matter how problematic it may seem to subscribe to this view,
it would be far more problematic to reject it out of hand. It does not seem sub-
ject to doubt that, in the universe of semiotic activity, meaning comes ‘before’
wording in some sense of the word; let me, however, add immediately that in
this locution, ‘before’ cannot refer to some temporal phenomenon; witness,
however, the awareness of ‘slips of the tongue’, searching for ‘that word’ (‘which
word?’ ‘Well, the one whose meaning the speaker needs to encode here and
now!’); saying: ‘I don’t know what to say’, does not mean: ‘I happen to have
forgotten the formal structures of this language’, but simply that the situation
defies the possibility of choosing some appropriate meaning: the occasion is
‘breath-taking’. For the receiver, however, meaning cannot come first16, no mat-
ter how instantaneous its grasp is in everyday use of language: what allows the
addressee to get to the speaker’s meanings is the speaker’s wording17.
Realization works somewhat differently in the two directions. In the encod-

ing view, it is an activation of some possible choice at the next lower level: thus
in the production of an utterance, context activates meaning, meaning activates wording.
By contrast, in the reception of the utterance, realization is construal of the
relevant choice at the higher level: thus in decoding an utterance, the choice in word-
ing construes meaning, the choice in meaning construes context. It is relevant to note
here that in SFL, the definition criteria for a category are said to be located at
the level above it (Halliday, 1988): for example, it is the context of the utter-
ance ‘Join the navy and see the world’ that alerts us to the fact that despite
‘being’ worded as coordinated imperative clauses, the utterance is not a set
of orders; it is an advertisement in the form of exhortative invitation supple-
mented by a condition. By contrast, the lower level category furnishes recogni-
tion criteria for the category at the above stratum: thus the clause ‘Do it now!’
is an imperative clause which overwhelmingly functions as recognition crite-
rion for – that is, construes – the semantic category demand; goods/services,
unless there is good reason for doing otherwise, as in the above advertisement.
Meta-redundancy (Lemke, 1984) often invoked in the discussion of meaning
and metaphor is a technical label that refers to precisely the relations just para-
phrased. The fact that realization between strata does not display a 1:1 relation-
ship is a critical feature ensuring the semogenic power of language (further
discussion in Section 16.3.1). Further, if the relationship between the catego-
ries on two strata were universally biunique, there would be no ground for the
recognition of two distinct strata (Halliday, 1988). By the same token, in an
important respect, it would reduce the semiotic system of language to the same
level of semological complexity as that found in the systems of traffic lights.
Hjelmslev (1961) suggested wisely that a necessary condition for recognizing
separate strata is their non-conformality.
Although the difference between the arbitrary and dialectical realization
relation has not been made explicit in SFL, Firth’s (1957, p. 33) distinction
between major and minor ‘functions’ of meaning interestingly does suggest
a division between the strata which seems to bear a close resemblance to the
suggestion I have made here: Firth treats phonetics as the minor function18
of meaning, and ‘lexical, morphological, and syntactical (. . .), and the function
of a complete locution in the context of situation . . .’ as the major functions of
meaning.19 Firth does not offer an explicit justification for his division between
the major and minor functions of meaning. My discussion of the special kind of
solidary relation between context, meaning and wording, clearly assigns the
triad an important role in the process of meaning, while the arbitrary relation
between the triad and the strata at the expression level is a good reason for
thinking of the latter as not having the same importance for making meaning:
their function is, rather, the relay of worded meaning.
By focusing on the triad as central to the meaning process, I do not mean to
imply that expression is altogether irrelevant to the process of meaning; in fact,
both in production (activation) and in reception (construal), the expression

strata of phonology and phonetics act as the point where the sign’s materiality
works as the initial gateway for entry to the levels of wording and meaning which would
otherwise remain inaccessible. In this sense they do have a function in the
encoding and decoding of meaning, but that function is, as pointed out by
Firth, a minor one. For, if we leave prosodic phonology (Halliday and Greaves,
2008) out of this discussion, it is obvious that segmental phonology and phonet-
ics play little or no part in the construal of wording (hence of meaning) in the
sense in which the term ‘construal’ is being used here. The expression strata
simply signal the presence of a sign unit but even this signalling is not always
unambiguous.20 Their function is thus not content-forming; it is simply content-
expressing. This is the price paid for arbitrary realization. However, the uncer-
tainty thus created is constrained by the second property of sign identified by
Saussure, that is, linearity. By supporting the syntagm, linearity gives us the
invaluable co-text or Firth’s serial context; not many complete speech events
consist of one morpheme, leave aside one phoneme. In normal use of language
phonemes occur in each other’s company, so do syllables, morphemes, words,
clauses, questions and responses; they thus create the ‘linguistic context’ or
become the co-text for each other. In doing real linguistics, isolated linguistic
categories21 – along with the man from Mars – should be banished as irrelevant
to linguistic explanation.
16.2.3 Concepts for meaning in language: (b) system and instance

The langue v. parole dichotomy is as old in linguistics as the modern discipline
itself. The two have been viewed as irreconcilable; thus ‘langue’ – a code or
system – was taken as ‘a fixed set of rules’ – sometimes essentially innate, and
‘parole’, its instance, as an unpredictable flow of signs, mercurial, and governed
only by the individual’s will; the former synoptic, the latter dynamic. However,
the problem with language as system is that it needs to be both stable/synoptic and chang-
ing/dynamic: it must remain in use to stay alive,22 and that is possible only if
it changes along with the changing conditions for its use (Butt, 2004; Hasan,
2009b). The SFL concept of instantiation resolved this seeming paradox:
Halliday (1987; 1999) argued that a system and its instance are ‘the same thing’
seen from different perspectives; their behaviour displays deep complementar-
ity (Halliday, 2008). While the possibility of speakers moving beyond the sys-
tem, whether by variation or innovation, is empirically evident, it is equally
obvious that instances allow speakers some view of the system. Nor is there any
doubt that the system is the resource by which parole as instance is interpreted
(Hasan, 1984; 2009b; 2009c). It is in and by the production of the instance that
both conformity and innovation play their significant roles in maintaining and
renewing the system. If synchronic variation and diachronic change in language
are related phenomena (Labov, 1972; Fasold and Schiffrin, 1989; McMahon,
1994), then we may have to concede that the view of system as synoptic and of
Figure 16.3 Two systems and their relationships: realization and instantiation
process as dynamic needs careful re-consideration. This debate is also relevant

to semantic change, which does not always go hand in hand with lexicogram-
matical change, as evident from the history of items such as host, silly, condescen-
sion and others where the primary categorization of the items does not change
but the meaning does. Neither is it clear whether such semantic changes in
meaning are responsive to some contextual changes; if so what they are, and
what is the nature of the process of such responsive changes. Figure 16.3 cap-
tures Halliday’s (1999, p. 8) view of the complex relationships between systems
and instances.
The different stages of the development of this relationship are discussed in
Halliday (1987; 1999; 2002; 2008), Hasan (1984; 1995; 2009a), O’Donnell
(1994), Matthiessen and Nesbitt (1996), and Matthiessen (2007). In terms of
current sociology, reflection on Figure 16.3 leads to a theory of the social reproduc-
tion of language (Hasan, 2009a): the social permeates both text as instance,
which realizes its context of situation, and language as system which realizes
(aspects of) its context of culture, thus supporting the hypothesis of linguistic
functionality as an inherent aspect of the organization of content –that is, the
strata of meaning and wording (Halliday, 1973b; Hasan, 1995; 2005b; 2009b).
The SFL view of complementarity between system and instance bears relevance
to a good many important issues; however, I will single out only the following
two for discussion here: (i) the description of the semantic stratum and (ii) vali-
dation of linguistic description.
16.2.4 Concepts for meaning in language: (c) representing

the potential
Saussure’s Course in General Linguistics is an evolving discourse: having
described the sign as a two-part entity – the signified and the signifier – he is at
pains in Part Two to describe the complex nature of each part, insisting that the
identity and value of each unit of signified and signifier is forged by its relations
to other units. Characteristically, Saussure produces a binary system of relations:
one based on combination or syntagm ‘supported by linearity’ (ibid, p. 123)
later also referred to as the syntagmatic bond, the other based on opposition or
association, later referred to as the associative bond. Halliday’s chain and choice
(1963) resonate well with this Saussurean pair, though it is Hjelmslev’s syntagm
and paradigm (1961) that are the prevalent terms for referring to the relations
essential for defining the identity and value of a sign. Since units must be identi-
fied at every stratum of language, it follows that the syntagmatic and paradig-
matic axes apply to all language internal strata.
Linearity is a feature of discourse: instances are linear, thus relations postu-
lated between the units of parole are in praesentia. By contrast, paradigmatic
relations are supported by associative bonds (relations of differentiation): these
relations, not evident in units within a discourse, are in absentia. Pertaining
to a whole paradigm, the associative bonds are closer to the system aspect of
language. Since it is system that acts as the potential, the idea of choice is
clearly critical to its description. In SFL, the potential for each linguistic stratum is
represented in the form of system networks,23 which are in effect an account of the
oppositional/ associative relations that are significant to the value and identity
of the categories of that stratal unit which acts as the point of origin for that
particular system network. Thus the system network of MOOD specifies cate-
gories of the unit clause at the lexicogrammatical stratum, and so on. Each tra-
versal of the system that is valid in terms of the logic of that system network
constitutes a selection expression (SE) of the set of choices made from it: the
SEs instantiate the system network, and are themselves realized as structures
of some kind. Again much has been written in SFL about system networks24, so
no further details will be presented here.
I begin with the assumption that given the status of semantics as a stratum
of language, its units too can be described paradigmatically to represent their
potential, as demonstrated by Halliday (1973b), and later by Hasan (1983),
Cloran (1994) and Williams (1995). However much one might disagree with
the basic premises of componential semantics (Katz and Fodor, 1963; Bierwisch,
1970; Leech, 1974), there is, in that literature too, suggestive evidence of para-
digmatic organization at the semantic level. When, in some of his writing, Firth
(1957) dissolves semantic description into a series of contextualizations of
some specific instance, the implication is that there is nothing left to describe;
there is, in other words, no stratum of meaning so that from Firth’s context
we move directly into ‘lexical, morphological and syntactic’ categories: in my
view the suggestion lacks validity. First, in accepting this suggestion, we would
have to reject the postulate of a stratum of linguistic meaning; for indeed a
stratum without any units is an anomaly; and the recognition of units entails
some existing or feasible description of their paradigmatic relations calling for
some system networks. However, if there are no defined categories of meaning,
talking about Firth’s ‘semantic’ phenomena is fraught with problems; it would

be impossible, for example, to say which meanings are at risk where, because
we have already abolished by fiat the need for a language of description
(Bernstein, 2000) for semantic categories. One might suggest that semantics is
interpretative: thus it will make use of semantically sensitive lexicogrammatical
categories whose value is established by their place in some system networks.
Against this too, there are some objections: as soon as every semantic unit – but
what is that? – can be unequivocally paired with some delicate lexicogrammati-
cal category, the justification for postulating both the level of semantics and of
lexicogrammar will disappear. And even assuming that we can produce a seman-
tically sensitive grammar which (a) is not in a 1:1 relation with the semantic
categories, and (b) is also sensitive to the semantic contribution of the syntag-
matic environment (cf. discussion of ‘Join the navy and see the world’), we are
still left with problems: how to refer to that category of meaning which we
are attempting to match? If the semantic category is itself ‘notional’, that is, has
no basis in any principled description, then is our semantically sensitive gram-
mar open to being treated as a ‘notional grammar’? In any event, grammar
must come to an end somewhere! It is difficult to imagine a lexicogrammar
whose units are capable of describing larger semantic units such as text and
rhetorical unit (Cloran, 1994): neither texts nor rhetorical units can be described
in terms of the largest lexicogrammatical unit, viz., the clause complex. Even if
these problems can be surmounted, one basic issue remains unresolved: How
desirable is that modelling of language which would allow the interpretation of
utterances without enabling their production? For the fact remains that for the
speaker the perception of meanings relevant to the context of discourse is the
starting point for an utterance. I conclude that an adequate linguistic theory
must allow the possibility of both activation and construal – of production and
of interpretation – of meaning.
But is there a unit scale at the semantic stratum? Hasan (1996) suggested a
scale of four units: proceeding from the largest to the smallest, they are text, rhe-
torical unit (RU), message and message component (MC)25. Like all postulated
unit scales on the different linguistic strata, the units here too are in a constitu-
ency relation: thus a text consists of one or more than one RUs, an RU consists of
one or more than one message, a message consists of one or more than one MC.
The best paradigmatically described semantic units are message (Hasan, 1983;
Williams, 1995) and RU (Cloran, 1994): both have been represented as a set of
system networks at the level of semantics (Hasan et al., 2007). Figure 16.4 pres-
ents a fragment of message semantics which describes semantic options in
demanding goods/services: the realization statements for the entire network
require too much space (for some indicative statements, Hasan, 2009a; Chapter 7:
Table 1: A–C; p. 284). The options in the network activate some choice(s) from
the MOOD system network at the lexicogrammatical stratum. Further discussion
of options from Figure 16.4 will be presented in following sections.
282
Figure 16.4 Simplified system network of options in demanding goods/services
16.2.5 Concepts for meaning in language: (d) validating description

To sum up the Saussurean position: the identity and value of any unit at any
linguistic stratum is the function of the paradigmatic and syntagmatic relations
the unit enters into. This is what it means to say, as Saussure did, that there are
no pre-existing categories: it is the point of view from which the facts of lan-
guage are described which creates the categories we recognize. This makes the
activity of description appear rather subjective; but if the description itself is
governed by some explicit principles and while following those principles yields
a picture of language that has observational and explanatory adequacy, then
the point of view is proved to be one which may be treated as a valid modelling
of that language26 for continuing further into its description. We therefore
need a device which explicitly states the conditions for treating a description as
valid. Halliday’s trinocular perspective is one such device: this involves consult-
ing the relations of the category under description from three relevant perspectives, view-
ing relations at the stratum above –that is, what the category realizes; viewing
relations at the stratum below –that is, what realizes the category; and viewing the
relations at the same stratum – that is, where the category under description is
positioned with regard to other categories of its kind; these indicate its paradig-
matic relations (its associative bonds). I offer here an illustration of how these
perspectives function by reference to the semantic option27 suggestive (see
Figure 16.4: path a1:b1 . . .) which is one possible element of meaning in the
unit message: progressive: demanding; goods-&-services: the message type
identified construes contextual acts of commanding, requesting, persuading,
cajoling, humouring someone into doing something the speaker specifies.
Looking above the semantic level, the contextual ‘reality’ which the option
suggestive realizes – or signifies – is an element of the tenor of discourse: the
speaker’s rhetorical stance, whereby s/he appears to place the addressee
under some pressure without using physical or symbolic violence so as to get
the addressee to do something by representing herself/ himself as complicit in
that action (e.g. saying to someone who is about to carry something fragile: ‘lets
be careful in lifting this’, or a mother instructing her child: ‘lets not tease the
pussy cat, darling!’). At least two questions need to be asked: Do speakers natu-
rally utter messages of this kind or is it simply a figment of the linguist’s imagi-
nation? Second, is there any evidence that on listening to a command type
utterance with the semantic feature suggestive, the addressee would ratify the
above contextual interpretation of this feature by some action or locution?
A negative response to either of these questions should send the analyst back to
the workbench, but for obvious reasons, such a response to the first question is
more troublesome than to the second. The former would invoke doubt as to
the existence of the option itself, the latter simply points to an error in descrip-
tion. On the assumption made by many well-known linguists (among them,
Firth, 1957) that a difference in linguistic ‘form’ equals a difference in meaning,
all that is needed is to consult corpora and observe human interaction in many
different contexts to establish the interpretation of the semantic element that
differentiates commands such as ‘lets be careful in lifting this’ v. ‘do be careful
in lifting this’ v. ‘would you be careful in lifting this’, v. ‘I’d like you to be careful
in lifting this’, ‘better be careful in lifting this’, and so on.
Looking at the same level, the most immediate category in an oppositional rela-
tion to the option suggestive, is nonsuggestive, which refers to the contex-
tual fact of the projected action being presented as the responsibility of the
addressee alone albeit under instruction, as in ‘lift this carefully’, ‘better lift
it carefully’, ‘could you lift it carefully’, and so on. This semantic option too
is realized along with most of the others described by Figure 16.4 by some
choice(s) in the MOOD system network. But looking at the same stratum can-
not be reduced to simply noting its difference to the other term(s) within the
same individual system28. Two points need to be clearly stated with regard to
viewing the facts at the same stratum: first, what this means, in effect, is examin-
ing the over-arching system network the options of which act as the systemic
description of the unit that served as its point of origin, for example, in this case
message progressive as opposed to punctuative (Hasan, 1996). There are
certainly other units on the semantic stratum (Section 16.2.4), but for checking
on the validity of the description of some element of a message, the categories
of the unit message itself are more relevant than those of a lower or higher unit.
At the same time, there are many different message types: one way of identify-
ing a primary type is by focusing on the least delicate oppositional features. In
Figure 16.4, these primary oppositional systems of options are placed within the
first right facing open brace. Although we are concerned here with the message
type demand; goods-&-services (the primary systemic options in Figure 16.4),
more specifically, the option suggestive applies to any message type one of
whose primary semantic choices is goods-&-services: more simply the giving
OR demanding of goods-&-services is the environment in which the choice between
SUGGESTIVE v. NON-SUGGESTIVE has significance. Thus the system suggestive v. non-
suggestive will also form part of the systemic description of messages function-
ing as offers and invitations whose primary options are give; goods-&-services
(Hasan, 1986), and in this network too, the terms will refer to the same contex-
tual tenor-related facts. Second, the value of suggestive/non-suggestive is
determined by their positioning vis-a-vis the other terms within the network as
a whole: for example it is significant that although the options may be freely
co-selected with the systemic options assertive v. consultative they are not
freely co-selected when it comes to the options that depend on exhortative;
and the more delicate options in the system network are not available to
those messages which have the feature suggestive. Thus we do not find mes-
sages such as ‘we’d like us to be quiet’ (speaker-oriented: desiderative) and
options dependent on the feature addressee-oriented are logically impossi-
ble with the option suggestive.
Table 16.1 Lexicogrammatical realization statements for some semantic options

Semantic options Realization statements in terms of lexicogrammar
demand; goods-&-services enter MOOD system network

non-exhortative indicative
assertive declarative
consultative interrogative
exhortative imperative
suggestive insert Subject
Subject preselects Thing
non-suggestive Subject implied/present refers to addressee
suggestive; non-exhortative Thing at Subject preselects item we
suggestive; exhortative imperative: inclusive
Thing at Subject preselects item lets
Finally, looking at the stratum below, that is, at lexicogrammar, reveals how
the semantic option(s) are realized. Simplifying somewhat, the realization state-
ments may take different shapes depending on the logic of the network. The
system suggestive/ non-suggestive is concurrent with two others, so the
realization statements will differ according to what terms are co-selected. At
message rank, two kinds of realizational statement are typical: (i) choice of an
option from a clause rank lexicogrammatical system network, and/or (ii) state-
ment of pre-selection(s) at some clausal element including its category identifi-
cation and/or realization. Table 16.1 illustrates some cases with a few semantic
options from Figure 16.4; the options are in column 1 of the table, and the
realizations in column 2.
Choices in the system network at the stratum of lexicogrammar entail the
formation of structural information. Thus to say that something is realized as
declarative is to say that a structure needs to be formed with the elements Subject
and Finite in that order, represented formulaically as S^F. The default realiza-
tion of element Subject is by a nominal group; that of Finite is by an auxiliary
verb functioning either as primary tense (e.g. ‘is, was, will be’) or as a modal
auxiliary (e.g. ‘can, may, would, should, have to, must . . .’). The pre-selecting
realization is used when the option can be realized by some unique element of
structure and/or linguistic item. A realization statement such as ‘Subject prese-
lects Thing’ can be rephrased as ‘Subject must be realized as a nominal group
which “has” the element Thing in it’; in other words an elliptical nominal group
such as in ‘those two’ is not allowed.29 The preselecting realization statement
can also call for a unique item or specify the selection of some item from a para-
digm with a particular feature, for example, ‘Thing at Subject preselects item
we’ may be worded as ‘a clause realizing these semantic elements must have the
element Subject and Subject must be the item we’, as for example in ‘we have to
be very quiet’ whose semantic description is represented in the SE below30:
[demand;goods/services:nonexhortative:assertive;suggestive:affable;necessity]
Strictly speaking, naming unique items such as ‘you/we’ is a short cut: for
example, underlying ‘we’, are the word rank lexicogrammatical options for the
category pronoun as shown below (after Halliday and Hasan, 1976, p. 44):
[person:interactant:speaker:inclusive]
and it is this lexicogrammatical ‘reality’ that is arbitrarily signalled by the sound

structure represented in writing as ‘we’. Of course the realizational path should
be pursued right through to the expression level, but enough has been said to
establish that the description of suggestive as presented here is explicit, and
likely to be valid in matching the ‘facts’. But how about explanatory adequacy?31
To me it seems that explanatory adequacy in the description of language
calls for at least three kinds of evidence: (i) global evidence that the language
internal description is not only adequate for one variety of some one état de
langue but also offers principled path(s) within the theoretical framework for
granting the potential for variation; at the same time, there is the possibility of
productive movement across the many varieties which correlate with spatio-
temporal phenomena as both time and space ‘reshape’ while language responds
to the many social demands in the living of life in society; (ii) evidence of
exotropic connections whereby the modelling and description of language is
such as to entertain the findings of modern research about the interconnec-
tions between language, culture and consciousness: this chapter began with
a description of the ways in which language has been central to the evolution
of persons and cultures; a theory with explanatory adequacy should be able to
show what aspects of language are instrumental in playing this part, and how
this comes about;32 and (iii) local evidence: description of language explaining
the significance of the instance in the linguistic systems and its relation to con-
text of culture and situation. The discussion here has pertained largely to this
third aspect.
16.3 The Meaning of Grammar and

the Description of Meaning
In the description of the semantic feature suggestive as presented above, the
trinocular perspective focuses primarily on the paradigmatic relations of that
category; thus the inter-stratal relations between semantics and lexicogrammar
are observed largely in terms of systemic choices. If context has not participated
in this celebration of the paradigmatic here, this is only because SFL so far does
not have any developed contextual networks easily accessible for such use. Note,
however, Saussure’s suggestion that the identity and value of a unit is the func-
tion of the relations the unit enters into both paradigmatically and syntagmati-
cally. Of course, some syntagms were mentioned above, for example, in the
realization statement regarding the option declarative, but they were realizationally
entailed and so form part of the syntax. Significantly Saussure adds that there is
more to syntagm than syntax (ibid, p. 123, note 5). This makes sense: syntax is
unit-internal, and even the largest lexicogrammatical unit has a limit, recursion
notwithstanding. The syntagmatic axis is not bound by any such constraints –
especially if ‘syntagm’ is viewed simply as a flow of signs within some specifiable
social environment, it would represent the textual environment for those
patterns/units one might choose to examine. This is not the place to develop a
critique of the predilection in linguistics for peering simply inside the bounds
of a single unit isolating it from all else to find ‘objective’ evidence of the valid-
ity of some proposed description. Firth pointed out half a century ago that
words are known by the company they keep; perhaps this generalization should
be extended to most linguistic units; the trick is to establish the principles of
consistency in deciding what counts as ‘company’.
I pointed out in Section 16.2.2 that there can be no 1:1 realization relation
between linguistic strata. Ample proof of this will be found in the very few real-
ization statements presented in Table 16.1. For example, imperative clause is
generally regarded as that lexicogrammatical category which realizes demand;
goods-&-services, that is, a ‘proposal’ of some kind. Considered as a probabi-
listic statement, this is very likely the default position. Nonetheless, neither is
every proposal an imperative, nor is every imperative a proposal. And this state-
ment could be extended to all primary categories of MOOD. Thus a demand
for goods and services may be assertive, realized as declarative (e.g. ‘you must
not shout’) or consultative, realized as interrogative (e.g. ‘could you do my shoe
laces, Mum?’). Faced with such descriptions, the suggestion has been to treat
these as some kind of interpersonal grammatical metaphor. This is not the place
to go into this debate; the problems in this solution are many and certainly
some of them appear to threaten the concept of grammatical metaphor itself,
which has proved so useful in the analysis of discourse (Halliday and Martin,
1993). In any event, what seems to clinch the matter is the fact that this solution
may not be possible for all lexicogrammatical categories. The category negative
polarity is one such. In the following sections, I will discuss what semantic options
the category of negative polarity is able to realize in just one semantic context,
namely, demand; information. First, a brief description of the category negative.
16.3.1 ‘Not’ as a lexicogrammatical category

The system of polarity, with the options positive v. negative, is well known in SFL
as a part of the interpersonal description of the English clause. It is thus posi-
tioned within the MOOD system network, with the clause itself having a func-
tion in role exchange. The functional element Mood is central to the description
of the categories used for role exchange (Halliday, 1994): potentially Mood
expands into the elements Subject and Finite, the former realized as some nomi-
nal category, the latter by the first auxiliary of a finite verbal group, which will
either select for tense or modality. What happens actually by way of expansion –
for example, are both elements there? if so, in what order of sequence? if not,
which may be assumed where and on what evidence? – all this is realizationally
relevant to the identification of the categories of role exchange. Mood is the
element that according to Halliday (1994, p. 71) is ‘tossed back and forth’ in a
series of conversational moves, for example, ‘he is, is he’? The Subject is the
element on which the validity of the rhetorical move rests, while the Finite ties
down each such move to some specific point in space and time by reference to
the utterance act (for discussion, Halliday, 1994, p. 68ff; Halliday and Matthiessen,
2004, p. 106ff). The system of polarity is considered ‘an essential concomitant
of finiteness’ (Halliday and Matthiessen, 2004, p. 116) since ‘for something
to be arguable it has to be specified for polarity: either it is so or it isn’t so’
(Halliday, 1994, p. 75). Figure 16.5 presents a simplified system network for
MOOD with ‘cryptic’ realization statements.
16.3.2 Options in polarity: from grammar to meaning

In Figure 16.5, the realization of the option negative is not stated fully: ‘not/n’t’
are certainly the most frequent realizations of negative polarity but there are
quite a few other items, such as ‘neither/nor’, ‘never’, and in some contexts
even items such as ‘hardly, scarcely, seldom’, and so on, though they have not
been thought of as realizing negative polarity.33 However these latter items
construe some meaning that ‘modifies’ the negation: in other words, they go
beyond the default meaning of negative polarity. But what exactly is that mean-
ing? In their descriptions, both Halliday and Matthiessen – indeed, most SFL
practitioners, including myself – not only show the value and identity of the
lexicogrammatical unit under description, they also give some statement about
its ‘function’. These concern the default construal of meaning by that unit, and
form part of what ‘looking above’ means in the trinocular perspective. With
regard to polarity, Matthiessen (1995, p. 487) suggests that
the choice between positive and negative . . . is clearly interpersonal . . . the

speaker in principle chooses negative if s/he judges that s/he has to cancel
what the addressee believes or will do. The choice of polarity thus depends
on the speaker maintaining and revising a model of the relationship between
himself/herself and the addressee: what is the semiotic distance between
them in their construal of experience and their readiness to act? – what is the
balance between consensus and conflict?
According to Halliday (1994, p. 89):
In a proposition, the meaning of the positive and negative poles is asserting

and denying: positive ‘it is so’, negative ‘it isn’t so’. . . . In a proposal, the
Figure 16.5 A simplified system network of options in MOOD
meaning of the positive and negative poles is prescribing and proscribing:

positive ‘do it’, negative ‘don’t do it’.
The default meanings of positive/negative polarity suggested by Halliday here

are also echoed by most dictionaries, for example, Chambers, Collins Cobuild,
Macquarie and others. The description certainly captures the meaning of posi-
tive/negative polarity in the context of the default realization of proposition
and proposal. These meanings are actually category specific; thus denial is
the meaning construed by negative polarity with statements; prescribing and
prohibiting are specific to commands. However, the accounts by the grammar-

ians and the lexicologists make two outstanding omissions, viz., neither is there
an interpretation of negative in demanding information, that is, in questions,
nor in giving goods/services, that is, offers. But of course there too, the
choice of polarity makes a difference to the meaning of the realizing clause.
In the following sections I shall discuss the function of negative polarity in
demand; information, that is to say, in question. Time and space will not allow
similar review of giving goods/services, that is, making offers or in demand
goods/services, that is, the command family.
16.3.3 The uses of not/n’t in demanding information

Table 16.2 presents the functions of polarity suggested above by Halliday together
with one naturally occurring example of each category. To this account I have
added an interpretation for both negative interrogatives and imperatives: the
former is typically a question with an attitude and the latter ceases to be an offer.
As Table 16.2 shows, interrogatives with negative polarity construe the seman-
tic feature assumptive. Contextually the feature is related to the tenor of dis-
course, referring specifically to the speaker’s stance to her/his own question
and her/his expectation of what the answer could reasonably be. Here are three
examples from naturally occurring dialogues which substantiate this point:
1: Mother: didn’t you see me go out

Karen: no!
Mother: you must be blind!
2: Mother: wasn’t that a big fishy?
Pete: yeah
Mother: it was a whopper!
3: Mother: you aren’t tired?
Donna: nope
Mother: what! After all that running around
Table 16.2 The meaning of polarity with some primary MOOD options in English
Proposition Positive Negative
demand meaning: asking a question meaning: assumption re response

did you see me go out? didn’t you see me go out?
information
give meaning: telling meaning: rejecting some thesis
the water’s getting too cold no, it isn’t for me
Proposal Positive Negative
demand meaning: prescribing meaning: proscribing
do as you’re told! don’t tell me ‘no’ all the time!
goods/services
give meaning: offer meaning: restriction
have another piece of toast! hey, don’t eat up all those nuts!
Thanks to the textual syntagm, we can be quite certain that in each case the
mother thought she ‘knew’ what her addressee’s mental state should be, thus
revealing one subtle aspect of her social relation to her child. In the Macquarie
empirical studies of sociolinguistic variation in semantic orientation, the fea-
ture assumptive showed up statistically as a highly significant element of the
sociolinguistic variable (Hasan, 2009a). This is an impressive result in view of
the fact that in a corpus based study Halliday and James (1993) demonstrated a
skewed probability (9:1) for positive v. negative polarity choice; in their study
covering over 1.5 million cases of finite clauses, they found 87.6 per cent occur-
rences of positive and only 12.4 per cent of negative, which included ‘never’,
‘seldom’ and other such items as well.
Example 3 offered above is an instance of the category ‘question3’ in Halliday
and Matthiessen (2000, p. 22; for discussion, Section 16.4); and its inclusion
here indicates that not all questions are realized as interrogative, whatever
their polarity choice. At the same time it should be noted that negative polarity
does not always construe the feature assumptive in all kinds of questions.
Figure 16.6 presents a simplified system network of options in asking English
questions, while Table 16.3 shows the functions of not/n’t in the semantic envi-
ronment of demand; information along with lexicogrammatical information.
Table 16.3 is best examined in three parts by reference to the three systemic
options in column 2 (A, B and C). Every entry in the table systemically ‘assumes’
the choices progressive: demand; information (see Figure 16.6). The fea-
ture verify (column 2(A)) is the entry point for the systemic choice between
reassure v. probe (column 3); verify is itself in systemic opposition to enquire
(column 2 (B)) the entry point for this system being the selection of the feature
confirm (column 1), the realization of which calls for an indicative clause. For
each example, column 5 provides a lexicogrammatical selection expression
from the MOOD system network, which counts as the realization of the seman-
tic selection expression stated across columns 1 to 3. It is clear from this table
progressive
Figure 16.6 Simplified system network of options in demanding information

292
Table 16.3 The meaning of ‘not’ in questions: semantic and grammatical views
simplified semantic SEs simplified grammatical SEs
(i) you were sad, weren’t you? declarative: tagged: reversed

reassure (ii) you don’t like this T-shirt, do you? declarative: tagged: reversed
verify (A)
probe (iii) you left it on the train did you? declarative: tagged: constant
(iv) you don’t like sugar, don’t you? declarative: tagged: constant
confirm
ask: nonassumptive (v) did you see me go out? interrogative: polar; positive
ask: assumptive (vi) wasn’t that a big fishy? interrogative: polar; negative
enquire (B)
check: nonassumptive (vii) was it a big fish? declarative: non-tagged: positive; Tone 2
check: assumptive (viii) you aren’t hungry? declarative: non-tagged: negative; Tone 2
explain: nonassumptive (ix) why did you say ‘look’? interrogative: non-polar: R-interrogate; positive
apprize precise (C)
explain: assumptive (x) why aren’t you hungry? interrogative: non-polar: R-interrogate; negative
that any message with the feature verify would have a declarative: tagged clause
as its lexicogrammatical realization. The semantic option reassure is realized
by the MOOD features tagged: reversed, irrespective of the location of the nega-
tive; thus in (i) the Tag is negative, in (ii) the declarative is. The semantic option
probe is realized by the MOOD features tagged: constant, which for most variet-
ies of English would imply a zero instance instead of (iv) in column 4 – however,
(iv) is a fairly common occurrence in informal Australian English, especially
where speakers are not concerned about the danger of being perceived as
speaking a sub-standard dialect. Note that the negative in these two environ-
ments appears to be the converse of assumptive. If the feature assumptive
implies that the speaker already knows what the ‘reasonable’ response is, the
features reassure and probe imply that the speaker is in need of verifying
something. So what is the difference between reassure and probe? It is subtle
but nonetheless important as an indicator of interpersonal relationships; mes-
sages with the feature reassure may be paraphrased as follows: ‘I think some-
thing is the case, and I believe you do, too; is that so?’, whereas those with the
option probe are actually probing for ‘correct’ information: so, for example
(iii) may be paraphrased as: ‘I understand you left it on the train. Is that what
happened?’. We can see why questions with the feature reassure would be
relevant to what Bernstein (1971) called ‘sympathetic circularity’ which is a
form of interaction only among people with intimate relationship – as an aspect
of tenor relation, it has been described as construing minimal social distance
between the interactants (Hasan, 1973; 1985); by contrast, messages with the
feature probe could not have this contextual significance; the relationship here
is not as close as in the previous case.
The option enquire (B) is the entry point for two simultaneous systems, ask
v check and also non-assumptive v assumptive (see Figure 16.6). The choice
of negative polarity in a clause realizing enquire will always construe the seman-
tic feature assumptive. We note that ask is realized by an interrogative clause
and check by declarative + Tone 2. But what difference does this make to their
semantic character? A message with the feature ask is a yes/no question with-
out any strings attached – what is sometimes referred to as a ‘straight’ question;
check, on the other hand, is not a ‘crooked’ question – it is simply one with
additional information about some attitude on the part of the speaker regarding
that question: this may be surprise, disappointment, doubt, enthusiasm . . . –
the exact nature is realized by the choice in the system of key (Halliday and
Matthiessen, 2004; Halliday and Greaves, 2008).
This leaves us with the option precise (C in column 2). As Figure 16.6 shows,
the option apprize is in systemic contrast to confirm: unlike the latter it does
not construe a yes/no-question, but what is known as a wh-question: it thus
seeks some specific element of information, and is lexicogrammatically realized
as a non-polar clause. In questions which have the feature precise, the query
point is explicitly realized by the preselection of a wh— word as Theme, such as
‘why, who, what, when, where’ etc. The option precise is the entry point for a
choice between explain and specify. The feature explain is realized by an

interrogative clause in which ‘why/what for’ would be preselected as Theme. As
Figure 16.6 shows, the option explain is one of the entry points into the system
whose choices are assumptive and non-assumptive (note in Figure 16.6 the
disjunct entry condition, one of which leads from explain to the systemic
choice between assumptive and non-assumptive, while the other entry condi-
tions is enquire. Further details concerning the realization of other options
in Figure 16.6 will be found in Hasan, 2009a; Hasan et al., 2007; Cloran, 1994;
Williams, 1995). The feature R-interrogative mentioned in the realization of
explain (last two lines of column 3) stands for Residue-interrogative; in a clause
with this feature some Residue element forms the query point, as in ‘what did
you eat’?, ‘when/where did you go?’, ‘what did you do?’ and so on. This feature
is in systemic contrast with S-interrogative (see Figure 16.5 for this system) which
stands for Subject-interrogative; in a clause with this feature wh-element con-
flates with Subject as, for example, in ‘who/what/which was it?’. As is obvious
from the examples, the structural realization of S-interrogative is distinct from
that of R-interrogative: the structure of S-interrogative is Subject/Wh—ˆFˆPred
as in ‘who said that?’ (which resembles a declarative except for its wh— ele-
ment) while that of R-interrogative is Adjunct or Complement/wh—ˆFˆSˆPred
as in ‘when did he say that?’ or ‘what did he say?’, which are ‘typical’ of the
structure associated with interrogative, that is, F^S.
The most relevant characteristic of messages with the feature specify is their
‘resistance’ to the selection of negation, as Table 16.4 attempts to show. It is not
that the selection of negative polarity in clause types that realize messages with
the semantic option specify is necessarily ‘ungrammatical’, simply that the
right context for them is not easy to imagine, and in many cases it seems to shift
the rhetorical stance of the message. Compare, for example, ‘what didn’t you
do then?’ with ‘what didn’t you do to stop that kid from smoking!’ It seems
appropriate to suggest that the syntactic ‘conditions’ on the selection of nega-
tive polarity in such interrogatives could benefit by the kind of semantic research
Table 16.4 The meaning of ‘not’ with demand information . . . apprize: specify
Semantic options Positive polarity Negative polarity
Global What’s going on? What’s not going on?

What happened? What didn’t happen?
Particpt: circs Where have you been? Where haven’t you been?
When did you visit Rome? When didn’t you visit Rome?
With whom did you go there? With whom didn’t you go there?
What were you talking about? What weren’t you talking about?
Particpt: event What did you do then? What didn’t you do then?
Particpt: actant Who was calling you? Who wasn’t calling you?
What did you eat? What didn’t you eat?
underlying networks such as presented here in Figures 16.4 and 16.6. Both
pertain to the interpersonal metafunction and are realized by some feature of
the system of mood, but the meanings pertaining to all four metafunctions can
be described in a similar way (e.g. Hasan, 1983).
16.3.4 The meaning of ‘not’ is not in ‘not’

The above account of the meaning of ‘not’ in English is both condensed and
incomplete. For example, time does not allow a discussion of the choice of
negative polarity in the context of demanding or giving goods and services.
Reversing the meaning construal in many types of commands, negative polarity
in offers is more often than not polite, precisely because it insists as in ‘do have
some more, won’t you?’. But it seems clear on the basis of the evidence offered
here that categories of wording are typically multivalent. Their semantic iden-
tity and value become evident only on an examination of the range of their
paradigmatic and syntagmatic relations: this is what I have attempted to show by
a case study of the meanings of ‘not/n’t’ in the above discussion. This particu-
lar category of wording is not exceptional in having such a varied set of mean-
ings: on the basis of the semantics of lexicogrammatical categories such as
declarative, interrogative and imperative, we may safely conclude that the mean-
ings of a category of wording are not in that category itself; they take particular
shape in the specific environment in which the wording occurs, and signifi-
cantly, this environment is most parsimoniously identified by reference to the
semantic level. To extend a Firthian claim, the meaning of not only lexical
categories but most probably of all linguistic categories is known by the com-
pany the category keeps.
16.4 Concluding Remarks
I accept without reservation the claim that it is impossible to mean any mean-
ings without lexicogrammar: both making meaning and reading meaning
depend to a very large extent on lexicogrammar. This, however, does not imply
that the only approach to meaning has to be ‘from below’, concerned only
with how sets of lexicogrammatical categories are to be interpreted. True, that
a metafunctional lexicogrammar will be semantically oriented: this follows
logically from the genesis of metafunctions in contextually situated discourse
(Halliday, 1973b; Hasan, 2005b; 2009a); but reflection on the acts of speaking
and ‘hearing’ will suggest that the two outermost strata of language are in a
systematic relation to these two aspects of the activity of languaging. To present
a full account of meanings, we need to know not only how lexicogrammar con-
strues meanings but also what meanings are construable in a given language
variety, when, and by whom, and how they become accessible to the listener.
If the trinocular perspective on the validity of semantic description links con-

text to lexicogrammar via semantics, treating the latter as an interface between
the two, this clearly implies that the point of origin for the speaker’s meaning
is not specifiable by reference to lexicogrammar, much less phonology or pho-
netics: the origin is found in the speaker’s perception of the occasion of talk,
which entrains a complex of social factors, a meeting of mental dispositions
typically shaped by the culture that is instantiated in that context of situation,
whose ‘reading’ is the identifier of the meanings at risk, and for the speaker to
get to this point is to be half way into the lexicogrammar. To account both for
the speaker’s choice of meanings and the addressee’s interpretation of speaker’s
wordings both the construal and the activation of meaning are equally impor-
tant issues for a functional linguistics. From this perspective, one question that
demands immediate attention is what criteria are to be used for identifying
that point in the descriptive enterprise where it is maximally productive to draw
the line between a ‘most delicate lexicogrammar’ and a systemic semantics
capable of representing the potential for meaning.
In the course of lexicogrammatical description, we often encounter state-
ments regarding a category’s default meaning. These, however, cannot be
treated as more than excursions into semantics; I say ‘excursions’ because such
statements typically do not – and cannot – state the complete meaning potential
of any lexicogrammatical unit34 for reasons presented above. The non-default
meanings of a category are many and just as important as the default ones;
I hope the discussion of ‘not’ has indicated that looking up into semantics
from lexicogrammar provides a different view into the meanings of some cate-
gory from that which is presented when the main concern is with categories of
semantic systems, because it is here that one asks: What possibilities of meaning
are available for realizing some feature of the context of discourse: the focus is
on ‘what possible meanings are pertinent to this semantic unit’, not ‘what are
the meanings of this category of wording’. This is a subtle and important differ-
ence. For example, in SFL following Halliday (e.g. 1966–8) the systemic choices
of MOOD have been widely regarded as construing speech role exchange
and certainly they do that, but as the above discussion has shown options from
this system network are capable of doing far more than just speech role
exchange: they act as a rich resource for creating, maintaining and changing
social distance between the interactants by varying the nuances of a speech act
thus revealing subtle aspects of speaker attitude to the addressee(s).
Theoretically, SFL is committed to the idea of language as an ever-renewing
system: innovation, change, developments of different kinds are phenomena
inherent in language use. In accounting for linguistic change, both diachronic
linguistics and dominant sociolinguistics have largely been concerned with the
expression level. By contrast, Halliday’s concept of semogenesis has developed
in the context of the description of wording and meaning (Halliday, 1985), that
is, in reflection on the content level: an over-arching concept, embracing differ-
ent dimensions of human history (Halliday and Matthiessen, 2000), semogenesis
refers to essential processes whereby the dynamic nature of language as system

is able to be maintained. In their illuminating discussion of the process of semo-
genesis with specific reference to the level of content, Halliday and Matthiessen
(2000, pp. 21–2) draw attention to ‘the dissociation of associated features in the
wording’ whereby over time the ‘map’ of wording and meaning in English has
been changed. The two associated features of wording they consider are those
from the system of MOOD and of KEY: the Mood of an English interrogative
is ‘known by’ the Mood configuration F^S with Rising Tone; the default inter-
pretation of the pattern is demanding information. The dissociation of the
features of Mood and Key and their ‘re-configuration’ expands the total mean-
ing-wording potential of the language, giving the speaker three different ways
of saying and meaning by way of asking questions:
Here again one meaning has been replaced by three: we now have (say) ques-
tion1 interrogative x rising tone; question2 interrogative x falling tone;
and question3 declarative x rising tone, e.g., is she cóming? Is she còming?
She’s cóming? (Halliday and Matthiessen, 2000, p. 22)
Semantic networks of the type presented here appear to be ideally suited to

describe the paradigmatic relations of the different semantic categories as well
as their lexicogrammatical realizations which treat systems of MOOD and of
KEY as simultaneous, thus allowing the possibilities of dissociating those choices
of the two which represent their default association at some given état de langue
while permitting their innovative reconfigurations. Note also that it is not easy
to find a rational basis for insisting on the one hand that the lexicogrammatical
patterns F^S/Tone 2, F^S/Tone1 and S^F/Tone 2 should be accommodated in
their full detail within the description of that stratum while also maintaining
that the expansion of Mood F^S must be treated as the only pattern capable of
construing the ‘meaning proper’ of demanding information. As I see it, there
are only three options: (i) subscribe to the absence of bi-uniqueness, implying
that semantic functions such as demanding information may be construed by
more than one lexicogrammatical pattern, all of which are paradigmatically
related but distinct from each other in some way; (ii) insist that the ideal lan-
guage system is bi-unique in respect to meaning-wording relation, which poses
serious problems in accounting for empirical data which clearly contradict that
position; or (iii) wherever bi-uniqueness is demonstrated to be absent, explain
the fact away as some kind of grammatical metaphor, implying that grammati-
cal metaphor is something other than a part of lexicogrammatical description.
Accepting any of these options has a payback; in the context of interpersonal
role-related meanings, I have considered the first as the best solution: it stays
close to the observed facts of language use, and leaves the possibility of system-
atic variation open, a desirable characteristic of language as dynamic system.
The outstanding characteristic of the primary modes of meaning – that is,
those modes of meaning which form the object of enquiry in kinesics and
linguistics – is that they have no other function in human life except that of
enabling exchanges of meaning. Facial, physical and vocal gestures preceded
that modality of meaning which we call language: the non-linguistic modes of
meaning were central to the life of our ancestors, as linguistic meanings are to
that of ours. With evolving, developing and differentiating cultures, human life
today is infinitely more complex, discourse is more diverse, communications,
more complex. Although ‘body language’ remains in use – in fact, cannot be
eschewed – and although secondary modes of meaning such as images, graphs,
tables, and so on make considerable contribution to some categories of discourse,
there is no exaggeration in the claim that today it is verbal language that acts as
the foundation of our complex social existence. Linguistic meaning appears to
surpass all other modes of meaning – primary or secondary – in its efficacy35 pre-
cisely because of the limitless-seeming possibilities of subtle re-configurations of
vast but limited variables. It is interesting to ask: How can semantic description
and representation capture this dynamic quality of linguistic meaning?
Notes
1
This chapter is a revised version of the talk I presented at the Free Linguistics
Conference 2008 convened by Ahmar Mahboob and Naomi K. Knight on 11–12
October at the University of Sydney.
2
I use ‘syntax’ and ‘expression’ because these terms can be applied to a large
variety of ‘signing’/’symbolic’ phenomena; by contrast, ‘wording’ and ‘sound’ are
restricted to language.
3
I use here Russell’s (1940) distinction between the ‘sens-ible’ and the ‘intellig-
ible’, the former refers to sensing acts and the latter to acts of intellection.
4
The literature on semiotic mediation is vast, and opinions vary on the interpreta-
tion of the expression ‘semiotic mediation’. For some discussion see Lucy, 1987;
Mertz and Parmentier, 1991; Wertsch, 1985b; 1997. On Vygotsky in an SFL per-
spective, see Hasan, 2005a.
5
‘Meaning’ is thus coming to acquire another meaning, namely, ‘the representa-
tion / trace of experience in the human brain’.
6
Of course Vygotsky talked about the relationship between language and the
development of higher mental functions in individuals, but Vygotsky was chiefly
concerned with specialized concepts dear to official pedagogy (Hasan, 2005a).
Halliday’s concern is with the growing child as a social agent living and developing
in a society, who needs higher mental functions for coherently handling the logic
of both the ideological and the interpersonal.
7
Although we may attribute communicative intent to such symbols, it may not be
possible to offer objective criteria for the claim.
8
Halliday’s use of protolanguage is qualitatively different from its use in formalistic
linguistics, such as Bickerton’s (1990).
9
From this point of view, a review of experiments in psychology concerning the abil-
ity to recall will most probably provide support in the finding that a meaningful
series is significantly easier for humans to ‘internalize’ than a ‘meaningless’ one:
to the digital machine meaningfulness makes no difference.
10
It is convenient to quote this often voiced ‘definition’ of sign/symbol, but in fair-
ness to Noble and Davidson (1996) I must point out that their subsequent
discussions of the linguistic sign go far beyond the claim quoted here.
11
Saussure’s linguistic ‘sign’ is in fact much more carefully defined; it is perhaps
the textual organization of this intricate discourse that tempts infelicitous
interpretations.
12
From now on in this chapter, SFL will refer to Halliday’s SFL as interpreted by
me, without any implication that Halliday is necessarily in agreement everywhere
with my interpretation. For further discussion, see Halliday, 1996; Halliday and
Matthiessen (2004, Chapter 1).
13
Hjelmslev (1961) recognized more delicate distinctions in each of his primary
strata: his ‘content substance’ and ‘content form’ approximate Halliday’s ‘seman-
tics’ and ‘lexicogrammar’, as also ‘expression form’ and ‘expression substance’
approximate ‘phonology’ and ‘phonetics’; there are however some significant
differences (for some discussion, Hasan, 1995).
14
To be more precise, up till quite recently, expression was seen in SFL as mono-
stratal, with phonology subsuming phonetics. Recent developments (Halliday
and Matthiessen, 2004; Halliday and Greaves, 2008) have led to the recognition
of a separate stratum of phonetics.
15
This is what gives us the licence to say that registers are defined by the ‘meanings
at risk’ in some context.
16
Let me emphasize again that despite the use of these ‘serial/sequential’ expres-
sions, I am not suggesting at all that the units of the various strata appear in
temporal sequence. I am tempted to cite Firth (1957, p. 147) on this issue:
The utterance happens in time. The stream of speech with all its items
integrated unrolls itself . . . on the time track of occurrence. But the systemic
abstraction which we isolate in language systems are not limited by the time
track dimension of the utterances from which they are taken. The statement
of the systems when we talk or write about them have their own time track, since they
are speech events (last emphasis mine, RH).
17
Though our concept of discourse in context is capable of handling this complex-
ity (Hasan, 2005a; 2005b), in the interest of saving space, we assume here that the
occasion of talk does not involve semantic distance (Hasan, 2004).
18
Sadly phonology does not figure in this list, but by reference to his co-text, we
may conclude that Firth recognizes the contribution of prosodic phonology to
meaning and is referring here simply to the minor function of segmental phono-
logy and phonetics (for some supporting discussion by Firth, see 1957, p. 192) .
19
I agree with the spirit of this division but take issue (Section 16.2.3) with the
‘absence’ of semantics itself from the list: although Firth is not consistent in this
practice, often in his writing, semantics is dissolved into a ‘comprehensive descrip-
tion of a speech event’, which seems to be a move lacking in validity.
20
Space does not allow discussion of examples, but the potential of segmental pho-
nology and phonetics as a recognition criterion for categories of wording is
highly varied from near 1:1 conventional association with strong lexical items to
near nil with ‘fully grammatical’ ones. One is safe in venturing a semantic gloss
on ‘honesty’ without looking for its co-text or context, but it would be foolhardy
to suggest one single gloss for such phonetic units as the suffix ‘–s’ in ‘walks’, or
‘-ing’ as in ‘painting’ or even for the item ‘not’ as I propose to show in Section
16.3, and ‘not’ is not alone – we have to think of words such as ‘see, turn, give,
have, can’ and many others. Contra formalistic grammarians, fully grammatical
items do make a contribution to the meaning of the utterance by virtue of realiz-
ing a grammatical relation, though the nature of this relation can only be
determined by what is going on up above in worded meaning in context.
21
Always excepting the bright linguist’s imagination, who when it suits him can
turn the exception into a rule.
22
In fact this may be a feature of all social systems; numerals do not get ‘archaic’
but words, and social practices do.
23
I do not wish to imply that the description of the potential of each unit in terms
of such paradigmatic relations is anywhere near complete for each stratum. Such
description in SFL is most developed for lexicogrammar; much less so for the
strata above or below it. Further, as I see it, this is one of the points where models
such as Fawcett’s (2000) or Martin’s (1992) differ from Halliday’s SFL. For the
former there are no systems at any other stratum than that of what he calls seman-
tics; for the latter, either there is no system network applicable to semantics as a
whole, or it is to be represented by taxonomies of already identified semantic
units.
24
For example, to name a very few: Fawcett, 1980; 1988; 2000; Halliday, 1961; 1966;
1969; 1973a; 1975; 1979; Hasan, 1983; 1987; 1989; 1996; 2009; Martin, 1992;
Matthiessen, 1995; 2007.
25
A fifth unit lower than MC would be seme, with categories such as entity, process,
quality, and so on. However until some description is attempted, it is going to
be difficult to decide whether this unit is needed at the semantic level: it is just
possible that lexis as delicate grammar can describe such units more economi-
cally without residual problems for semantics.
26
This would neither be claimed as ‘God’s Truth’ nor as a ‘Hocus Pocus’ theory
of language: simply a pro-tem ‘real picture of language’, offering an account of
some unit that may be said to possesses ‘descriptive adequacy’, until proven
wrong.
27
Semantic options are presented in small capitals to distinguish them from the
ordinary and non-technical uses of the same orthographic shape.
28
I make a consistent distinction between system and system network: a system net-
work is a logically related set of two or more individual systems. Individual systems
are viewed as ‘forming’ a system network as soon as two or more individual sys-
tems are logically related by concurrence and/or dependence. The complexity
of the system network increases as the relations of dependence and/or concur-
rence multiply.
29
Subject ellipsis may occur but only if an option explicitly calls for it.
30
The semantic description has two options not discussed before: affable which
contrasts with unctuous. Compare ‘we have to be very quiet’ (affable) with
‘we have to be very quiet, don’t we?’ (unctuous); and necessity as opposed to
possibility as opposed to prediction. These options are realized by choices in
the system of modality.
31
Note the absence of ‘descriptive adequacy’: the reason is simple. Descriptive
adequacy is what is being established by showing that the description has obser-
vational and explanatory adequacy.
32
This clearly leads us into the issue of ‘appliable linguistics’ (Halliday, 2005) – a
linguistic theory that is able to explain the efficacy of language in human life.
33
However, see Halliday and James (1993). Hardly, scarcely and other such items
do not easily co-occur with ‘not/n’t’ (? ‘He was not hardly happy’) though it is
fine to say ‘he was hardly not happy’ with two ‘negatives’ making a positive (= ‘he
was certainly happy’).
34
It seems likely that we can go furthest within grammar perhaps in ‘lexis as most
delicate grammar’ (Hasan, 1987b; Matthiessen, 1995; Tucker, 1998) but dis-
courses are not a series of lexical items, and encountered in a syntagm, even
lexical meaning is likely to display ‘shifts’ from its default meaning.
35
In fact it might be better to talk about ‘effectivity’ rather than ‘efficacy’; the latter,
with its ‘purr’ connotations gives the impression that the power of language
is benign. This is not necessarily the case: its power is neutral as to creation or
destruction; it can accomplish both equally effectively.
References
Bateson, G. (1972). Steps to an Ecology of Mind. New York: Ballantine Books.
Bernstein, B. (1971). Class, Codes and Control, Volume 1: Theoretical Studies towards a
Sociology of Language. London: Routledge & Kegan Paul.
Bernstein, B. (1990). The Structuring of Pedagogic Discourse, Volume IV: Class, Codes and
Control. London: Routledge.
(revised edn). Oxford: Rowman & Littlefield.
Bickerton, D. (1990). Language and Species. Chicago: University of Chicago Press.
Bierwisch, M. (1970). ‘Semantics’, in J. Lyons (ed.), New Horizons in Linguistics.
Harmondsworth: Penguin, pp. 166–84.
Butt, D. G. (2001). ‘Firth, Halliday and the development of systemic functional
grammar’, in E. F. K. Koerner (ed.), An International Handbook on the Evolution
of the Study of Language from the Beginning to the Present. Berlin: Walter de Gruyter,
pp. 1806–38.
Butt, D. G. (2004). ‘How our meanings change: school contexts and semantic evolu-
tion’, in G. Williams and A. Lukin (eds), The Development of Language: Functional
Perspectives on Species and Individuals. London: Continuum.
Butt, D. G. (2008). ‘On the robustness of realization systems’, in J. J. Webster (ed.),
Meaning in Context: Implementing Intelligent Applications of Language Studies. London:
Continuum.
Cloran, C. (1994). Rhetorical Units and Decontextualisation: An Enquiry into Some Rela-
tions of Context, Meaning and Grammar. Monographs in systemic linguistics, No: 6.
University of Nottingham: Department of Linguistics.
Davidson, I. (2003). ‘The archaeological evidence of language origin: states of arts’,
in M. H. Christiansen and S. Kirby (eds), Language Evolution. Cambridge: Cam-
bridge University Press, pp. 140–57.
Davidson, I. (2005). ‘Beyond mysticism?: Review of Jackendoff, R. (2002)’. Linguis-
tics and the Human Sciences 1 (2), 337–345.
Davidson, I. (2007). ‘Scales and balances: archaeology, cultural heritage and sustain-
ability’. Proceedings of Academy of the Humanities (Annual Lecture 2007), 113–43.
Dawkins, R. (2005). The Ancestor’s Tale: A Pilgrimage to the Dawn of Life. London:
Phoenix.
Fasold, R. W. and Schiffrin, D. (eds) (1989). Language Change and Variation.
Amsterdam: Benjamins.
Fawcett, R. P. (1980). Cognitive Linguistics and Social Interaction: Towards an Integrated
Model of a Systemic Functional Grammar and the Other Components of an Interacting
Mind. Heidelberg: Julius Groos & Exeter University.
Fawcett, R. P. (1988). ‘What makes a good system network good?: Four pairs of
concepts for such evaluations’, in J. D. Benson and W. S. Greaves (eds), Systemic
Functional Approaches to Discourse: Selected Papers from the 12th International Systemic
Functional Workshop. Norwood, NJ: Ablex, pp. 1–27.
Fawcett, R. P. (2000). A Theory of Syntax for Systemic Functional Linguistics. Amsterdam:
Benjamins.
Firth, J. R. (1957). Papers in General Linguistics 1934–1951. London: Oxford Univer-
sity Press.
Greenfield, S. (1997). The Human Brain: A Guided Tour. London: Weidenfield and
Nicolson.
Greenfield, S. (2008). The Quest for Identity in the 21st Century. London: Scepre.
Halliday, M. A. K. (1961). ‘Categories of the theory of grammar’. Word, 17.3; 241–92.
Reprinted in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 1: On
Grammar. London: Continuum.
Halliday, M. A. K. (1963: Reprinted in full 2002). ‘Class in relation to the axes of
chain and choice’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday,
Vol 1: On Grammar. London: Continuum.
Halliday, M. A. K. (1966–8: Reprinted in full 2005). ‘Notes on transitivity and theme
in English, Parts 1–3’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday,
Vol 7: Studies in English language. London: Continuum.
Halliday, M. A. K. (1969: Reprinted in full 2005). ‘Options and functions in the
English clause’, in J. J. Webster (ed.), The collected works of M. A. K. Halliday, Vol 7:
Studies in English Language. London: Continuum.
Halliday, M. A. K. (1973a). Explorations in the Functions of Language. London:
Arnold.
Halliday, M. A. K. (1973b: Reprinted in full 2003). ‘Towards a sociological seman-
tics’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 3: Language
and Linguistics. London: Continuum.
Halliday, M. A. K. (1975: Reprinted in full 2004). ‘Learning How to Mean: explora-
tions in the development of language’, in J. J. Webster (ed.), The Collected Works of
M. A. K. Halliday, Vol 4: The language of Early Childhood. London: Continuum.
Halliday, M. A. K. (1979: Reprinted in full 2002). ‘Modes of meaning and modes of
expression: types of grammatical structures and their determination by different
semantic functions’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday,
Vol 1: On Grammar. London: Continuum.
Halliday, M. A. K. (1980: Reprinted in full 2004). ‘Three aspects of children’s lan-
guage learning: learning language, learning through language, learning about
language’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 4: The
Language of Early Childhood. London: Continuum.
Halliday, M. A. K. (1985). An Introduction to Functional Grammar. London: Arnold.
Halliday, M. A. K. (1987: Reprinted in full 2003). ‘Language and the order of

nature’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, (1987, 1999)
Vol 3: Language and Linguistics. London: Continuum.
Halliday, M. A. K. (1988: Reprinted in full 2002). ‘On the ineffability of grammatical
categories’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 1:
On Grammar. London: Continuum.
Halliday, M. A. K. (1992: Reprinted in full 2002). ‘How do you mean?’ in J. J. Webster
(ed.), The Collected Works of M. A. K. Halliday, Vol 1: On Grammar. London:
Continuum.
Halliday, M. A. K. (1994). An Introduction to Functional Grammar (2nd revised edn).
London: Arnold.
Halliday, M. A. K. (1996: Reprinted in full 2002). ‘On grammar and grammatics’, in
J. J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol 1: On Grammar.
London: Continuum.
Halliday, M. A. K. (1999: Reprinted in full 2007). ‘The notion of “context” in lan-
guage education’, in J. J. Webster (ed.), The Collected Works of M. A. K. Halliday,
Vol 9: Language and Education. London: Continuum.
Halliday, M. A. K. (2002). ‘Computing meanings: some reflections on past experience
and present prospects’, in G. Huang and Z. Wang (eds), Discourse and Language
Functions. Shanghai: Foreign Language Teaching and Research Press, pp. 3–25.
Halliday, M A K. (2004). ‘Representing the child as a semiotic being (one who
means)’, in Jonathan J. Webster (ed.), The Language of Early Childhood: The
Collected Works of M A K Halliday, Vol 4. London: Continuum.
Halliday, M. A. K. (2005). ‘On matter and meaning: the two realms of human
experience’. Linguistics and the Human Sciences 1 (1), 59–82.
Halliday, M. A. K. (2008). Complementarities in Language. Beijing: The Commercial
Press.
Halliday, M. A. K. and Hasan, R. (1976). Cohesion in English. London: Longman.
Halliday, M. A. K. and James, Z. L. (1993: Reprinted in full 2005). ‘A quantitative
study of polarity and primary tense in the English finite clause’, in J. J. Webster
(ed.), The Collected Works of M. A. K. Halliday, Vol 6: Computational and Quantitative
Studies. London: Continuum.
Halliday, M. A. K and Martin, J. R. (1993). Writing Science: Literacy and Discursive
Power. London: The Falmer Press.
Halliday, M. A. K. and Matthiessen, C. M. I. M. (2000). Construing Experience through
Meaning: A Language based Approach to Cognition. London: Continuum.
Grammar (3rd revised edn). London: Arnold.
Halliday, M. A. K and Greaves, W. S. (2008). Intonation in the Grammar of English.
London: Equinox.
Hasan, R. (1973). ‘Code, register and social dialect’, in Basil Bernstein (ed.), Class,
Code and Control, Vol 2: Applied Studies towards the Sociology of Language. London:
Routledgee & Kegan Paul. Reprinted in Jonathan J Webster (ed.) (2005a).
Language, Society and Consciousness: The Collected Works of Ruqaiya Hasan, Vol 1.
London: Equinox.
Hasan, R. (1983). ‘A semantic network for the analysis of messages in everyday talk
between mothers and their children’. Mimeo.
Hasan, R. (1984). ‘What kind of resource is language?’ in M. Garner (ed.), Austra-

lian Review of Applied Linguistics, Vol 7.1. The Applied Linguistics Association of
Australia.
Hasan, R. (1985). ‘Lending and borrowing: from grammar to lexis’, in J. E. Clark
(ed.), The Cultivated Australian: Festschrift in Honour of Arthur Delbridge. Amsterdam:
Helmut Buske.
Hasan, R. (1986). ‘Offers in the making: a systemic functional approach’. Mimeo.
Hasan, R. (1987b). ‘The grammarian’s dream: lexis as delicate grammar’, in
M. A. K. Halliday and R. P. Fawcett (eds), New Developments in Systemic Linguistics,
Vol 1: Theory and Descriptions. London: Pinter.
Hasan, R. (1995). ‘The conception of context in text’, in P. H. Fries and M. Gregory
(eds), Discourse in Society: Systemic Functional Perspectives. Norwood, NJ: Ablex.
Hasan, R. (1996). ‘Semantic networks: a tool for the analysis of meaning’, in
C. Cloran, D. Butt and G. Williams (eds), Ways of Saying, Ways of Meaning: Selected
Papers of Ruqaiya Hasan. London: Cassell, pp. 104–31.
Hasan, R. (2001). ‘Wherefore context? The place of context in the system and pro-
cess of language’, in S. Z. Ren, W. Gutherie and I. W. Ronald Fong (eds), Grammar
and Discourse: Proceedings of the International Conference on Discourse Analysis. Macau:
Universidad de Macau.
Hasan, R. (2004: Reprinted in full 2005a). ‘Reading picture reading: a study in
ideology and inference’, in J. J. Webster (ed.), The Collected Works of Ruqaiya
Hasan, Vol 1: Language, Society and Consciousness. London: Equinox.
Hasan, R. (2005a). The Collected Works of Ruqaiya Hasan, Vol 1: Language, Society and
Consciousness. London: Equinox.
Hasan, R. (2005b). ‘Language and society in a systemic functional perspective’,
in R. Hasan, C. Matthiessen and J. J. Webster (eds), Continuing Discourse on
Language, Vol 1: A Functional Perspective. London: Equinox.
Hasan, R. (2009a). Semantic Variation: Meaning in Society and in Sociolinguistics: The
Collected Works of Ruqaiya Hasan, Vol 2. London: Equinox.
Hasan, R. (2009b). ‘Wanted: A theory for integrated sociolinguistics’, in J. J. Webster
(ed.), Semantic Variation: Meaning in Society and in Sociolinguistics: The Collected
Works of Ruqaiya Hasan, Vol 2. London: Equinox.
Hasan, R. (2009c). ‘On semantic variation’, in J. J. Webster (ed.), Semantic Variation:
Meaning in Society and in Sociolinguistics: The Collected Works of Ruqaiya Hasan, Vol 2.
London: Equinox.
Hasan, R., Cloran, C., Williams, G. and Lukin, A. (2007). ‘Semantic networks: the
description of meaning in SFL’, in R. Hasan, C. Matthiessen and J. J Webster
(eds), Continuing Discourse on Language, Vol 2: A Functional Perspective. London:
Equinox.
Hjelmslev, L. (1961). Prolegomena to a Theory of Language. Translated by F. J. Whitfield.
Wisconsin: University of Wisconsin Press.
Katz, J. J. and Fodor, J. A. (1963). ‘The structure of a semantic theory’, in J. A. Fodor
and J. J. Katz (eds), The Structure of Language: Readings in the Philosophy of
Language. Englewood Cliffs, NJ: Prentice-Hall.
Labov, W. (1972). Sociolinguistic Patterns. Oxford: Blackwell.
Lave, J. (1997). ‘Learning, apprenticeship, social practice’. Journal of Nordic Educa-
tional Research 3, 140–51.
Leech, G. N. (1974). Semantics. Harmondsworth: Penguin.
Lemke, J. L. (1984). Semiotics and Education. Toronto Semiotic Circle Monographs.

Working Papers and Pre-publications. Toronto: Victoria University.
Lucy, J. A. (1987). ‘Vygotsky and Whorf: a comparative analysis’, in M. Hickman
(ed.), Social and Functional Approaches to Language and Thought. London: Academic
Press.
Marwick, B. (2005). ‘The interpersonal origins of language: social and linguistic
implications of an archaeological approach to language evolution’. Linguistics
and the Human Sciences 1 (2): 197–224.
Matthiessen, C. M. I. M. (1995). Lexicogrammatical Cartography: English Systems. Tokyo:
Internal Language Science Publishers.
Matthiessen, C. M. I. M. (2007). ‘The architecture of language according to
the systemic functional theory: developments since the 1970s’, in R. Hasan,
C. M. I. M. Matthiessen and J. J. Webster (eds), Continuing Discourse on Language,
Vol 2. London: Equinox.
Matthiessen, Christian and Nesbitt, Christopher (1996). ‘On the idea of theory-
neutral descriptions’, in Ruqaiya Hasan, Carmel Cloran and David Butt (eds),
Functional Descriptions: Theory in Practice. Amsterdam: John Benjamins.
Mertz, E. and Parmentier, R. J. (1991). Semiotic Mediation: Sociocultural and Psychologi-
cal Perspectives. New York: Academic Press.
McMahon, A. M .S. (1994). Understanding Language Change. Cambridge: Cambridge
University Press.
Noble, W. and Davidson I. (1996). Human Evolution, Language and Mind: A Psycho-
logical and Archaeological Inquiry. Cambridge: Cambridge University Press.
O’Donnell, M. (1994). ‘Sentence analysis and generation: a systemic perspective’.
Unpublished PhD thesis, Department of Linguistics, University of Sydney.
Russell, B. (1940). An Enquiry into Meaning and Truth. London: Allen & Unwin.
de Saussure, F. (1966). Course in General Linguistics (W. Baskin, trans.). New York:
McGraw-Hill.
Tucker, G. H. (1998). The Lexicogrammar of Adjectives: A Systemic Functional Approach to
Lexis. London: Cassell.
Vygotsky, L. S. (1978). Mind in Society: The Development of Higher Psychological Processes.
Cambridge, MA: Harvard University Press.
Wertsch, J. V. (1985a). Vygotsky and the Social Formation of Mind. Cambridge, MA:
Harvard University Press.
Wertsch, J. V. (1985b). Culture, Communication and Cognition: Vygotskian Perspectives.
Wertsch, J. V. (1997). ‘Collective memory: issues from a sociohistorical perspective’
in M. Cole, Y. Engeström and O. Vasquez (eds), Mind, Culture and Activity.
Williams, G. (1995). Joint Book Reading and Literacy Pedagogy: A Socio-Semantic
Examination. Unpublished PhD thesis, Department of Linguistics, Macquarie
University.
This page intentionally left blank
Index
NOTE: Page references in italics refer to tables and figures.
3 × 3 maps 186–7, 197 affinity group 82–3

theoretical underpinnings of 187–92 affinity space 82–3
training for literacy tutors 8, 30, 186 agent role
appliable linguistics 13–14
abstract coding orientation 225, 227–9 appraisal analysis 54–5, 118, 150–1,
abstract nouns 211–12 158–9, 173–4
academic domain 187, 189–90, 191 academic writing 172–3, 182
ideational meanings in 193–5 affiliation and 164–5
interpersonal meanings in 195–6 media’s portrayal of sports star’s
textual meanings in 196–7 transgressive behaviour 125–6
academic writing visual aids 7–8, 151, 161–4
as assessment tool 169–70 appreciation in appraisal system 54, 55,
evaluative work 8, 172–3, 182 118, 133, 173
model texts of 31–4, 36, 37 conversational humour 134, 139–40,
student writing vs. syllabus 141, 142–3, 145
documents 176–7 student writing 176, 178, 179–80,
see also Scaffolding Literacy in Adult 181, 182
and Tertiary Environments arbitrary realization 275, 276, 277
(SLATE) project archaeological anthropology 267–8
adolescent literacy practices assessment criteria
cultural context modelling 80–4, 89 development of 26, 29, 30–1
in social activism 6, 77–8 attitude 118, 133
affect in appraisal system 54, 55, 118, 173 conversational humour 134–6
student writing 176, 178, 179, 180, notion 155, 173
181–2 student writing 175–81
youth justice conferencing 60, 68 syllabus documents 174–5
affiliation 6 youth justice conferencing 60–1,
conversational humour and 7, 130–2, 68, 155
146–7 attributes 239, 240, 244–6, 248
levels of 136–8
of rap artist to consumers 246–8 bond(s) 134
role of evaluative language in 164–5 conversational humour and 134–5,
social process of 133–4, 140 138, 143–4, 145, 146
within civic domain 6, 82–3 syntagmatic 278, 280, 281, 286–7
308 Index
bondicon 61, 62, 157, 158 complementary structures 258

brightness (modality marker) 225 conjunctions 201–3, 208, 216
subjective uses 205
causal conjunctions 210 conjunctive adjuncts 201, 202
causal connectives 205 see also logical connectives
therefore 213–16 constituency 252, 257–60
causal relations 9, 202, 206, 207, constrained-based causation 211
210–13 content 274
nominalization of 212 content domain 203
causal verbs 210, 217 context 27, 275
types 210–11 models of 76
channel context of culture 76, 275
aural and visual modalities 96 models 77–8, 80–4
news text and 97–8 context of situation 76, 81
choice contextual configuration 94–5
of coding orientation in visual contextualization (modality
modality texts 226, 231 marker) 224–5, 228
meaning as 4–5, 14–18 conversational humour
in selection of vocalists in affiliation and 7, 130–2, 133–6,
construction of rap songs 240–1, 146–7
243 analytical approach to research
in translation 18–19, 21–4 data 132–3
civic domain naming and 136–8
communicative goals and 81–2 notion 133
social affiliations within 82–3, 84 coordinating conjunctions 201, 202
clause 258, 259 logico-semantic relations and 202
rank and 259–61 couplings 132, 134–6, 164
coding orientation 221, 225, 231–2, 271 Cousins, Ben
in pedagogic texts 225–31 cross modal phoric references 96
coherence 200, 203–5, 207–8, 216–17 cross modal resonances 96
cohesive harmony analysis 95 cultural context 76, 275
of news text 99–100, 102 cultural context model 77–8
colour differentiation (modality of adolescent literacy practices 80–4,
marker) 224, 227, 228–9 89
colour modulation (modality cultural domain 81–4
marker) 224, 227 culture
colour saturation (modality meaning and 269–70
marker) 224, 226–7, 229 role in negotiation of communal and
commissioned recounts 6, 47, 154 relational identities 132, 137–45
examples of 48–53
generic structure of 47–8, 154 default meanings 288–90, 296
ideational focus of 53, 54 depictive nominal groups 99
community identities depth (modality marker) 225, 228
conversational humour and 7, 137, description of language 283–6
141, 144–5, 146 dialectical realization 276, 277
compact prosodies 150 diffuse prosodies 150–1
comparative interpretation discourse markers 205
SLATE project 34–41 discourse relations 203–4, 205
Index 309
discourse semantics 27, 133, 190 field 80, 94–5

ideational meanings and 194 of news text 97–100
interpersonal meanings and 195 notion 27
textual meanings and 196 of production 93
discursive codes 103 of symbolic control see symbolic control
focus of appraisal system 174
educational linguistics 8, 185–6 student writing 176–7
elaboration 204 force of appraisal system 174
enforced causation 210 formative verbs 211
environment of translation 16
episode rank 253, 255 genre(s) 5, 76–7
epistemic domain 203 in academic domain 185–6, 189, 191
ergativity analysis 117–18, 119–25 distinction between register and 95
evaluative language ideational perspective 193, 194
affiliation and 164–5 interpersonal perspective 195
commissioned recounts 5–6, 45–6, Martin’s model 76, 92, 95
54–5, 69–70, 152 notion 27
commissioned recounts, gestural and pedagogical implications 27–8
phonological analysis 159–60 social identity and 83
commissioned recounts, prosodic textual perspective 196
structure 7–8, 150–1, 158–9 within civic domain 82, 89
commissioned recounts, gestural analysis
visualization 7, 163–4 of commissioned recounts 159–60
high-stakes examination writing 8, graduation of appraisal system 173
172–3, 175–81 notion 174
examination writings student writing 176–7
attitude and graduation 8, 176–7 syllabus documents 175
evaluative language 8, 172–3, 175–6 grammatical metaphors 209, 287
inscribed attitude 177–8
interplay between inscribed and Halliday, M. A. K.
invoked attitude 178–81 on appliable linguistics 1, 3–4
exophoric references 96 on linguistic meaning 271–2
experiential meaning potential on modality 222–3
affiliation and 248 on polarity 288–9
artist’s semiotic profile and 9–10, 245–6 rank grammar of 253–4, 255
notion 237 on register 92
of pitch register 239–40 reinterpretation of rank grammar
of rhythm 239 of 256–61
of roughness of voice 240 on relationship between systems and
experiential metaphors 209 instances 279
explicit attitude see inscribed attitude on semiogenesis 296–7
expression 267, 274, 275–6, 277–8 social activity model 76
external conjunctions 203, 205 trinocular perspective of 283–5, 296
humour see conversational humour
feedback hyper-themes 196–7
SLATE project’s structured approach
to 31, 36, 38–9 ideation 146
on students’ writing 170 notion 133, 136
310 Index
ideational meaning 27, 191 juvenile offenders

conversational humour 134–6 language use 53–4
images 253, 261 restorative justice 5–6, 44, 68–9
pathway of realization 193–5
identity construction knowledge
social affiliation and 7, 137–8, 146 common-sense vs. formal 171
sports stars’ transgressive behaviour
and 7, 115–16, 127–8 language resources 186
ideological identities 132, 137 of domain experts 208–10
of gender 143–4 stratification 190
of religion 143 langue vs. parole dichotomy 278
illumination (modality marker) 225, 228 lexical metaphors 209, 212
implicit attitude see invoked attitude lexicogrammar 27, 190, 274, 281
inchoative verbs 211 domain experts usage of 208–10
inscribed attitude ideational meanings and 194–5
commissioned recounts 63, 64–7, interpersonal meanings and 195–6
158–9 meaning and 276, 285–7, 295–6
conversational humour 133, 134, textual meanings and 196–7
138, 141, 142–3, 146–7 linguistic meaning 10, 273, 298
in student writing 177–81 from protolanguage to 271–2
instantiation 278–9 realization 275–8
interactional identities stratification 274–5
conversational humour and 137 see also meaning
internal conjunctions 203, 205 linguistic modality 222–3
interpersonal grammatical literacy domains 187, 189–90
metaphors 287 literacy pedagogy
interpersonal meaning 27, 150, 191 issues 168–72
modality and 9, 223–4 rationale and background 168–9
pathway of realization of 195–6 senior secondary school context 169
student writing 175–6 logical connectives 8–9, 200–1, 216–17
syllabus documents 174–5 alternative terminology 201
syllabus documents vs. student macro-categories of 202–4
writing 176–7 in sentence-initial position 206–7
intransitive verbs 210, 211 subjective functions of 205
invoked attitude as surface-level fillers 206
commissioned recounts 158–9 underuse by native speakers 207–8
conversational humour 133, 135, 146 logical metaphors 209
student writing 178–82 logico-semantic relations 202, 209
coordinating conjunctions and 202
judgement in appraisal system 54, 55, logogenesis 85–6, 151, 158–9, 161
118, 156, 173 long-range appraisal prosodies 151
media portrayal of sports stars’
transgressive behaviour 125–6 macrogenres 46–7, 83
student writing 176, 178, 179, 180, master identities
181–2 conversational humour and 137–8,
youth justice conferencing 56–63, 143, 144
68, 155, 158–9, 163–4 MCD see membership category devices
Index 311
MDA see multimodal discourse analysis negative polarity (not)

meaning 13 as lexicogrammatical category 287–8
centrality to human life 267 meaning of 10–11, 289–90, 295
as choice 14–18 realization of 288
history 10 use in demanding
lexicogrammar and 276, 285–7, information 290–5
295–6 neuroscience 268–9
making mind 268–9 news 93
mind and culture 269–70 as pedagogic discourse 93–4
modes of 271–2, 297–8 register perspective 6, 92–3, 102–4,
popular notion 272 105–11
in social life 267–8 on sports stars’ transgressive
member rank 253, 255 behaviour 7, 115–16, 126–8
membership category devices within SFL framework 92–3
(MCD) 134, 143–4
see also couplings paradigmatic relations 280, 283, 286,
message component 281 297
message semantics 281–2 PDA see positive discourse analysis
meta-redundancy 277 peri-causal verbs 211
modal verbs 222 personal recounts genre 53–4
modalities other than language persuasion 6, 77–8
(MOL) 10, 251, 261, 262 in academic domain 189
mode 94–5 in civic domain 82
of news text 95–7 in political testimony 85–9
notion and definition 27, 95 sample text 78–9
model of cultural context 6, 76, 77–8 photographic naturalism 224, 225
adolescent literary practices 80–4, 89 pitch register 239–40, 242–3
MOL see modalities other than polarity 287–8
language choice of 288–90
MOOD systems network 94, 287–8, political testimony 84–5
289, 291, 293, 296, 297 appeal element 85
multimodal discourse analysis identification element 86
(MDA) 10, 152, 221, 251 involvement 87–8
rank debate 252–6 legitimation 86–7
rank debate implications 261–2 motivating elements 85–6
multimodal texts 10, 261, 262 reassurance 88–9
unfolding path of 260–1 polysemy 201, 202–4
positive discourse analysis (PDA) 78
naming 132 predicated theme 215
conversational humour and 136–8, prosodies 191
146–7 protolanguage 271–2
narrative 80
native speakers rank 251–2
underuse of logical in multimodal discourse analysis 10,
connectives 207–8 252–6, 261–2
naturalistic coding orientation 225, not an isolated parameter 256–7
229–31 rank grammar 253, 254–5, 256
312 Index
rapping voice 240 social activity 97, 99, 100, 191

pitch register 242–3 Halliday’s model 76
rhythm 241–2 notion 190
roughness 243–4 social identity
realization 13, 190, 274, 275–8, 285–7 conversational humour and 131
instantiation and 278–9 role in genre determination 83
of negative polarity 288 role in persuasive political
recount genres testimony 86
stages 47–8, 80, 89 social networks 135
reflection-based activity 98–9 social process
text as 99 of affiliation 133–6, 140
register of news 103–4
distinction between genre and 95 role of language in 97
ideational meanings and 193, 194 social relations
interpersonal meanings and 195 conversational humour and 130–1
news discourse and 6, 92–3, 102–4, modality and 231
105–11 news item and 101
notion and definition 27, 92 social semiotics
register analysis 94–5 modality and 221–5
rhetorical structure theory (RST) 202 social variation 270–1
rhetorical unit (RU) 281 sociogenesis 270
rhetorical unit (RU) analysis 95, 203 sociolinguistics 1
of news text 97 criticism 2
rhythm 241–2 speeches of adolescent social activists
RST see rhetorical structure theory persuasiveness 6, 77–8, 85–9
RU see rhetorical unit stages 80
text 78–9, 88
scaffolding sports stars
notion 28 damaged heroes 127
in youth justice conferencing 49, media portrayal of transgressive
51, 71 behaviour 7, 115–16, 128
school genres transgressive behaviour of 114–15, 127
literacy teaching and 170–2 stacked area graphs 151, 161–4
semantics 274, 276, 280–1 stake out presentation mode 96, 97,
semiogenesis 296–7 98, 103
semiology 273 stratification 190, 274–5
semiotic mediation 10, 269–70 structure 95
sensory coding orientation 225, 226–7, register analysis 102–3
231 student writing see academic writing
SFL see systemic functional linguistics subordinating conjunctions 201, 202
shame 68–9 Sydney School pedagogy 26, 27–8, 190
sign/symbol 273, 275, 278 syllabus documents of disciplines
identity and value of 278, 279–80, appraisal profiles of 172, 178–82
283, 286–7 appropriate language and concepts
singing voice 240 in 180–1
pitch register 242–3 interpersonal meanings of 174–5,
rhythm 242 176–7
Index 313
symbolic control theme 196–7

Bernsteinian 93 predicated 215
discursive codes 103 tone group analysis 152, 153, 160,
publicly owned media and 93, 94 165
text producing interactants 101 tone setting subtexts 98–9
sympathetic circularity 293 transitive verbs 201, 209, 210–11
synoptic perspective 163, 260, 278–9 transitivity analysis 116, 118–25
syntagmatic relations 278, 280, 281, translation
286–7 appliable linguistics and 19
syntax 267, 287 problem of choice 14–15
system networks 234–5 problem of choice, examples 18–19
notion 280 tourist guides 16
of options in demanding goods/ see also Chinese-English tourist guide
services 281, 282, 284–5, 287, translation
290–1 translation equivalence 16, 17, 18–19
select expression 280 translation shifts 16, 17, 18–19
systemic functional linguistics (SFL) 4, trinocular perspective 283–6, 296
11, 116, 185–6, 221, 234, 296
definition criteria for categories 277 virtual addressee 96, 101
strata and metafunctions 27, 28, visual grammar 10
190–1, 257–8 of Kress and Van Leeuwen 253,
254, 256
tag(s) 293 of O’Toole 253, 254–5, 256
tagmemics 4, 257 visual modality 223–5
teaching-learning cycle 28–9 visual modality texts 9, 232
tempo 238, 239, 241–2 coding orientation in 225–31
tenor 94–5 visualization
academic domain 195 appraisal prosody 7–8, 151, 161–4
news item and 100–2 voice over presentation mode 96, 97, 98
notion 27 vox pop presentation mode 96, 97, 98
text visualization 7, 161
methods 161 YJC see youth justice conferencing
textual coherence youth justice conferencing (YJC) 5–6,
logical connectives and 201, 208 44–5, 72–3, 151–4
textual meaning 27, 191 criminological research on 45
pathway of realization 196–7 critique of 70–1
texture 95 emotional dynamics 69
register analysis 102–3 negotiation of meaning in 45–6
thematized adjunct 214–15 ritualization 69, 70–2

Appliable Linguistics (PDFDrive)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Appliable Linguistics (PDFDrive)

Uploaded by

Copyright:

Available Formats

Appliable Linguistics

Related titles available from Continuum:

© Ahmar Mahboob, Naomi K. Knight and contributors 2010

All rights reserved. No part of this publication may be reproduced or transmitted

British Library Cataloguing-in-Publication Data

ISBN: 978-1-4411-6415-5 (hardcover)

Library of Congress Cataloging-in-Publication Data

Typeset by Newgen Imaging Systems Pvt Ltd, Chennai, India

List of Figures vii

Chapter 1 Appliable Linguistics: An Introduction 1

Chapter 10 Evaluative Stance in Humanities: 168

Figure 3.1 Strata and metafunctions in SFL 28

Figure 11.3 Pathway of realization of ideational meanings in Text 11.1 194

Table 4.1 Inscriptions of affect in the Interpretation stage of the 60

David Caldwell is a PhD candidate in Linguistics at the University of Sydney

Yumin Chen is a lecturer and post-doctoral research fellow in the School of

Paul Dwyer is currently Chair of the Department of Performance Studies at the

Shoshana Dreyfus is a postdoctoral fellow in the Department of Linguistics at

intersections between critical discourse analysis and systemic functional linguis-

Emeritus Professor Ruqaiya Hasan taught Linguistics at Macquarie University,

Sally Humphrey is a post-doctoral fellow in the Department of Linguistics at

Alan Jones is a Senior Lecturer in the Department of Linguistics at Macquarie

Naomi K. Knight is a PhD candidate in the Department of Linguistics at the

Annabelle Lukin is a Senior Lecturer and Macquarie University Research Fellow

Ahmar Mahboob is a Senior Lecturer in the Department of Linguistics at the

J. R. Martin is Professor of Linguistics (Personal Chair) at the University of

Michele Zappavigna is a research fellow in the Department of Linguistics at

Sumin Zhao is currently completing her PhD. thesis on children’s educational

An edited volume is the joint effort of a number of people. It would be difficult

Ahmar Mahboob and Naomi K. Knight

Appliable Linguistics: An Introduction

This chapter introduces the need to establish Appliable Linguistics as the

on the other’ (pp. xiv–xv). However, by mid-1970s, researchers in applied lin-

It would not be too great an exaggeration to say that even as recently as

selective in their use of work in linguistics and sociolinguistics. For example,

I have always tried to work with a functional orientation to language; not

a comprehensive and theoretically powerful model of language which, pre-

applied to the problems, both research problems and practical problems,

The present volume takes this understanding of appliable linguistics as a

a realizational relationship whereby ‘a theory is made manifest, or “realized”, in

A concern with the communication of media is also expressed in Dreyfus

connectives), on the other hand, is current in ‘expert’ writing as evidenced in

semantic identity and value. Hasan concludes, ‘the meanings of a category of

Pinpointing the Choice: Meaning and the

2.1 Meaning as Choice

Given the demands we make on language, in almost everything we do, it is

in poets’ notebooks, or in think-aloud protocol records of translators – they are

be explained in the discussion of the examples in which they occur. There

2.2 Some Examples of Choice in Translation

Text A Chinese original, (i) in characters, (ii) in Pinyin romanization

Text B English translation given in the tourist guide

Appliable Linguistics and English Language

3.2 Background to the SLATE Project

3.3 Theoretical Underpinnings to the SLATE Project

In continuing to develop a pedagogy of empowerment, the SLATE project

Figure 3.1 Strata and metafunctions in SFL

Figure 3.2 Teaching–learning cycle (Rothery, 1996)

(i.e. disciplinary knowledge in academic contexts) needs to be built up. In the

3.4 Developments and Achievements of the SLATE Project

3.4.1 SLATE support for language coaches

Each of these criteria addressed a number of language features that were

provided in-depth feedback on these. The coaches were encouraged to start

3.4.2 SLATE support for students – modelling