BookPDF Available


"Methods of approaching the study of discourse have developed rapidly in the last ten years, influenced by a growing interdisciplinary spirit among linguistics and anthropology, sociology, cognitive and cultural psychology and cultural studies, as well as among established sub-fields within linguistics itself. Among the more recent developments are an increasing ‘critical’ turn in discourse analysis, a growing interest in historical, ethnographic and corpus-based approaches to discourse, more concern with the social contexts in which discourse occurs, the social actions that it is used to take and the identities that are constructed through it, as well as a revaluation of what counts as ‘discourse’ to include multi-modal texts and interaction. Advances in Discourse Studies brings together contributions from leading scholars in the field, investigating the historical and theoretical relationships between new advances in discourse studies and pointing towards new directions for the future of the discipline. Featuring discussion questions, classroom projects and recommended readings at the end of each section, as well as case studies illustrating each approach discussed, this is an invaluable resource for students of interdisciplinary discourse analysis."
Advances in Discourse Studies
Advances in Discourse Studies brings together contributions from top scholars
in the fi eld, investigating the historical and theoretical relationships between
new advances in discourse studies and pointing towards new directions for
the future of the discipline. Covering areas such as conversation analysis,
corpus-based discourse analysis and genre analysis, this book provides a
unique survey of the most recent advances in methodology and approach to
discourse analysis.
Featuring clear section introductions, discussion questions, classroom
projects and recommended readings at the end of each section, as well as
case studies illustrating each approach discussed, this will be an invaluable
resource for students of interdisciplinary discourse analysis as well as to
academics in a wider range of disciplines including linguistics, sociology,
anthropology, psychology, communication studies and cultural studies.
Vijay K. Bhatia and Rodney H. Jones are Professors in the Department of
English and Communication at the City University of Hong Kong. John
Flowerdew is Professor at the School of Education, University of Leeds.
Advances in Discourse Studies
Edited by
Vijay K. Bhatia,
John Flowerdew and
Rodney H. Jones
First published 2008
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
270 Madison Ave, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group, an informa
© 2008 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record for this book has been requested
ISBN 10: 0–415–39809–6 (hbk)
ISBN 10: 0–415–39810–X (pbk)
ISBN 13: 978–0–415–39809–1 (hbk)
ISBN 13: 978–0–415–39810–7 (pbk)
To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to”
This edition published in the Taylor & Francis e-Library, 2008.
ISBN 0-203-89229-1
Master e-book ISBN
List of contributors viii
Acknowledgements x
1 Approaches to discourse analysis 1
Conversation analysis 19
2 Conversation analysis: overview and new directions 22
3 Being out of order: overlapping talk as evidence of trouble
in airline pilots’ work 36
Suggestions for further work 51
Ethnographic-based discourse analysis 53
4 Ethnographic-based discourse analysis: uses, issues and
prospects 56
5 Using ethnography in the analysis of pedagogical practice:
perspectives from activity theory 67
Suggestions for further work 81
vi Contents
Corpus-based discourse analysis 83
6 Corpora and discourse analysis: new ways of doing old
things 86
7 Corpus-based analyses of discourse: dimensions of
variation in conversation 100
8 Corpora and context in professional writing 115
Suggestions for further work 128
Multimodal discourse analysis 129
9 Some thoughts on personal identity construction: a
multimodal perspective 132
10 Multimodal discourse analysis: the case of ‘ability’ in UK
secondary school English 149
Suggestions for further work 161
Genre analysis 163
11 Towards critical genre analysis 166
12 Genre evolution? The case for a diachronic perspective 178
Suggestions for further work 192
Contents vii
Critical discourse analysis 193
13 Critical discourse analysis and strategies of resistance 195
14 Mediation, text and action 211
Suggestions for further work 228
Mediated discourse analysis 229
15 Discourse itineraries: nine processes of resemiotization 233
16 Good sex and bad karma: discourse and the historical
body 245
Suggestions for further work 258
Index 259
Carol Berkenkotter , Professor, Department of Rhetoric, University of
Minnesota, Minneapolis, MN, USA. Email:
Vijay K. Bhatia , Professor, Department of English and Communication, City
University of Hong Kong, China. Email:
Douglas Biber , Regents’ Professor of Applied Linguistics, Department of
English, Northern Arizona University, Flagstaff, AZ, USA. Email: Douglas.
Lilie Chouliaraki , Professor and Chair of Media and Communication,
Department of Media and Communication, London School of Economics
and Politics. Email:
Traci Curl , RCKU Fellow, Department of Language and Linguistic Science,
University of York, UK. Email:
Paul Drew , Professor, Department of Sociology, University of York, UK.
Email: .
John Flowerdew , Professor of Applied Linguistics, University of Leeds, UK.
Lynne Flowerdew , Senior Lecturer, School of Education, University of Leeds,
UK. Email:
Carey Jewitt , Reader in Education and Technology, University of London,
UK. Email:
Ken Jones , Professor, Department of Education, Keele University, UK.
Rodney H. Jones , Associate Professor, Department of English and Com-
munication, City University of Hong Kong, China. Email: enrodney@
David Y.W. Lee , Assistant Professor, Department of English and Com-
munication, City University of Hong Kong, China. Email: davidlee@
Contributors ix
Angel Lin , Associate Professor, Department of Curriculum and Instruction,
Faculty of Education, Chinese University of Hong Kong, China. Email:
Maurice Nevile , Research Fellow Division of Communication and Education,
University of Canberra, Australia. Email:
Sigrid Norris , Associate Head of School (Research), Auckland University of
Technology, New Zealand. Email:
Ron Scollon , Retired. Email:
Graham Smart , Associate Professor. School of Linguistics and Applied
Language Studies, Carleton University, Ottawa, Canada. Email: gsmart@
We would like to thank the Department of English and Communication of
City University of Hong Kong for fi nancial and administrative assistance in
preparing this volume. In particular we would like to express our gratitude
to Ms Kitty Leung, for her effi cient and enthusiastic administrative support,
and to Ms Annabel Knibb and Mr Richard Forest for helping to edit and
proofread the manuscript.
1 Approaches to discourse
Vijay K. Bhatia, John Flowerdew and
Rodney H. Jones
In recent decades the social sciences have undergone a ‘discursive turn’
and become increasingly interested in the part played by language in the
creation of the reality that surrounds us. This interest has been accompanied
by the development of new theories and methods for the study of language
use and its role in human society. Discourse analysis, though often seen as
located within the discipline of linguistics, is in fact an interdisciplinary
eld of inquiry. With a history of less than 50 years it has acquired the
status, stability, signifi cance and integrity of a well-established discipline,
extending the conventional boundaries of linguistics. Dating back to the
1960s, it has been defi ned as the analysis of linguistic behaviour, written and
spoken, beyond the limits of individual sentences, focusing primarily on the
meaning constructed and interpreted as language is used in particular social
This defi nition really contains two main ingredients: the idea that language
can be analysed not just on the level of the phoneme/morpheme, the word,
the clause or the sentence, but also on the level of the text, and the idea that
language ought to be analysed not as an abstract set or rules, but as a tool for
social action. Although early conceptualizations of discourse analysis were
seen as an offshoot of linguistic analysis, focusing more on the ‘language as
text’ side of the equation and drawing on the work of early text analysts
like Propp (1958) and Jakobson (1937), in its present form it has moved to
more of a focus on ‘language in use’, drawing on insights from sociology,
psychology, semiotics, communication studies, rhetoric, as well as disciplines
such as business and marketing, accountancy, organizational studies, law and
information technology, to name only a few. In this regard, it has evolved as a
fruitful way of understanding the use of language in a variety of institutional,
academic, workplace and professional settings.
Another interesting aspect of the development of discourse analysis has
been that it has attracted the attention not only of linguists and applied
linguists, but also socio-political theorists, sociologists, anthropologists,
computer experts, business and legal specialists, communication experts and
organizational theorists. In this context, it is hardly surprising that discourse
analysis has, in the last four decades developed into a variety of schools
2 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
using different approaches, frameworks, procedures and methodologies and
focusing on different kinds of semiotic data, with the aim of deriving insights
for a variety of purposes.
The focus of most contemporary approaches to discourse on ‘language
in use’ has its roots in a number of larger developments in the twentieth
century in the fi elds of philosophy, anthropology, sociology and linguistics
itself. The roots of this view of language are perhaps best traced to the
work of Wittgenstein (1951/1972), who saw language as a series of ‘games’
through which people construct what he calls ‘forms of life’, particular ways
of being in relation to others and their surroundings. Less than two decades
later, with the publication of Austin’s 1962 classic How to Do Things with
Words, the notion that the study of language should involve more than
just its structure but also the way it is used and the way social standards
and practices shape and give rise to it became more prominent, at least
in philosophical circles. Later, thinkers like Foucault and Derrida, though
diverging considerably from the tradition of Austin, also made language, and
in particular ‘discourse’, central to their understanding of social practice.
Just as the notion of language as social practice began to take hold among
philosophers of language, social scientifi c disciplines particularly concerned
with social practice began to recognize the centrality of language in much of
what they were studying. Anthropologist Gregory Bateson and psychiatrist
Jürgen Ruesch (Ruesch and Bateson 1951) argued that social and psychological
phenomena cannot be separated from the ‘matrix of communication’ in
which they occur. They were followed by a host of social and behavioural
scientists, among them Goffman (1959) and Garfi nkel (1967), who focused
on the role of language in social behaviour and social formations. By the
1970s psychology, sociology, and anthropology had all taken a decidedly
‘discursive turn’, infl uenced not just by the structuralist linguistics of de
Saussure, but also by a new breed of linguists who at around the same time
were becoming more and more concerned with the relationship of language
to social actions and to the socio-cultural worlds of those who use it.
In America this concern had given rise to the work in the early part of the
twentieth century of Edward Sapir and Benjamin Whorf as well as others
in the Boasian tradition of anthropological linguistics. In Europe this new
concern for language in use was later exemplifi ed by the work of Michael
Halliday. Departing from structuralist and cognitive paradigms in grammar
which saw language systems as autonomous and independent of language
use, Halliday insisted that ‘language is as it is because of its function in the
social structure’ (1973: 65) and called for the development of a ‘sociological
linguistics’, a discipline which will allow us to see language on two levels,
a macro-sociological level in which language ‘serves to transmit the social
structure, the values, the systems of knowledge, all the deepest and most
pervasive patterns of the culture’ (1973: 45), and a micro-sociological level
in which meanings are seen as specifi c to particular contexts and situations.
Halliday’s systemic functional grammar (SFG) has had a profound infl uence
Approaches to discourse analysis 3
on many contemporary schools of discourse analysis, including critical
discourse analysis, mediated discourse analysis and multimodal discourse
analysis. Although they were primarily developed at the level of the clause,
the analytical tools of Halliday’s grammar have been found to be well
adapted to tracking participants, logical relations, processes, qualities, and
evaluations of these by speakers and writers as they develop throughout a
given text or across a group of texts (see e.g. Martin and Rose 2003).
The discourse analytical approaches that have grown out of these
interdisciplinary developments are many, including register and genre
analyses, critical discourse analysis, discursive psychology, conversation
analysis, interactional sociolinguistics, the ethnography of communication,
stylistics, mediated discourse analysis, corpus-based analysis, narrative
analysis, multimodal discourse analysis, rhetorical-grammatical analysis,
argumentation analysis, and many others, and no book on discourse could
hope to cover all of them. Our intention in this volume is to explore seven
major approaches to the study of discourse that we believe represent a range
of directions which the intellectual traditions we described above have taken.
Although all of them, to varying degrees, represent a concern for language
use in the social world, they focus on widely varying aspects of its use and
often defi ne the social world in widely varying ways, from the immediate
conversational context to the larger political, social or economic context.
They are corpus-based approaches to discourse, genre analysis, conversation
analysis, critical discourse analysis, multimodal discourse analysis, mediated
discourse analysis and ethnographic approaches to discourse. Some of them,
like genre analysis and conversation analysis, are decades old, while others,
like mediated discourse analysis and multimodal discourse analysis, are more
recent to the scene.
As the title of this book suggests, we are not as interested in describing the
type of work that has been done using these various approaches as we are in
showing how those working in these areas are charting new courses, which
often involve borrowing from other fi elds and other schools of discourse
analysis. In doing this we hope to understand not just what is unique about
each of these approaches, but also where future possibilities for convergence
and interdisciplinarity are opening up.
In order to do this, however, it is fi rst necessary to understand the main
questions upon which these approaches diverge and the different roads they
have taken from common intellectual roots. As will become clear in our
discussion, despite a common commitment to the study of texts and their
use in social contexts, those working in the different approaches diverge on
two of the most basic issues in this formulation: the question of what a text
is, and the question of what counts as the social context in which that text
is used. In a sense these differences can be seen as the result of the infl uence
of multiple disciplines on the development of discourse analysis. Sociology
and anthropology have encouraged analysts to view the use of language as
a function of the context in which language is used, whereas linguistics has
4 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
constrained discourse analysts to focus primarily on text, with context in
the background. In recent years another factor has entered the equation,
that is, the role of semiotic modes, other than written or spoken text, which
has opened up the possibilities of looking at nonlinear extra-linguistic forms
of communication such as pictures, diagrams, gestures, colours, differing
fonts and their sizes, to name only a few. Finally, discourse analysts are
faced with a variety of new media of communication including computer-
mediated communication, SMS messaging and other new communication
In what follows, we would like to look briefl y at where these different
approaches ‘have been’ in terms of their historical and intellectual
development in relation to the other approaches. We will also examine where
these approaches stand on these fundamental questions of text and context,
in preparation for, in the remainder of this volume, considering where they
might be going and how their trajectories might converge.
Conversation analysis
Conversation analysis (CA) was developed in the late 1960s and early 1970s
by Harvey Sacks, Emanuel Schegloff and Gail Jefferson (Sacks 1974; Sacks,
Schegloff and Jefferson 1974; Schegloff and Sacks 1973). It has its roots in
ethnomethodology, a branch of sociology developed by Harold Garfi nkel
(1967) which, like the ‘ethnography of communication’ of Gumperz and
Hymes, chiefl y concerns itself with the basic competences and interpretative
processes members of a culture use to interact and interpret their experience.
The central goal of conversational analysis, as Atkinson and Heritage
(1984:1) emphasize, is ‘the description and explication of the competences
that ordinary speakers use and rely on in participating in intelligible, socially
organized interaction’.
In contrast to earlier social scientifi c traditions, sociologists like
Goffman (1981) and Garfi nkel (1967) were adamant that people’s lives
should be studied only on their own terms without reference to theoretical
preconceptions. Rather than starting with a theory and analysing people’s
behaviour through it, those in this tradition advocate intentionally setting
aside theory to try to get at what’s actually going on based on close analysis
of people’s (often mundane) speech and actions.
By analysing the properties of conversation, conversation analysts attempt
to understand the patterns in social life. The assumption is that such patterns
can be used to develop procedural rules governing talk-in-interaction. Echoing
Austin and Steele, conversation analysts regard discourse as a kind of social
action – we are always ‘doing things with our words’. What is unique about
their approach is their concern with the sequential organization of actions,
and, in particular the mechanics of turn-taking. CA’s guiding analytical
principle, as Nevile reminds us in his chapter, is asking of each utterance in
a conversation the question, ‘Why that now?’ Utterances are seen as ‘paired
Approaches to discourse analysis 5
actions’ (Sacks, Schegloff and Jefferson 1974), always dependent on what
has come before and what will come after. The most basic expression of this
fundamental idea is the concept of the adjacency pair, a pair of utterances
that are not just linguistically related but ‘socially’ related because they
accomplish particular social actions. The ways utterances (and the actions
they accomplish) are put together sequentially follow rules of conditional
relevance, each utterance displaying a particular understanding of the
previous utterance and creating the conditions for subsequent utterances.
In contrast with the ethnographer’s data, which consists of interviews,
eld notes, lived experiences or narratives of participants, conversational
analysts work on naturally occurring and closely transcribed conversational
data. They regard observational data as prone to manipulation by researchers
or the subjects themselves. As Atkinson and Heritage (1984) point out,
observational data is often based on preconceived notions of what is probable
or important. Although like ethnographers, conversation analysts also work
on a selection of data, their data is conversational and not observational,
consisting of very detailed transcriptions of natural talk.
The most dramatic difference between linguistic ethnographers and
conversation analysis is on the issue of context. Whereas, for ethnographers,
the wider social context is used to inform their understanding of why language
is used the way it is, conversation analysts view context as constructed
moment by moment through conversational moves, and argue that those
aspects of context not conversationally attended to by participants should
not be part of the analysis. That is not to say that conversation analysts do
not concern themselves with larger issues of social identity and power (such
as gender and institutional communication). Rather, they believe that the
way to understand these issues is through a close analysis of the mechanics
of interaction rather than with reference to larger social structures or
Ethnographic approaches to discourse analysis
Although there are several ethnographic approaches to discourse analysis,
most of them draw their inspirations from anthropology and social
psychology and regard social context as the central aspect of communication.
Present ethnographic approaches to discourse owe much to an American
anthropological linguistic tradition that gave rise to the work of scholars
like Gumperz and Hymes (1986), whose ‘ethnography of communication’
aims to provide a description of how members of a particular community
are expected to perform linguistically in order to be considered ‘competent’
members. Communicative competence involves not just mastery of the
linguistic system, but the ability to use language in conjunction with social
practices and social identities in ways which others in the community will
recognize to perform a myriad of social activities such as engaging in small
talk, making transactions, joking, arguing, teasing, and warning. It is learnt
6 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
within communities through participating in communication, anticipating
others’ responses, and incorporating generalities into our own repertoire of
actions and meanings (Mead 1934).
Most traditions in ethnography from anthropology and linguistics aim
to understand the social world in terms of the ‘lived experience’ of those
who inhabit it. In this regard, they take what Pike (1967) calls an emic
approach to language, seeking to discover patterns in language use based
on the observation of natural social events of native participants. The
participants themselves have learnt these patterns by participating in their
communities and themselves observing by other members of the community.
Despite this commitment to ‘native’ ways of understanding communication,
ethnographic approaches to language invariably involve a process of selection
based on the analyst’s practical concerns and theoretical preoccupations. Data
gathering typically involves documenting, describing and interpreting social
practices through observation and analysis of a selection of socio-linguistic
behaviours. This selection invariably takes place in the fi eldwork, and often
consists of what researchers fi nds interesting, and of course, depends on
what the subjects allow them to observe. So there is a process of selection at
both ends, by the observers and the ones being observed. In addition to these
processes of selection, there is also another area of decision-making: when
the observer prepares a record of their fi eldwork, which consists of evidence
for the identifi cation, description and interpretation of social practices, they
invariably make use of a specifi c theory or framework within which they
select and interpret their observations. These observations are seen as the
most important tool for ethnographic analysis of communication.
However, more recently several other instruments have also been added
to the armoury of the ethnographer, which include structured or semi-
structured interviews with participants in social interactions, focus group
interviews, accounts of experienced participants, or what is often referred
to as ‘lived narratives of experience’, to name a few. Using a combination of
these tools, ethnographers interpret social behaviour of people in a specifi c
society or culture to reach conclusions and make generalizations.
Smart (1998 and this volume) distinguishes several kinds of ethnographies,
such as analytical, refl exive, naturalistic, institutional and interpretative.
Following the Geertzian tradition of the interpretation of cultures (Geertz
1973), he uses ‘interpretive ethnography’ to ‘explore a particular social
group’s discourse practices – as these are instantiated in writing, speaking,
or other symbolic actions – in order to learn how members of the group view
and operate within their mutually constructed conceptual world’ (Smart).
From the point of view of a conventional understanding of discourse
(textual) analysis, ethnographic analysis relies relatively less on actual
analysis of linguistic data and more on text-external social and contextual
factors. Ethnographers view purely linguistic analysis of data as less than
satisfactory for understanding language as it is used in social and cultural
contexts. At the same time, ethnographic approaches to discourse analysis
Approaches to discourse analysis 7
have had an important infl uence on other approaches to discourse such
as critical discourse analysis, genre analysis, mediated discourse analysis
and multimodal discourse analysis, and they have also made important
contributions to the fi elds of English for Specifi c Purposes (ESP) (Widdowson
1978; Swales 1981, 1990; Bhatia 1993) and what have come to be known
as the ‘new literacy studies’ (Barton and Hamilton 1998; Gee 1996; Street
1984). Lynne Flowerdew (this volume) also makes a case for introducing an
ethnographic element into corpus linguistics.
Corpus-based discourse analysis
Corpus-based analysis, which works with large amounts of machine-readable
text, was initially used primarily in the fi elds of lexicography and grammar.
It is only relatively recently that there have been extensive applications of
corpus approaches to discourse analysis (Baker 2006). The earliest initiatives
in corpus-based analysis of language use began with the creation of large
(by the standards of those days) general corpora representing language use
in a variety of contexts, both written as well as spoken, to draw insights
from observations about how people use language, both in terms of
lexico-grammar features and their functional variations. However, corpus
development over the years has changed in several important ways. First
the size of corpora has become much greater. The ‘Bank of English’ corpus
contains about 450 million words, whereas the British National Corpus has
about 100 million words. These large-scale general corpora are effective
and reliable in providing insightful information about the preferred use of
specifi c lexico-grammatical patterns in everyday language use. The most
important aspect of this approach is that it makes it possible for linguists
and discourse analysts to go beyond the analysis of sentences and short texts
to the analysis of huge amounts of text. It is thereby possible to corroborate
intuitions about individual instantiations concerning the functional value of
particular language patterns by recourse to very large numbers of instances.
Work with large corpora has demonstrated that language follows to a
large extent very regular patterns consisting of pre-constructed phrases.
This is referred to as the ‘idiom principle’ by Sinclair (1991), in contrast to
the ‘open choice’ principle, which refers to word-by-word ‘slot and fi ller
combinations. According to Sinclair (1991) speakers primarily adhere to the
idiom principle and only switch to the open choice principle when some
constraint occurs which makes the idiom principle fail to function.
One implication of the idiom principle view of language is that these pre-
constructed phrases may become the unit of analysis rather than individual
words. In analysing such units it has become clear that certain words and
phrases may take on particular ‘semantic preferences’ (typical areas of
semantic meaning). Thus the word glass typically occurs with a lexical set of
words to do with drinks, e.g. sherry, lemonade, water, champagne, milk, etc.
(Baker 2006: 86). At the same time words and phrases may carry ‘semantic
8 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
prosodies’ (typical areas of pragmatic meaning, or connotations). Thus,
a word like cause typically collocates with negatively loaded words – e.g.
accident, concern, damage, death, trouble – and thereby takes on a negative
semantic prosody; provide, on the other hand, is typically used with positive
collocates – e.g. aid, care, food, opportunities, relief, support – and thus takes
on a positive semantic prosody (Stubbs 1996). It is only through isolating
many examples of use derived from large amounts of text that observations
such as these regarding semantic preference and prosody can be made.
More recently corpus-based analysis has also become useful in the study
of language variations in specifi c academic and professional genres. These
corpora are usually much smaller. Connor and Upton (1996) make a strong
case for this kind of study of specialized corpora.
While general corpora are important and provide a critical foundation
for the study of language structure and use, they are less conducive for
analysing language use in specifi c academic and professional situations.
Consequently, there is now a strong and growing interest in compiling
specialized corpora that focus on specifi c types of genres within specifi c
contexts. Instead of being compiled for representativeness of language
across a large number of communicative purposes, specialized corpora
often focus on one particular genre … or specifi c situation … .
(Connor and Upton 1996: 2)
One important advantage of working with smaller corpora is that corpus-
specifi c semantic prosodies may be thrown up. For example, Flowerdew
(1997) showed how, in a corpus of the discourse of the last British governor
of Hong Kong, Chris Patten, words such as economy, individual, and wealth,
all carried a positive semantic prosody.
One of the reasons for the popularity of corpus-based discourse analysis
is the facility that it provides to handle and analyse large quantities of data
with minimal effort. Discourse and register analyses in their early days were
constrained by the fact that any manual processing and analysis of data was
seen as an impossible task, but with the availability of computers and a variety
of analytical software, these tasks have become not only less cumbersome,
but the results have also become more reliable and convincing.
Biber (this volume), one of the well-known specialists in this fi eld, proposes
what he calls multidimensional studies of register variation, by identifying
not simply the ‘salient linguistic co-occurrence patterns in a language’, but
also by ‘comparing spoken and written registers in the linguistic space defi ned
by those co-occurrence patterns’. By doing so, it is possible, he claims, not
simply to construct distinctive grammars of individual registers, but also co-
variant lexico-grammatical patterns across registers.
Approaches to discourse analysis 9
Multimodal discourse analysis
Multimodal discourse analytical approaches regard text as just one of the
many modes of communication available for social interaction. Although
the three approaches we have seen so far vary signifi cantly in terms of their
focus on text and context, one factor that is still restrictive in all of them is
the fact that they all take textual (linguistic) data to be the primary resource
for social interactions. There is a widespread belief now that textual data
is not necessarily the most important mode used for the construction and
interpretations of social meaning. Kress and Van Leeuwen (1996: 34) rightly
point out:
The new realities of the semiotic landscape are ... primarily brought
about by social and cultural factors: the intensifi cation of linguistic
and cultural diversity within the boundaries of nation-states, and by
the weakening of these boundaries, due to multiculturalism, electronic
media of communication, technologies of transport and global economic
developments. Global fl ows of capital dissolve not only cultural and
political boundaries but also semiotic boundaries.
Semiotic modes other than text can include gestures, posture, proxemics,
visual images, document layout, music and architectural design, to name a
few. Multimodality is especially important when one considers media such
as fi lm and TV, not to mention the increasing dynamics of electronic media.
Idema (2003: 33) sums up the difference in traditional text- or language-
based approaches to discourse analysis and the new multimodal approach to
discourse analysis as follows:
In general terms, the trend towards a multimodal appreciation of
meaning making centres around two issues: fi rst, the de-centring of
language as favoured meaning making; and second, the re-visiting and
blurring of the traditional boundaries between and roles allocated to
language, image, page layout, document design, and so on …. This
blurring of boundaries among the different semiotic dimensions of
representation has been linked, on the one hand, to changes in our
‘semiotic landscape’, and, on the other hand, to analysts’ realization that
our human predisposition towards multimodal meaning making, and
our own multi-semiotic development or ontogenesis, requires attention
to more than one semiotic than just language-in-use.
Another interesting perspective on multimodal discourse analysis is that it not
merely attempts to integrate all the possible semiotic modes of expressions,
but can also integrate various other approaches to discourse analysis. Some
working in multimodality have been infl uenced by conversational analysis,
most notably Goodwin (1981). Others, like Kress and van Leeuwen (2001)
10 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
and O’Toole (1994) borrow heavily from the systemic functional grammar
of Halliday and the work of Australian ‘social semioticians’ like Kress and
Hodge (1979) whom he inspired. Still others like Kendon (1990) owe a
considerable debt to anthropologists like Birdwhistell (1970) and Schefl en
Finally, more recent approaches, like that of Norris (2004), owe a great
deal to interactional sociolinguists like Tannen (1984) and mediated discourse
analysts like Scollon (2001). The two chapters represented in this volume
come from two of these broad traditions, Jewitt working in the tradition of
social semiotics, and Norris claiming her debt to interactional sociolinguistics
and mediated discourse analysis.
Genre analysis
In its earlier form, genre analysis was seen as an extension of linguistic analysis
to study functional variation in the use of English in academic contexts. Swales’
earliest work (1981) on research article introductions marked the beginning
of the genre analytical model for a grounded description of academic research
genres. The motivation was to use the fi ndings for the teaching and learning
of English for Specifi c Purposes. Unlike registers, which were identifi ed on
the basis of a specifi c confi guration of three main contextual categories of
eld, mode and tenor of discourse, Swales identifi ed genre on the basis of
its communicative purpose. There are two other versions of genre analysis
that emerged more or less the same time, one in Australia within systemic
functional linguistic theory, and the other in the United States within the
eld of rhetoric. Although these three frameworks draw their inspirations
from different sources, they seem to have considerable overlapping concerns
and perspectives.
Genre analysis, whether defi ned in terms of typifi cation of rhetorical
action, as in Miller (1984), Bazerman (1994) and Berkenkotter and Huckin
(1995), regularities of staged, goal oriented social processes, as in Martin et
al. (1987) and Martin (1993), or consistency of communicative purposes, as
in Swales (1990) and Bhatia (1993), can be viewed as the study of situated
linguistic behaviour in institutionalized academic or professional settings.
These are attempts to offer increasingly more complex (‘thicker’) descriptions
of language use, incorporating, and often going beyond, the immediate
context of situation, taking analyses beyond mere linguistic descriptions
to offer explanation for specifi c uses of language in conventionalized and
institutionalized settings. As we can see, the most important feature of this
approach in all three manifestations of genre theory is the emphasis on
In more recent years, genre analysis has developed in the direction of a
more comprehensive exploration of what Bhatia (2004) specifi es as ‘social
space’ to raise a number of other interesting issues, including some about
the integrity of generic descriptions. He proposes a multi-perspective and
Approaches to discourse analysis 11
multidimensional three-space model for the analysis of discourse as genre
integrating social professional space, social space and textual space. One
of the interesting aspects of this multi-perspective is the way it attempts to
integrate a number of other approaches to discourse analysis into a single
framework, some of which include ethnographic discourse analysis (Swales
1998), critical discourse analysis, corpus-based discourse analysis, and
multimodal discourse analysis.
Critical discourse analysis
Critical discourse analysis (CDA) focuses on socio-political domination,
which includes issues of social change, power abuse, ideological imposition,
and social injustice by critically analysing language as social action. It
is thus based on the assumption that the analysis of discourse provides
insightful information on such social issues as they are largely constituted
in language.
CDA regards discourse as an essential component of the constitution of
society and culture and is viewed (along with material action) as a major
form of social action. By studying discourse and society, CDA aims to
challenge inequality, injustice, unfairness and lack of democracy in society
by investigating social practices through a critical analysis of discourses and
social actions. Van Dijk, one of the founders of CDA, typically characterizes
its focus as the study of relationship between discourse, power, dominance
and social inequality (van Dijk 1993, 1998).
In line with this view, Fairclough (1989: 20) views language as a form
of social practice. He also regards CDA as exploring relationship between
discourse and social actors. For him discourse has potential for the expression
of particular ideologies and identities. This view of discourse is also in line
with social psychological theories of discourse (cf. Potter and Wetherall
1987), which regard discourse as a primary vehicle for the construction of
social and individual identities.
There are several approaches to critical discourse analysis. Fairclough
(1995: 2) offers a multi-dimensional framework for studying discourse
by mapping three separate forms of analysis onto one another: ‘analysis
of discourse, analysis of discourse practice (processes of text production,
distribution and consumption) and analysis of discursive events as instances
of sociocultural practice’. Van Dijk (1998) regards ideology as the basis for
the representation of social groups and hence he fi nds a useful link between
social structures and discourse structures. In general, van Dijk adopts a more
socio-cognitive approach to analysis. Wodak (1996), on the other hand, uses
CDA to study the issues of racism and anti-semitism by looking critically at
the historical dimension of discourse.
CDA is not without its critics. Probably most notably Widdowson (2004
for the latest version) has criticized CDA for bias. The starting point for
CDA, Widdowson claims, is a particular ideological commitment, which is
12 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
then supported by the selection of texts that are suitable for presenting the
desired analysis. There is a total lack of objectivity, according to Widdowson.
There are a number of answers to this critique. Meyer (2001) provides
some, namely that the analyst is necessarily subject to a certain bias, given
that all human beings are socially positioned, that CDA is at least open
about its commitment, and that, in the tradition of Kant, ‘pure’ cognition
is unattainable. Meyer (2001) also provides a list of criteria for assessing
the quality of CDA. This includes representiveness, reliability, validity,
completeness, accessibility and triangulation (see also J. Flowerdew 1999).
Mediated discourse analysis
Mediated discourse analysis (MDA) shares the goals of CDA, but focuses on
social action rather than on discourse. Like CDA, mediated discourse analysis
takes the analysis, interpretation and explanation of social problems as its
central concern; however, as Scollon (1998, 2001) points out, it does not
regard that these social issues are constituted primarily in discourse. Instead
it views discursive practice as just one form of social practice, not necessarily
the main form of practice out of which society creates its institutions and
power relations. Along with discourse, MDA argues, society and culture
are constituted in the material products and a myriad of non-discursive
Drawing heavily from Vygotskian psychology and sociocultural approaches
to the mind (see, for example, Wertsch 1991), MDA aims to understand
how discourse is used to take concrete social actions, and how, in those
social actions social structures and ideologies are created and re-created.
Like conversation analysis, it is concerned with the fundamental mechanics
of human action and its sequential organization in ‘chains of action’. At the
same time, it also borrows from the other end of the disciplinary spectrum
of discourse analysis, sharing with CDA and ethnographers of language a
commitment to understanding the relationships between these concrete,
situated social actions and social practice that reproduce larger patterns of
social relations and ideology within the ‘historical body’ of the social actor.
On the question of ‘text or context’, mediated discourse analysis claims
neither, choosing to focus instead on where text and context come together
in mediated actions.
The main concerns of mediated discourse analysis, as laid out by Jones
and Norris (2005) are: fi rst, mediated actions themselves, the concrete
things we do when we interact in the world; second, mediational means,
the ‘cultural tools’ (which may or may not be texts) with which we take
actions, and which enable or constrain these actions; third, the ‘practices’
that develop through these actions as they become part of the ‘historical
body’ of the social actor; fourth, the ‘sites of engagement’ in which multiple
social practices converge, opening a window for a mediated action to occur
Approaches to discourse analysis 13
(Scollon 2001); and fi nally, the way ‘agency’ in social actions is distributed
over individual social actors and cultural tools.
In terms of methodology, mediated discourse analysis uses resources from
different frameworks, which include critical discourse analysis, ethnography
and interactional sociolinguistics, and multimodal discourse analysis (Scollon
2001; Kress and Van Leeuwen 2001). According to Scollon (2001), MDA
is itself a nexus of practice at which the perspectives of mediated action
theory, anthropological linguistics and the ethnography of communication,
conversational analysis and ethnomethodology, critical discourse analysis
and the social practice theory of Bourdieu (1977) meet. He writes:
… mediated discourse as a theoretical framework mirrors the social
world that it hopes to analyse… it has taken on an identity through the
linkages overall that are made through concrete actions and projects over
time … we should not see this nexus of practice as a set of objectivized
or structural relationships among different schools. On the contrary,
these relationships exist only in and through concrete intersections of
these practices in specifi c research projects.
(Scollon 2001)
Interrelationships across discourse analytical approaches
At the beginning of the chapter we pointed out that these seven approaches
were quite distinct and were the result of very different motivations and that
they drew inspirations from different sources. However, as we moved along
each one of these approaches, we discovered that in spite of these differences,
some of these approaches were either infl uenced by others, or developed as
reactions to some of them. Conversation analysis, for instance, was a reaction
to an overwhelming concern with broad social structures and theory in then
current approaches to sociology. Similarly, the multimodal approach to
discourse analysis was a reaction to an equally overwhelming concern with
text in other forms of discourse analysis such as conversation analysis and
corpus-based discourse analysis. The corpus-based approach in itself was a
reaction to a number of approaches that confi ned themselves to the detailed
analysis of rather small sets of data. Genre analysis, in a similar manner,
was a reaction to analyses of de-contextualized lexico-grammatical features
of language, providing a way to make the analysis of texts more functional
and grounded in professional contexts. Critical discourse analysis was an
attempt to combine discourse analysis with social analysis, with implications
for the understanding of socio-cultural practices. Finally, mediated discourse
analysis was a reaction to what was seen as an overemphasis on the analysis
of discourse without a suffi cient understanding of the concrete social actions
people use discourse to carry out.
14 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
If we look at these approaches more closely, we fi nd that all of them can
be plotted along two major dimensions of Text/Context and Semiotic Mode.
These can be represented in the diagram shown in Figure 1.1.
If we look at these approaches as visually displayed in Figure 1.1, we fi nd
that all of these in some sense distinct approaches to discourse analysis differ
from each other depending upon the extent to which they regard social
context and/or semiotic forms that are used to construct discourses. Corpus-
based analyses of discourse are almost entirely focused on textual materials
(although see L. Flowerdew this volume), whereas multimodal analyses of
discourse extend to include other semiotic modes. If we go to the extreme
of social context, we fi nd ethnographic approaches to discourse focusing
almost entirely on social contexts, whereas conversation analysis shifts the
focus to the other extreme, focusing almost entirely on textual data. The
remaining three approaches to discourse analysis – genre analysis, critical
discourse analysis and mediated discourse analysis – seem to be paying
varying attention to both textual and other semiotic modes, on the one hand,
and social context on the other. The other factor common to all these three
is that they use varying combinations of frameworks and methodologies,
giving a kind of multidimensional perspective on discourse. However, they
Ethnographic analysis
Multimodal discourse
Conversation analysis
Corpus-based analysis
Te x t
Figure 1.1 Approaches to discourse analysis: text, context and semiotic mode
Approaches to discourse analysis 15
differ essentially in terms of the objectives they serve and the applications to
which they are suited. All of them pay some attention to texts and the social
contexts in which they are grounded, and in turn provide interesting insights
about the use of both language and social practices.
We have made a very brief attempt to introduce some of the main
approaches to discourse analysis. There was neither an intention to offer
detailed accounts of these approaches, nor to survey variations within these
individual approaches. More detailed accounts of all these approaches
will be offered in the chapters that follow, each one indicating how these
approaches are developing from the basic theoretical roots traced here and
are being exploited to analyse different forms of discourse in new ways.
These chapters focus less on the historical development of these approaches
and more on what lies ahead for them, and at the end of each section we
provide our own suggestions for further work that might be done in these
What should become clear in the chapters that follow is that these
approaches are not developing in isolation, but rather in constant dialogue
with one another, and it is in this conversation among approaches, we
argue, that the real advances in discourse studies will be made. In one sense,
although each individual approach provides a useful and credible view of
the elephant, as we might say, none of them, on its own, can provide a full
view of the elephant. This volume is an invitation to consider how these
different approaches can be harnessed and integrated in order to have as
comprehensive a view of the beast as is possible.
Atkinson, J.M. and Heritage, J. (eds) (1984) Structures of Social Action: studies in
conversation analysis, Cambridge: Cambridge University Press.
Austin, J.L. (1962) How to Do Things with Words: the William James Lectures
delivered at Harvard University in 1955, J.O. Urmson (ed.) Oxford: Clarendon
Baker, P. (2006) Using Corpora in Discourse Analysis, London: Continuum.
Barton, D. and Hamilton, M. (1998) Local Literacies: reading and writing in one
community, London: Routledge.
Bazerman, C. (1994) ‘System of genres and the enhancement of social intentions’,
in Aviva Freedman and Peter Medway (eds) Genre and New Rhetoric, London:
Taylor & Francis, 79–101.
Berkenkotter, Carol and Huckin, Thomas N. (1995) Genre Knowledge I Disciplinary
Communication: Cognition/Culture/Power, Hillsdale, NJ: Lawrence.
Bhatia, Vijay K. (1993) Analysing Genre: language use in professional settings,
London: Longman.
—— (2004) Worlds of Written Discourse: a genre-based view, London: Continuum.
Birdwhistell, R. (1970) Kinesics in Context, Philadelphia, PA: University of
Pennsylvania Press.
Bourdieu, P. (1977) An Outline of a Theory of Practice, trans. R. Nice, Cambridge:
Cambridge University Press.
16 Vijay K. Bhatia, John Flowerdew and Rodney H. Jones
Connor, U. and Upton, T.A. (1996) Discourse in the Professions: perspectives from
corpus linguistics, Amsterdam: John Benjamins Publishing Company.
Fairclough, N. (1985) ‘Critical and descriptive goals in discourse analysis’, Journal of
Pragmatics, 9: 739–63.
—— (1989) Language and Power, London: Longman.
—— (1995) Critical Discourse Analysis: the critical study of language, London:
Flowerdew, J. (1997) ‘The discourse of colonial withdrawal: a case study in the
creation of mythic discourse’, Discourse and Society, 8(4): 493–517.
—— (1999) ‘Description and interpretation in critical discourse analysis’, Journal of
Pragmatics, 31: 1089–99.
Garfi nkel, H. (1967) Studies in Ethnomethodology, Englewood Cliffs, NJ: Prentice-
Gee, J.P. (1996) Social Linguistics and Literacies: ideology in discourses, London:
Taylor & Francis.
Geertz, Clifford, (1973) The Interpretation of Culture, New York: Basic Books.
Goffman, E. (1959) The Presentation of Self in Everyday Life, Garden City, NY:
—— (1981) Forms of Talk, Oxford: Basil Blackwell.
Goodwin, Charles (1981) Conversational Organization: interaction between speakers
and hearers, New York: Academic Press.
Gumperz, J.J. and Hymes, D. (1986) Directions in Sociolinguistics: the ethnography
of communication, Oxford: Blackwell.
Halliday, M.A.K. (1973) Explorations in the Functions of Language, London: Edward
Idema, Rick (2003) ‘Multimodality, resemiotization: extending the analysis of
discourse as multi-semiotic practice’, Visual Communication, 2 (1): 29–57.
Jakobson, R. (1937) Lectures on Sound and Meaning, Cambridge, MA: MIT Press.
Jones, R. and Norris, S. (2005) Discourse in Action: introducing mediated discourse
analysis, London: Routledge.
Kendon, A. (1990) Conducting Interaction: patterns of behaviour in focused
encounters, Cambridge: Cambridge University Press.
Kress, G. and Hodge, R. (1979). Language as Ideology, London: Routledge.
Kress, G. and Van Leeuwen, T. (1996) Reading Images: the grammar of visual design,
London: Routledge.
Kress, G. and Van Leeuwen, T. (eds) (2001) Multimodality, London: Sage.
Martin, J.R. (1993) ‘A contextual theory of language in B. Cope and M. Kalantzis
(eds) The Powers of Literacy: a genre approach to teaching writing, Pittsburgh, PA:
University of Pittsburgh Press, pp. 116–36.
Martin, J.R., Christie, F. and Rothery, J. (1987) ‘Social processes in education: a
reply to Sawyer and Watson (and others)’, in I. Reid (ed.) The Place of Genre in
Learning: current debates, Geelong: Deakin University Press, pp. 46–57.
Martin, J.R. and Rose, D. (2003) Working with Discourse: meaning beyond the clause,
London: Continuum.
Mead, G.H. (1934) Mind, Self and Society from the Standpoint of a Social Behaviorist,
Chicago, IL: University of Chicago Press.
Meyer, M. (2001) ‘Between theory, method and politics: positioning of the approaches
to CDA’, in Wodak, R. and Meyer, M. Methods of Critical Discourse Analysis,
London: Sage, pp. 14–31.
Approaches to discourse analysis 17
Miller, C.R. (1984) ‘Genre as social action’, Quarterly Journal of Speech, 70: 157–78.
Norris, Sigrid, (2004) Analysing Multimodal Interaction: a methodological framework,
London: Routledge.
O’Toole, M. (1994) The Language of Displayed Art, Madison, NJ: FDU Press.
Pike, K.L. (1967) Language in Relation to a Unifi ed Theory of the Structure of Human
Behaviour, The Hague: Mouton and Co.
Potter, J. and Wetherall, M. (1987) Discourse and Social Psychology Beyond Attitudes
and Behaviour, London: Sage.
Propp, V. (1958) Morphology of the Folktale, Austin, TX: Texas University Press.
Ruesch, J. and Bateson, G. (1951) Communication: the social matrix of psychiatry,
New York: W.W. Norton and Company, Inc.
Sacks, H. (1974) ‘An analysis of the course of a joke’s telling in conversation’, in
J. Sherzer and R. Bauman (eds) Explorations in the Ethnography of Speaking,
London: Cambridge University Press.
Sacks, H., Schegloff, E.A. and Jefferson, G. (1974) ‘A simplest systematic for the
organization of turn taking in conversation’, Language, 50: 696–735.
Schefl en, A.E. (1974) How Behavior Means, New York: Anchor Books.
Schegloff, E.A. and Sacks, H. (1973) ‘Opening up closings’, Semiotica, 8: 289–327
Scollon, R. (1998) Mediated Discourse as Social Interaction: a study of news discourse,
London: Longman.
—— (2001) Mediated Discourse: the nexus of practice, London: Routledge.
Sinclair, J. (1991) Corpus, Concordance, Collocation, Oxford: Oxford University
Smart, G. (1998) ‘Mapping conceptual worlds: using interpretive ethnography to
explore knowledge-making in a professional community’, The Journal of Business
Communication, 35(1): 111–27.
Street, Brian V. (1984) Literacy in Theory and Practice, Cambridge: Cambridge
University Press.
Stubbs, M. (1996) Text and Corpus Analysis, Oxford: Blackwell.
Swales, J.M. (1981) Aspects of Article Introductions, Aston ESP Research Report
No. 1, Birmingham: Language Studies Unit, University of Aston.
—— (1990) Genre Analysis: English in academic and research settings, Cambridge:
Cambridge University Press.
—— (1998) Other Floors, Other Voices: a textography of a small university building,
Rhetoric, Knowledge and Society Series, London: Lawrence Erlbaum Associates.
Tannen, D. (1984) Conversational Style: analysing talk among friends, Oxford:
Oxford University Press, 2005.
Van Dijk, T. (1993) ‘Principles of critical discourse analysis’, Discourse and Society,
4: 249–83.
—— (1998) Ideology, London: Sage.
Wertsch, J.V. (1991) Voices of the Mind: a sociocultural approach to mediated action,
Cambridge, MA: Harvard University Press.
Widdowson, H.G. (1978) Teaching Language and Communication, Oxford: Oxford
University Press.
—— (2004) Text, Context, Pretext: critical issues in discourse analysis
, Oxford:
Wittgenstein, L. (1972) Philosophical Investigations, trans. G.E.M. Anscombe,
Oxford: Basil Blackwell and Mott.
Wodak, R. (1996) Disorders of Discourse, London: Longman.
Part I
Conversation analysis
In recent years, conversation analysts have developed the fundamental
principles of Sacks and Schegloff in various ways, often involving overlaps
with and borrowing from other fi elds of linguistics and discourse analysis.
There has been a considered movement, for example, from an almost
exclusive focus on the lexico-semantic features of talk to an interest in the
role of prosodic features (Couper-Kuhlen and Selting 1996), grammatical
features (Ochs et al. 1996), and non-verbal communication (such as gesture,
pausing and gaze) (Goodwin 2000; Heath 1986) in conversation. There
have also been considerable efforts to fi nd ways of applying conversation
analysis (CA) principles that were developed from data in English to other
languages such as Japanese (Tanaka 1999) and Finnish (Sorjonen 2001),
discovering how different language systems provide different resources for
the organization of talk.
Categorization analysis, which was, in fact, one of Sacks’s earliest preocc-
upations (Sacks 1972) is also becoming a more prominent feature of CA
in studies that show how social categories are used in interaction and the
conversational machinery of claiming, imputing, affi rming or challenging
social identity (Hester and Eglin 1997). Another important recent develop-
ment has been the movement towards using CA to understand people’s
behaviour in various workplace settings (Drew and Heritage 1992;
Heath 1997; Nevile this volume) and in the study of computer mediated
communication and human-computer interaction (HCI) (see, for example,
Luff et al. 1990; Norman and Thomas 1991).
At the same time, the approach to analysis and the theoretical principles of
CA have been extremely important in the development of other approaches
to discourse. They have had, for example, a profound infl uence on the
development of interactional sociolinguistics as practised by people like
Tannen (1989), and have also been adopted by discursive psychologist like
Potter and Weatherall (1987), who study how mental states are constituted
in discourse. Mediated discourse analysis, as well, owes a debt to CA for
providing a framework for understanding the workings of the ‘interaction
order’ and the sequential organization of mediated actions. Finally, CA has
been beginning a relationship with corpus-based discourse analysis, with
20 Conversation analysis
many conversation analysts working with rather large collections of texts
like telephone calls (Drew and Curl this volume) and pilot talk (Nevile
this volume) and looking for patterns over a range of texts. Corpora (and,
increasingly, computer based tools) help analysts to identify common
forms of conversational organization and then to ‘distributionalize’ (Sacks
1974) these phenomena or forms, determining the positions or sequential
circumstances in which they are most likely to occur.
The two chapters in this section represent two of these new directions in
which CA is advancing, namely the movement towards taking into account
the grammatical forms utterances in conversation take and the application
of CA to workplace practices. In their chapter, Drew and Curl not only
expand CA’s methodological scope by describing how the syntax of turn
construction plays a systematic part in determining the actions which are
accomplished in turns, but also signifi cantly expand CA’s preoccupation
with the sequential organization of conversational actions by considering
‘initial actions’, asking where they come from and what forms they are most
likely to take. Through their examination of phone conversations, they are
able to show how the syntactic forms utterances take are associated with
characteristic and appropriate ‘places’ in interaction and typically used to
deal with particular interactional circumstances or contingencies.
Nevile’s chapter is an example of the application of CA’s methodology
to workplace interactions. In particular he focuses on the effect of sequence
and timing on organizing interactions in collaborative work, showing how
the temporal organization of turn taking is relative to demands and goals of
particular work settings, in his case, the fl ight deck of a commercial airliner.
He also argues convincingly that deviations from the normal fl ow and timing
of talk can act as a warning signal that workplace activities may not be
progressing as they should.
Useful further reading on conversation analysis can be found in Drew
(2005), Ford et al. (2002), Ochs et al. (1996), Wooffi tt (2005), ten Have
(1999) and Schegloff (forthcoming).
Billing, M. (1997) ‘Whose terms? Whose ordinariness? Rhetoric and ideology in
conversation analysis’, Discourse and Society, 10: 543–58.
Couper-Kuhlen, E. and Selting, M. (eds) (1996) Prosody in Conversation, Cambridge:
Cambridge University Press.
Drew, P. (2005) ‘Conversation analysis’, in K. Fitch and R. Sanders (eds) Handbook of
Language and Social Interaction, Mahwah, NJ: Lawrence Erlbaum, pp. 71–102.
Drew, P. and Heritage, J. (1992) Talk at Work: interaction in institutional settings,
Cambridge: Cambridge University Press.
Goodwin, C. (2000) ‘Action and embodiment within situated human interaction’,
Journal of Pragmatics, 32: 1489–522.
Ford, C., Fox, B.A. and Thompson, S.A. (eds) (2002) The Language of Turn and
Oxford: Oxford University Press.
Conversation analysis 21
Heath, C.C. (1986) Body Movement and Speech in Medical Interaction, Cambridge:
Cambridge University Press.
—— (1997) ‘Analysing work activities in face to face interaction using video’, in D.
Silverman (ed.) Qualitative Methods, London: Sage.
Hester, S. and Eglin, P. (eds) (1997) Culture in Action: studies in membership
categorisation analysis, Washington, DC: University Press of America.
Luff, P., Gilbert, G.N. and Frohlich, D.M. (eds) (1990) Computers and Conversation,
London: Academic Press.
Norman, M.A. and Thomas, P.J. (1991) ‘Informing HCI design through conversation
analysis’, International Journal of Man–Machine Studies, 35: 235–50.
Ochs, E., Schegloff, E.A. and Thompson, S.A. (eds) (1996) Interaction and Grammar,
Cambridge: Cambridge University Press.
Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology: beyond attitudes
and behaviour, London: Sage.
Sacks, H. (1972) ‘An initial investigation of the usability of conversational data for
doing sociology’, in D.N. Sudnow (ed.) Studies in Social Interaction, New York:
Free Press, pp. 31–74.
—— (1974) ‘An analysis of the course of a joke’s telling in conversation’, in R. Bauman
and J. Sherzer (eds) Explorations in the Ethnography of Speaking, Cambridge:
Cambridge University Press, pp. 337–53.
Schegloff, E.A. (forthcoming) Sequence Organization in Interaction: a primer in
conversation analysis I, Cambridge: Cambridge University Press.
Sorjonen, M.-L. (2001) Responding in Conversation: a study of response particles in
Finnish, Amsterdam: John Benjamins.
Tanaka, H. (1999) Turn-taking in Japanese Conversation: a study in grammar and
interaction, Amsterdam: John Benjamins.
Tannen, D. (1989) Talking Voices, Cambridge: Cambridge University Press.
ten Have, P. (1999) Doing Conversation Analysis: a practical guide, Thousand Oaks,
CA: Sage.
Wooffi tt, R. (2005) Conversation Analysis and Discourse Analysis: a comparative and
critical introduction, London: Sage.
2 Conversation analysis
Overview and new directions
Paul Drew and Traci Curl
The background and programme of conversation analysis
Conversation analysis (CA) is founded on a sociological conceptualization of
the basically social nature of language use in human interaction. However as
the work in CA has developed, it has come to be a truly multi-disciplinary
eld; and in this account of recent developments and the directions in which
we see CA heading, we will highlight particularly the implications which
CA’s methods and fi ndings are having for core areas in linguistics – focusing
especially on the syntax of social actions in talk. But to begin with, we will
briefl y review the background to CA’s programme.
CA is a fi eld of study concerned with the norms, practices and competences
underlying the organization of social interaction. Notwithstanding its
name, it is concerned with all forms of spoken interaction including not
only everyday conversations between friends and acquaintances, but also
interactions in medical, educational, mass media and socio-legal contexts,
relatively ‘monologic’ interactions such as lecturing or speech-making,
and technologically complex interactions such as web-based multiparty
communication. Regardless of the interaction being studied, CA starts
from the perspective that (contra both Chomsky and Parsons) the details
of conduct in interaction are highly organized and orderly and, indeed,
that the specifi cities of meaning and understanding in interaction would be
impossible without this orderliness.
The central sociological insight of CA is that it is through conversation
that we conduct the ordinary, and perhaps extraordinary, affairs of our lives.
When people talk with one another, they are not merely communicating
thoughts, information or knowledge. Our relationships with one another,
and our sense of who we are to one another, are generated, manifest,
maintained and managed in and through our conversations, whether face-to-
face or on the telephone. People construct, establish, reproduce and negotiate
their identities, roles and relationships in conversational interaction. In our
interactions with others, we don’t just talk; conversation is not, to adapt
Wittgenstein’s phrase, ‘language idling’. We are doing things, such as inviting
someone over, asking them to do a favour or a service, blaming or criticizing
Overview and new directions 23
them, greeting them or trying to get on fi rst name terms with them, disagreeing
or arguing with them, advising or warning them, apologizing for something
one did or said, complaining about one’s treatment, telling about troubles,
sympathizing, offering to help and the like. These and other such activities
are some of the primary forms of social action. They are as real, concrete,
consequential and as fundamental as any other form of conduct. So when we
study conversation, we are investigating the actions and activities through
which social life is conducted. It is therefore primarily an approach to social
action (Schegloff 1996).
Methodologically, CA seeks to uncover the practices, patterns and generally
the methods through which participants perform and interpret social action.
CA emerged from two intellectual streams in sociology. The fi rst derives most
proximately from the work of Goffman (e.g. 1983a), who argues that social
interaction constitutes a distinct institutional order comprised of normative
rights and obligations that regulate interaction, and that function in broad
independence from the social, psychological and motivational characteristics
of persons. The second is Garfi nkel’s ethnomethodology (Garfi nkel 1967; see
also Heritage 1984a), which stresses the contingent and socially constructed
nature both of action and the understanding of action, and the role of
shared methods in the production, recognition and shared understanding
of joint activities. CA’s fusion of these two perspectives (through the work
of Sacks; see especially Sacks 1992 and Schegloff’s 1992 account of Sacks’s
analytic programme) resulted in an appreciation of the ways in which the
Goffmanian interaction order structures the production, recognition and
analysis of action as it unfolds in real time through the use of shared methods
or practices. This process (and its analysis) is possible because participants
refl exively display their analyses of one another’s conduct in each successive
contribution to interaction.
At the most basic level CA seeks to discover patterns in social interaction
in order to fi nd evidence of practices of conduct, in the systematic design of
turns at talk. To be identifi ed as a practice, particular elements of conduct
must be recurrent, specifi cally situated, and attract responses that discriminate
them from related or similar practices. A central feature of this procedure is
that the analysis of the practices used to perform a social action (e.g. using a
gurative expression to close down a topic, prefacing an answer to a question
with oh, or identifying a co-interactant by name in the course of a turn) can
be validated through the examination of others’ responses.
In pursuit of the goal of identifying basic interactional and communicative
competencies, CA focuses on sequences of actions. In performing some
current action, participants generally project (empirically) and require
(normatively) the production of a ‘next’ or range of possible ‘next’ actions to
be done by another participant. Moreover, in constructing a turn at talk, they
normally address themselves to immediately preceding talk, and design their
contributions in ways that exploit this basic positioning. In the production
of next actions, participants display an understanding of a prior action and
24 Paul Drew and Traci Curl
do so at a multiplicity of levels – for example, by an ‘acceptance’ a speaker
can show an understanding that the prior turn was possibly complete, that
it was addressed to them, that it was an action of a particular type (e.g. an
invitation) and so on. Within this framework, the grasp of a ‘next’ action that
a current projects, the production of that next action, and its interpretation
by the previous speaker – are the products of a common set of socially shared
practices. CA analyses are thus simultaneously analyses of action, context
management and intersubjectivity – because all three of these features are
simultaneously, if tacitly, the objects of the participants’ actions.
Based on this framework, CA has developed as an empirical discipline
focused on a range of domains of interactional conduct, including turn-
taking (the allocation of opportunities to speak among participants), the
organization of conversational sequences, the internal structuring of turns
at talk and the formation of actions, the organization of repair (dealing with
diffi culties in speaking, hearing and understanding talk), story-telling and
narrative, phonetic and prosodic aspects of talk, body behaviour and so on.
These organizations form the technical bedrock on which people build their
social lives, and construct their sense of sociality with one another.
Interaction of any kind is made possible through participants sharing certain
communicative competencies. These consist partly of knowledge about the
language, of the ways that elements of language (including lexis, grammar
and syntax, intonation, prosody and so forth) are organized, combined and
deployed. But they include, most crucially, knowledge also of the structures,
patterns, norms and expectations concerning the social organization of
(verbal) action in sequences of interaction. Such ‘knowledge’ is not generally
something of which we are aware at any conscious level. It is, however, salient
to participants in interaction in their establishing a mutual understanding of
what they are saying and doing in the talk. Thus the coherence of talk, and the
mutual understandings which underlie it, rest on a ‘common set of methods
or procedures’ (Heritage 1984a: 241); and these in turn are the constituents
of our basic communicative competencies. So when we study conversation, we
are attempting to discover what are the essential and quite fundamental shared
competencies that make all coherent social action – including communication
– possible between members of a culture. These competencies or sense-making
practices (Pomerantz and Mandelbaum 2005) consist of the practices and
devices that are the focus of CA research. That is, the aim of research in CA
is to discover and explicate the practices through which participants produce
and understand conduct in interaction. These practices are uncovered, in large
part, through identifying patterns in talk.
In these respects CA lies at the intersection between sociology and other
cognate disciplines, especially linguistics and social psychology. Certainly
research in CA has paralleled developments within sociolinguistics,
pragmatics, discourse analysis and so forth towards a naturalistic, observation-
based science of actual verbal behaviour, which uses recordings of naturally
occurring interactions as the basic form of data (Heritage 1984a). All levels
Overview and new directions 25
of linguistic production (e.g. syntax, phonetics) can be related to the actions
(such as greetings, invitations, requests) or activities (instructing, cross-
examining, performing a medical examination and diagnosing, etc.) in
which people are engaged when interacting with one another. In this way
conversational organizations underlie social action (Atkinson and Heritage
1984); hence CA offers a methodology, based on analysing sequences in
which actions are produced and embedded, for investigating how we
accomplish social actions.
In summary, CA investigates the organizations of and interconnections
between four underlying characteristics of talk-in-interaction (see e.g. Drew
2005), which are:
turn design or construction
sequence and sequence organization
Grammar and interaction
It is worth highlighting two points in the account we have given so far of
CA’s programme. First, we noted that CA has focused largely (though not
exclusively) on sequence, and especially on responses, and how responses both
display understandings of prior actions and themselves create contingencies
for subsequent actions – thereby helping to shape the unfolding sequence
(sequential patterns). Second, we stressed that all levels of linguistic production
are involved in turn design – the construction of a turn-at-talk from a range of
elements or components, including word selection, syntactic and grammatical
features, phonetic and prosodic aspects, as well as (in face-to-face interaction)
gaze, posture, bodily orientation and the like. Until recently CA research
focused largely on word selection (lexis); with some rare exceptions, the ways
in which different linguistic resources were integrated into turn design, and
the role played by other levels of linguistic production in the management
of action, were not much investigated. For instance, Couper-Kuhlen and
Selting (1996: 11) pointed out that prosody had been largely neglected in the
empirical study of spoken interaction: despite its being no less important than
other (linguistic) turn design components, research had tended to focus on
lexico-syntactic features, due perhaps to the legacy of the infl uence of literacy
on studying language use. They further argued that:
… prosody can be seen as one of the orderly ‘details’ of interaction, a
resource which interlocutors rely on to accomplish social action and
as a means of steering inferential processes. Prosodic features... can be
reconstructed as members’ devices [authors’ emphasis], designed for the
organization and management of talk in social interaction. They can
be shown to function as part of the signalling system which – together
26 Paul Drew and Traci Curl
with syntax, lexico-semantics, kinesics, and other contextualizing cues
– is used to construct and interpret turn constructional units and turns-
(Couper-Kuhlen and Selting 1996: 25)
Although we mentioned that research has tended to focus on lexico-syntactic
features of talk-in-interaction, it would be more accurate to say that CA
has not much investigated grammar in general, or syntax in particular, at
all formally or systematically. Until recently, our approach to grammar has
been rather ad hoc, drawing on observations about grammatical or syntactic
features as and when they seemed salient, but without taking grammar/syntax
as the principal object of systematic inquiry. We should just note that the
emphasis in CA research on lexis should be taken to include all those kinds
of verbal objects, including oh (Heritage 1984b and 2002a), mm (Gardner
1997), particles (Sorjonen 2001) and laughter (Jefferson 1979; Haakana
2001; Glenn 2003), the use and semantics of which have generally been
beyond traditional theories of language; one of the strengths of CA has been
that it encompasses all those seemingly trivial, fl awed, ‘meaningless’ and
incidental details of speech, most of which cannot be handled or explained
through traditional linguistic theories. At any rate, only recently has CA’s
methodology, based on sequential analysis, been deployed to investigate at
all formally the ways in which grammar and syntax are practices for talk-in-
interaction. A certain awakening of how grammar and interaction intersect
came with the publication (like its companion volume by Couper-Kuhlen
and Selting on prosody, also in 1996) of a collection focusing on the ways
in which the grammar and syntax of various languages are employed to
perform various different practices in conversation (Ochs et al. 1996). Since
then, a small number of studies have begun to explore how the syntax of
turn construction plays a systematic part in determining both the actions
which are accomplished in turns, and the precise affordances (including
inferential character) of those designs/actions (examples are Heritage 2002b
and Tanaka 1999). It is this direction of research in CA that we want to
highlight and illustrate in the remainder of this chapter.
CA’s focus on patterns of unfolding sequences of interaction, most
notably perhaps in the accounts of various kinds of adjacency pairs and the
preference organization associated with the sequencing and formation of
alternative responses to ‘initial’ actions such as invitations, requests and
the like (e.g. Drew 1984; for perhaps the best account of adjacency pairs,
see Sacks’s lectures 1–5, Spring 1972, in Sacks 1992, vol. 2), has resulted
in our having given less attention to the ‘initial’ actions themselves, how
they are formed, and, in short, how such sequences are generated in the
rst place. Recent work in CA, including our own research, has begun to
investigate precisely the question of how ‘initial’ actions come to be formed,
syntactically, as they are (e.g. Heinemann 2006; Curl et al. forthcoming).
That is, we are beginning to ask, for instance, how speakers come to select
Overview and new directions 27
from among a number of alternative syntactic forms that might be available,
a particular form with which to initiate or perform a certain action. We are
putting ‘initial’ in inverted commas in order to highlight the fact that while
studies in CA have tended to treat actions such as assessments, invitations,
complaints, etc. as ‘fi rst’ actions (e.g. fi rst pair parts in adjacency pairs), in
reality they also emerge from ongoing interaction; they do not come from
nowhere – they arise out of whatever particular interactional circumstances
and contingencies that obtain up to that point. Goffman summarizes this
matter in his usual pithy manner, thus:
Most important of all, the sense in which current utterance is conditioned
by immediately prior turn’s talk – when, indeed, there is such talk –
does not speak to the many elements of the same current utterance that
are not in any way determined by prior turn (or prior utterance in the
same turn), yet are nonetheless determined in ways that satisfy Felicity’s
Condition. In any case, an account of second utterances in terms of their
contingency on a fi rst leaves unexplained how there could be any fi rsts;
after all, from where could they draw their design? Conversation could
never begin. Or, once begun, would be one utterance away from the
end. Tails would know how to wag, but there would be no dogs.
(Goffman 1983b: 50)
To illustrate this, consider the ways in which we make offers to others,
particularly when offering some kind of assistance. Reviewing a large corpus
(somewhat over 20 hours) of telephone calls recorded in homes both in the
US and UK, of conversations between family, relatives, friends, colleagues
and sometimes to service organizations of one kind or another (shops, banks,
doctors, etc.), we found that offers were made predominantly using one of
three syntactic forms: these three forms are illustrated in the following brief
Example 1 (Holt:2:3:1)
1 L
es: .hh And he now has: u-a:: um (1.1) I don’t think eez called
2 it consultancy (0.2) They fi nd positions for people: in the
3 printing’n paper (0.4) indus [try:,
4 Mar: [Oh I see: [:.
5 Les: [ hh An:d if: your
6 husband would li:ke their addre[ss.
7 Mar: [Y e : [: s,
8 Les: [<As they’re
9 specialists,
10 Mar: Ye::s?
11 (.)
12 Les: Uhm: my husband w’d gladly give it to him.
28 Paul Drew and Traci Curl
Example 2 (NB:IV:10:19)
Lot: =W’l listen (.) e-uh: dih you want me uh come dow’n getchu
to [morrow er] a n y th]ing?
Emm: [n:N o : :] dea:r.]
Emm: No [: I’m]fi ne]
Lot: [to the] sto::]re ’r anythi:ng?
Example 3 (NB:IV:4:4)
Emm: W’l anyway tha:t’s a’dea:l so I don’know what to do about
Ba:rbara .hhhhh (0.2) c’z you see she w’z: depe [nding on:
(L): [(ºYehº)
Emm: him taking ‘er in tuh the L.A. deeple s:- depot Sundee so
[‘e siz]
Lot: [I:’ll] take ‘er in: Sundee,
In the fi rst example, Lesley is offering to put Mary’s husband, who is evidently
in the printing trade but currently unemployed, in touch with an employment
agency specializing in that fi eld. She does so using the conditional form, if
(your husband would like …), (then) (my husband will …) (the contingent
clause is not in fact initiated with then; as is most commonly the case, it is
left implicit). Lottie constructs her offer in Example 2, to collect her sister,
Emma, to take her shopping, in an interrogative syntactic form as a do you
want construction. By contrast, Lottie’s offer in the third example, from a
different telephone call, is syntactically declarative, I’ll do … . We see, then,
that in each case, the offer is made through the selection of a particular
syntactic form: syntax is as much part of turn design as is the choice of words,
or prosodic features and so on. These three syntactic forms – conditionals,
Do you want interrogatives and declaratives (and related formats) – were the
most commonly occurring in our data corpus.
In order to investigate the interactional contingencies that may be
associated with a given form, which can often be pretty close to the
interactional ‘functions’ of a form or practice, we can do what Sacks termed
‘distributionalize’ a phenomenon/form (Sacks 1974). That is, we look to
see where in talk that practice or form tends to occur, or in what sequential
circumstances or positions that form is found/systematically used (see, for
example, Drew and Holt 1998 for an account of investigating the systematic
distribution of an object, there of fi gurative expressions in conversation).
When we examined the distribution of the occurrence of these three
syntactic forms for offering, it appeared that the fi rst, conditionals, occurred
at the beginnings of telephone calls, when an offer was the fi rst topic after
the call openings; by contrast, Do you want … ? interrogatives occurred
in call closings; whilst declarative forms occurred in what might loosely be
considered the ‘middles’ of calls, i.e. not close to the beginning or ends of
Overview and new directions 29
This emerging pattern turned out to be somewhat true, but misleading. We
began to fi nd cases that did not fi t this pattern of occurrence. For instance, in
the following example Lesley makes an offer using the conditional format,
although this is not during or even near the beginning of the call.
Example 4 (Holt:X(C):1:1:3:5)
1 L
es: The other thin:g (.) was (.) uhm .t.h we’ve had an
2 invoice: fr’m Scott’s .hh (.) Now if they deliver it
3 (0.5) to you: UH (0.7) we- (.) we wanted t’pay the
4 Les: carriage w’l they haven’t invoiced us f’r any
5 carriage.
6 Phi: .hwhh (.) We:l [l
7 Les: [So-
8 (0.2)
9 Phi: [they
10 Les: [if they come t’you an’ invoice you f’carriage say
11 that it’s we that’re paying oka:y?
Lesley makes this offer, to pay for some plants that she has asked to be
delivered to Philip (‘if they come t’you an’ invoice you f’carriage say that
it’s we that’re paying oka:y?’, lines 10–11; though notice that Lesley began
something like this offer in lines 2–3, ‘if they deliver it (0.5) to you: …’),
using the conditional form – the form which was emerging was that used
in call beginnings. However, she makes this offer 2 minutes 40 seconds
into the call, and therefore at some distance from the beginning. Space
prevents us showing other instances where forms were, as in Example 4,
used ‘inappropriately’, according to our emerging hypothesis; suffi ce it to
say that we found a few, not many, instances of declarative forms used in
call closings, and Do you want … ? interrogatives in ‘middles’. There was an
association, a correlation perhaps, between these three syntactic forms and
the three different positions in telephone conversations; but the association
did not hold for all.
It turned out that underlying the distributional pattern which we had
begun to suspect was another – a pattern which was itself associated with
beginnings, ‘middles’ and endings of calls, but one which – rather like an
intervening variable – was not the determining factor in speakers’ selections
of the appropriate syntactic form. A clue to this is to be found in Example 4,
and particularly Lesley’s initiating this topic, and her offer, with ‘The other
thin:g (.) was..’ (line 1). Her preface The other thing is a way to indicate
that this is part of her agenda, one of the reasons for calling. Right at the
beginning of the call, Lesley offers her condolences to Philip on the death
of his mother the day before; that topic, the funeral arrangements, etc.,
quite naturally take precedence. But now in Example 4 it turns out, and
Philip discovers, that Lesley has called specifi cally to make this offer. It is
30 Paul Drew and Traci Curl
that – that this is the reason for Lesley’s call – which informs the selection
of the conditional construction with which to make the offer. Of course, the
opening of a conversation, whether on the telephone or face to face, is where,
generally, a speaker introduces the reason for initiating the interaction. But
there can be circumstances, such as this in Example 4 (i.e. that the death of
Philip’s mother is more important, and takes precedence) in which a reason
for calling comes to be delayed.
We’ll see, then, that the conditional offer in Example 1 is the reason why
Lesley has phoned Mary.
Example 5 (Holt:2:3)
1 Mar: One three fi ve?
2 (.)
3 Les: Oh hello, it’s um: Leslie Field he:re,
4 Mar: Oh ^hello:,
5 Les: Hello, .tch.h I hope you don’t mind me getting in touch
6 but uh- we met your husband little while ago at a Liberal
7 meeting.
8 (0.3)
9 Mar: Ye: [s?
10 Les: [.hh And he wz: (0.3) i-he told us something of what’d
11 happen:ed,
12 (0.5)
13 Les: to him .hh An:’ I wondered haa- (0.2) i-he said he m::ight
14 have another position in vie:[w,
15 Mar: [Mmhm,
16 Les: .hh (.) Uhm (0.3) .tch Well I don’t know how that went, .h
17 uh (.) It’s just thet I wondered if he hasn:’t (0.3) uh
18 we have friends in: Bristol
19 Mar: Ye:s?
20 Les: who:-(.) uh: thet u-had the same experience.
Mar: Oh:_:.
22 Les: And they uhm: .t (0.2) .hh He worked f’r a printing an:’
23 paper (0.9) uh fi rm[u-
24 Mar: Ye:s,
25 Les: uh[:- which ih puh- uh: part’v the Paige Group.
26 Mar: [Yeh,
27 (.)
28 Les: .hh And he now has: u-a:: um (1.1) I don’t think eez called
29 it consultancy (0.2) They fi nd positions for people: in the
30 printing’n paper (0.4) indus [try:,
31 Mar: [Oh I see: [:.
32 Les: [ hh An:d if: your
33 husband would li:ke their addre[ss.
Overview and new directions 31
34 Mar: [Y e : [: s,
35 Les: [<As they’re specialists,
36 (.)
37 Les: Uhm: my husband w’d gladly give it to him.
Again, Lesley’s preface ‘I hope you don’t mind me getting in touch’ (line 5)
explicitly indicates this (the upcoming offer, which here as in all such cases
is preceded by some account of the circumstances of making the offer) is the
reason she has called.
Compare this with Example 3 above, in which Lottie offers to take
Emma’s daughter (Barbara) and her family to catch the Greyhound bus at
the LA depot.
Example 3 (NB:IV:4:4)
Emm: W’l anyway tha:t’s a’dea:l so I don’know what to do about
Ba:rbara .hhhhh (0.2) c’z you see she w’z: depe [nding on:
(L): [(ºYehº)
Emm: him taking ‘er in tuh the L.A. deeple s:- depot Sundee so
[‘e siz]
Lot: [I:’ll] take ‘er in: Sundee,
Lottie’s offer is made spontaneously, in response to Emma’s expression of
a problem ‘I don’know what to do about Ba:rbara’ (lines 1–2), regarding
Barbara getting to the bus depot (lines 2–4). This illustrates the distribution
of offers made in declarative forms; they occur in the sequential position
of responses to the other’s explicit account of a problem, and are thus
spontaneous offers generated interactionally by the other’s expression of a
problem or diffi culty (and not reasons for calling).
The other syntactic form, Do you want … ? interrogatives, also appear
to be interactionally generated, but not by some immediately preceding or
explicit expression of trouble, as for the declarative forms, but by some need
which the speaker educes the other might have. Here is a case in point;
Chloe and Claire are members of a group of women who meet regularly
in one another’s homes to play bridge; evidently Chloe is hosting the next
game, and at the point where their conversation is ending (lines 8–12) Claire
offers to bring more chairs for the event.
Example 6 (SBL:2:2:3:28)
1 Chl: We:ll it was [fu:n Clai [re,
2 Cla: [hhh [Yea:: [:h,]
3 Chl: M]mº
4 Chl: [(an’)
5 Cla: [I enjoyed every minute o [f it,
6 Chl: [Yah.
32 Paul Drew and Traci Curl
7 (0.4)
8 Cla: Okay well then u-wi’ll see: you: Sa’urde [e.
9 Chl: [Saturdee night.
10 Cla: Seven thirty?
11 (.)
12 Chl: Ya[h.
13 Cla: + [hhhh D’you want me to bring the: chai: [rs?
14 Chl: [hahh
15 Chl: Plea::: (.) NO: (0.2) ºYah,º
16 (0.3)
17 Chl: I:’ve got to get chairs. Bring ’em one more time.
Plainly Chloe does not indicate that she has any diffi culty as regards seating, or
that she needs chairs. Chloe is educing (inferring, or bringing out something
which was latent) from just a little earlier in their conversation (4 minutes
before), in which they resolved the matter of Claire not having score tallies
by Chloe agreeing to make and bring them.
Example 7 (SBL:2:2:25)
1 Chl: hhh Now waita minute Claire don’t we need tallies?
2 (1.8)
3 Chl: º( ),º
4 (0.5)
5 Cla: YEH w’l why dun’ [I j’s ] make ]up ] Why c- ]
6 Chl: [W’dju ] bring ]the ] tallie ][s?
7 Cla: [Why can’I d’s
8 make those thin:gs up I made bef:o::re. Dz it haftuh be so
9 ni:[ce?
10 Chl: [e-Heavens n [o make ‘em up.[h
11 Cla: [.hhhh [Oka:y.
Example 7 is taken from earlier in the call in which Example 6 occurs. From
this, and evidently from previous experience – see Chloe’s ‘I:’ve got to get
chairs. Bring ’em one more time.’ in line 17, Example 6 – Claire infers
something else (chairs) which Chloe might need (and apparently does so
correctly). So offers done syntactically as Do you want…? interrogatives are
made in response to a diffi culty or need which is found to be implicit (never
explicit) in something said earlier in the talk (never in the immediately prior
Thus each of the three different syntactic forms for offering have their
characteristic and appropriate ‘places’ or uses; each of the forms is used to
handle or deal with different interactional circumstances or contingencies. It
is evident that speakers orient to the appropriateness of a given form when
Overview and new directions 33
making an offer, in cases where they begin using a certain ‘inappropriate’
form (e.g. Do you want … ? in response to an explicit expression of trouble)
but do not complete it, and then self-repair their talk so as to select the
correct/appropriate construction.
The connections we’ve explored and illustrated briefl y here, between
syntactic form and interactional circumstances or contingencies, is a
relatively new direction for CA research – one which promises to open up
signifi cant aspects of the interface between CA and core linguistic areas, and
promises also signifi cant ndings regarding the role played by features of
linguistic design, here syntax, in participating meaningfully and coherently
in talk-in-interaction. This is by no means the only new direction taken
by CA research; others include the application of CA’s methodology to
our understanding of workplace or institutional interactions, especially
medical consultations (Heritage and Maynard 2006). But the direction we
have outlined here is one that promises fresh insights into the linguistic
design of social action in talk – by focusing not so much on responses
to actions, and the subsequent unfolding sequences, which have hitherto
been the main line of CA enquiry (as we noted before, for instance in
the research on adjacency pairs), but instead on how those initial forms
emerge in the fi rst place, and how they are constructed linguistically. This
complements, and might contribute to, the work on ‘emergent grammar
(e.g. Hopper 1998).
Atkinson, J.M. and Heritage, J. (eds) (1984) Structures of Social Action: studies in
conversation analysis, Cambridge: Cambridge University Press.
Couper-Kuhlen, E. (1996) ‘The prosody of repetition: on quoting and mimicry’,
in E. Couper-Kuhlen and M. Selting (eds) Prosody in Conversation, Cambridge:
Cambridge University Press, pp. 366–405.
Couper-Kuhlen, E. and Selting, M. (eds) (1996) Prosody in Conversation, Cambridge:
Cambridge University Press.
Curl, T.S. (2006) ‘Offers of assistance: constraints on syntactic design’, Journal of
Pragmatics, 38: 1257–80.
Curl, T., Drew, P. and Ogden, R. (forthcoming) Linguistic Resources for Social Action,
Cambridge: Cambridge University Press.
Drew, P. (1984) ‘Speakers’ “reportings” in invitation sequences’, in J.M. Atkinson
and J. Heritage (eds) Structures of Social Action: studies in conversation analysis,
Cambridge: Cambridge University Press, pp. 129–51.
—— (2005) ‘Conversation analysis’, in K. Fitch and R. Sanders (eds) Handbook of
Language and Social Interaction, Mahwah, NJ: Lawrence Erlbaum, pp. 71–102.
Drew, P. and Heritage, J. (1992) Talk at Work: interaction in institutional settings,
Cambridge: Cambridge University Press.
Drew, P. and Holt, E. (1998) ‘Figures of speech: fi gurative expressions and the
management of topic transition in conversation’, Language in Society, 27: 495–
34 Paul Drew and Traci Curl
Gardner, R. (1997) ‘The conversational object Mm: a weak and variable acknowledging
token’, Research on Language and Social Interaction, 30: 131–56.
Garfi nkel, H. (1967) Studies in Ethnomethodology, Englewood Cliffs, NJ: Prentice
Glenn, P. (2003) Laughter in Interaction, Cambridge: Cambridge University Press.
Goffman, E. (1983a) ‘The interaction order’, American Sociological Review, 48:
—— (1983b) ‘Felicity’s condition’, American Journal of Sociology, 89: 1–53.
Haakana, M. (2001) ‘Laughter as a patient’s resource: dealing with delicate aspects
of medical interaction’, Text, 21: 187–219.
Heinemann,T. (2006) ‘ “Will you or can’t you?” Displaying entitlement in interrogative
requests’, Journal of Pragmatics, 38: 1081–104.
Heritage, J. (1984a) Garfi nkel and Ethnomethodology, Cambridge: Polity Press.
—— (1984b) ‘A change-of-state token and aspects of its sequential placement’,
in J.M. Atkinson and J. Heritage (eds) (1984) Structures of Social Action:
studies in conversation analysis, Cambridge: Cambridge University Press,
pp. 299–345.
—— (2002a) ‘Oh-prefaced responses to assessments: a method of modifying
agreement/disagreement’, in C.E. Ford, B. Fox and S. Thompson (eds) The
Language of Turn and Sequence, Oxford: Oxford University Press. pp. 196–234.
—— (2002b) ‘The limits of questioning: negative interrogatives and hostile question
content’, Journal of Pragmatics, 34: 1427–46.
Heritage, J. and Maynard, D. (eds) (2006) Communication in Medical Care:
interaction between physicians and patients, Cambridge: Cambridge University
Hopper, P.J. (1998) ‘Emergent grammar’, in M. Tomasello (ed.) The New Psychology
of Language: cognitive and functional approaches to language structure, Mahwah,
NJ: Lawrence Erlbaum, pp. 155–75.
Jefferson, G. (1979) ‘A technique for inviting laughter, and its subsequent acceptance/
declination’, in G. Psathas (ed.) Everyday Language: studies in ethnomethodology,
New York: Irvington, pp. 79–96.
Ochs, E., Schegloff, E.A. and Thompson, S. (1996) Interaction and Grammar,
Cambridge: Cambridge University Press.
Pomerantz, A. and Mandelbaum, J. (2005) ‘Conversation analytic approaches to
the relevance and uses of relationship categories in interaction’, in K. Fitch and
R. Sanders (eds) Handbook of Language and Social Interaction, Mahwah, NJ:
Lawrence Erlbaum, pp. 149–70.
Sacks, H. (1974) ‘An analysis of the course of a joke’s telling in conversation’, in
R. Bauman and J. Sherzer (eds) Explorations in the Ethnography of Speaking,
Cambridge: Cambridge University Press, pp. 337–53.
—— (1992) Lectures on Conversation, 2 volumes, G. Jefferson (ed.) Oxford:
Schegloff, E.A. (1992) ‘Introduction’, in H. Sacks (1992) Lectures on Conversation
Volume 1, G. Jefferson (ed.), Oxford: Blackwell, pp. ix–lxii.
—— (1996) ‘Confi rming allusions: toward an empirical account of action’, American
Journal of Sociology, 104(1): 161–216.
—— (2007) Sequence Organization in Interaction: a primer in conversation analysis
I, Cambridge: Cambridge University Press.
Overview and new directions 35
Sorjonen, M.-J. (2001) Responding in Conversation: a study of response particles in
Finnish, Amsterdam: John Benjamins.
Tanaka, H. (1999) Turn-Taking in Japanese Conversation: a study in grammar and
interaction, Amsterdam: John Benjamins.
ten Have, P. (1999). Doing Conversation Analysis: a practical guide, Thousand Oaks,
CA: Sage.
3 Being out of order
Overlapping talk as evidence of
trouble in airline pilots’ work
Maurice Nevile
Research in conversation analysis (CA) uses recordings of naturally occurring
interaction to uncover the language, practices and processes of reasoning
by which people accomplish social actions and create the intelligible and
recognizable orderliness of everyday life (see Sacks 1992; Hutchby and
Wooffi tt 1998; ten Have 1999; Wooffi tt 2005). As Drew and Curl (this
volume) explain, CA shows how patterns in talk reveal how participants
produce and understand conduct in interaction, in real time. Drew and
Curl (this volume) point to the range of aspects of interaction on which CA
focuses, such as turn-taking, the organization of conversational sequences,
the structure of turns at talk, the actions that participants undertake, repair
(dealing with diffi culties in interaction), as well as prosodic and embodied
details of interaction. In this chapter I contribute to an expanding direction
for research in CA, interaction in institutions and workplaces (Drew and
Heritage 1992; McHoul and Rapley 2001; Arminen 2005), and focus on
an area where CA is well suited to advance discourse analysis: how talk for
work is organized in time. Participants at work can time their relative turns at
talk, moment-to-moment, in ways that realize the interests, demands, goals
and constraints of the setting, and so to accomplish work acceptably. I will
examine transcriptions from recordings of naturally occurring interaction
from one specifi c collaborative work setting: the airline cockpit.
Generally, I am interested here in the signifi cance of timing in the sequential
organization of turns in interaction as evidence for how airline pilots create
and understand the progress of their work. Specifi cally, I examine moments
when two pilots talk simultaneously, what CA describes as overlapping talk.
I will suggest that in the airline cockpit such moments can signify trouble
in the fl ow of talk for work. Pilots talking in overlap are talking out of
order, because overlapping talk is a departure from the typical temporal and
sequential order of cockpit talk. Pilots almost always allow one another’s
talk to emerge complete and in the clear, and so talk only one at a time
(Nevile 2007b). Overlapping talk is therefore evidence in talk’s very timing
of something non-routine, maybe even problematic, in pilots’ work together,
Overlapping talk among airline pilots 37
or for the fl ight’s progress. I conclude by pointing to the future potential
value of such CA-based research for developing work practice.
Conversation analysis and temporal order in interaction
An interest in temporal order is a fundamental feature of CA. Analysts have
shown the importance to participants themselves of just how and when this
or that happens in interaction, to know moment-to-moment just what it is
that they are doing and what is going on, what Arminen (2005: x) refers to
as the ‘time-bound fabric of social actions’. CA is fundamentally concerned
with how participants organize matters of when and what next, and this
is refl ected in CA’s guiding analytic principle: why that now? Some CA
studies of work, and related research in ethnomethodology, have focused
directly on the impact of time for organizing interaction for collaborative
work (e.g., Lynch et al. 1983; Ochs and Jacoby 1997; Button and Sharrock
1998; Goodwin 1994, 2002).
These studies examine how participants
collaborate to time their conduct appropriately to accomplish setting-specifi c
work goals, including how participants produce and coordinate their talk
and non-talk activities (e.g. gestures, gaze, etc.), use objects and resources,
act relative to physical features of the local environment, and respond to
evolving circumstances. This chapter furthers a line of research examining
how pilots accomplish timeliness and order for talk for work (Nevile 2004a,
2004b, 2005a, 2005b, 2006). I will focus on moments of overlapping talk as
evidence of occasions when that timeliness and order is vulnerable (see also
Nevile 2007a; Nevile and Walker 2005).
Overlapping talk
The model of turn-taking for ordinary conversation at the heart of CA
accounts for the fact that, while usually one party talks at a time, moments
of overlapping talk do commonly occur (Sacks et al. 1974). More often than
not, however, such moments can be accounted for (e.g. as interruption),
and participants have various resources for dealing with them (see Schegloff
2000 for a substantial overview of research on overlap, and also Schegloff
2002; Jefferson 2004; and Lerner 2004). In pilots’ talk, on the other hand,
overlapping talk is unusual, occurring (at most) less than a handful of times
per fl ight (Nevile 2007b; see data in Nevile 2004a). At fi rst this might seem
unsurprising, because the content, allocation and order of pilots’ turns at
talk are scripted for them in advance, and pilots are instructed in training
not to speak simultaneously. However, pilots do not always talk according to
the script (Nevile 2001, 2004a, 2005a), and to speak one party at a time is
something pilots must manage and accomplish in interaction, in situ in real
time (Nevile 2007b).
Research in CA has shown that a very common type of overlapping talk
occurs when one participant starts talking just as the other speaker comes to
38 Maurice Nevile
a recognizably possible end of their turn. A recipient projects the end of the
speaker’s turn and starts up as next speaker just a little early (see Schegloff
2000), for example to increase the chances of emerging as next speaker, or
to gain something by demonstrating and acting early on an understanding
of the turn’s trajectory. Airline pilots, however, seem not to do this, even
though their talk is highly predictable and projectable. Pilots follow offi cial
procedural wordings that act as a script specifying who is to say what, and
when. Instead, overlapping talk mostly does not occur because pilots are
oriented to precisely time new talk to start at the actual end of another’s
turn, when current talk is actually ended (Nevile 2007b). An absence of
overlapping talk refl ects pilots’ orientation to the strictly sequential nature
of their work. Pilots treat turns at talk as they treat the tasks they perform
to fl y their aircraft, as acceptably occurring such that a next one is begun
only when a prior one is complete. A new turn, like a new task, becomes
relevant only when another has actually been completed. So, what might
be happening on those occasions when pilots do nd themselves talking
simultaneously, in overlap?
The data
The transcriptions here are made from audio and video recordings of pilots
at work on actual routine passenger fl ights. Some transcriptions are from
recordings I made by arrangement with two Australian airlines. In total
I made 18 fl ights that varied in length from around 40 minutes to over
two hours.
Other transcriptions are from commercially produced cockpit
videos of German airlines. Segments here are therefore taken from different
airlines, fl ights and crews. The transcriptions include technical language,
and so for clarity and space I explain only the particular point of interest,
and occasionally omit lines of talk not relevant to that point (e.g. overheard
radio talk). To preserve anonymity of individual airlines I have changed
some wording and use the generic fl ight descriptor ‘Airline One Two Three’.
Where necessary I have provided translations from German to English. The
transcription system, simplifi ed from the notation originally developed by
Gail Jefferson (see e.g. ten Have 1999) is given in full at the end of this
Being out of order
I will consider segments when talk is treated as expected, relevantly next,
but delayed, and when one or both pilots treat circumstances as uncertain
and needing resolution. In either case, overlapping talk is a departure from
the orderly fl ow of talk and action and is evidence of trouble in the pilots’
progress through the tasks required to conduct the fl ight.
Overlapping talk among airline pilots 39
The fi rst two segments show overlapping talk occurring when one pilot
treats another’s talk as delayed, when a pilot does not respond with next talk
that could be expected according to formal procedures for the sequential
ordering for tasks. Delayed talk signals trouble because it can block the
pilots’ progress to the next talk and so through the sequence of tasks for
the fl ight (Nevile 2007a). The pilot last to speak treats the delayed talk as
accountable by starting up again as speaker to offer a new version of the prior
talk, or to initiate the next task regardless. The moment of overlapping talk
occurs when the other pilot almost simultaneously produces the expected
but delayed talk.
In this fi rst example the pilots fi nalize preparations for takeoff.
Captain (C) calls (‘it’s your go’, line 3) for the First Offi cer (FO) to say
information for conducting the takeoff, as the Pilot-Flying (PF) in control
of the aircraft on this fl ight (lines 5–6). The First Offi cer concludes with ‘no
changes to the brief’, (line 6), making relevant an acknowledging response
from the Captain. However, this acknowledgement from the Captain is
not immediately forthcoming. There are 1.9 seconds of silence before the
First Offi cer pursues the Captain’s acknowledgment (Pomerantz 1984) by
remodelling his concluding wording with ‘as discussed’. The First Offi cer
nds himself talking in overlap as the Captain almost simultaneously begins
the delayed acknowledgement.
1 (2.0)
2 C: ( ) give us seven four we’ll cross ah: one hundred on a heading bug
3 (0.9) both to ADF (0.5) a:nd it’s your go.
4 (2.5)
5 FO: okay: go-around nine thousand ASEL (0.3) right comma:nd (0.5) and
6 ah (0.8) no changes to the brief,
7 (1.9)
8 FO: a[s discussed].
9 C: [that’s::]: understood.
10 (4.3)
11 C: a:nd the checks when you’re ready hhh- =
12 FO: =checks,
13 (1.2)
In the second segment a Captain releases the ‘fl ight control lock’ (‘lock’s
coming off’, line 6), making it now possible for him to call for the First
Offi cer to resume a suspended checklist. That would be a next relevant task
once the lock comes off. After ten seconds pass the First Offi cer treats that call
as delayed. He begins to talk for the next item of the checklist (‘fl i-’, line 8),
which is the fl ight control lock, without being called by the Captain to do
40 Maurice Nevile
so. However, the Captain almost simultaneously makes the expected call for
resuming the checklist (line 9), and so the pilots talk briefl y in overlap.
1 (7.2)
2 C: (I) might just turn around eh?
3 (0.4)
4 FO: ye:p (.) sounds good.
5 (13.9)
6 C: o:kay: lock’s coming off,
7 (10.1)
8 FO: [i-
9 C: [and the rest of the check.
10 FO: ight controls,
11 (0.3)
12 C: checked,
13 FO: checked and takeoff clearance,
14 (0.5)
15 C: not required here:
16 (0.3)
17 FO: okay (.) check’s complete.
18 (0.5)
Here there is only brief overlapping talk because the First Offi cer cuts off
his talk immediately after the Captain starts up. The First Offi cer only
resumes once the Captain completes. Despite starting up fi rst, the First
Offi cer concedes speakership to the Captain, and so orients to the Captain
as the legitimate next speaker who, according to procedure, should fi rst call
for the checklist to be resumed. That call makes it acceptable for the First
Offi cer to then say the ‘challenge’ for the next checklist item ‘fl ight controls’
(line 10).
In these fi rst two segments overlap occurred, breaking the typically orderly
ow of non-overlapping cockpit talk. One pilot treated another’s talk as
expected but delayed, and so making trouble for the progression of talk and
action through the sequence of fl ight tasks. Overlapping talk resulted when
the expected talk and remedy talk were initiated almost simultaneously.
The next three segments concern a critical cockpit task, receiving clearance
(permission) from Air Traffi c Control (ATC), and in each case overlapping talk
occurs when one or both pilots treat an aspect of the clearance as uncertain.
In the fi rst two examples the trouble concerns receipt of the clearance, and
in the last example the trouble concerns the content of the clearance.
The general procedure is that clearance for some action is given, over
the radio, to the pilot responsible for radio duties and acting as the fl ight’s
Overlapping talk among airline pilots 41
spokesperson – usually the Pilot-Not-Flying (PNF). Any radio talk to and from
controllers is potentially hearable to both pilots, but pilots follow procedures
to talk for establishing a shared crew awareness of a clearance (see Nevile
2004a for examples of this). Clearance is required for many aspects and
stages of a fl ight, and to act without clearance can be an extremely serious
and professionally (even legally) sanctionable lapse, especially because it
could lead to fl ight-threatening situations.
In the fi rst two segments, the overlapping talk is associated with a moment
when a Pilot-Flying, the pilot actually in control of the aircraft and who
does not speak to the controller, asks the other pilot if clearance has been
Here a German crew prepare their aircraft for landing. The PNF calls the
air traffi c controller to announce their position (‘established’, line 2). The
controller’s reply is too faint to be transcribed but the PNF heard it because
he ‘reads back’ that the fl ight is ‘cleared to land …’ (line 5). Importantly,
note that this radio exchange does not become the subject of talk between
the two pilots. Instead, 20.5 seconds pass before talk for a new task, then a
further 18.8 seconds before the PNF tells the PF ‘>we have been< cleared
to land.’ (line 11). It would seem the pilots then establish a shared awareness
of the clearance (lines 12–14). However, overlapping talk occurs later as the
PNF concludes talk for a different task, the landing checklist, the third task
since the talk of the clearance (lowering landing gear, extending wing fl aps,
landing checklist). As the PNF says ‘completed’, the PF asks in overlap, ‘and
we’ve been cleared to land?’ (line 24).
1 PNF: Muenchen contro:l gruss Gott Airline (.) One Two ºThree Fourº
2 established.˚
3 (1.6)
4 ATC: {
5 PNF: cleared to land zero eight left Airline One Two ºThree Four
6 (20.5)
7 PF: aps two.
8 (0.8)
9 PNF: aps two.
10 (18.8)
11 PNF: >we have been< cleared to land.=
12 PF: =cleared to land?=
13 PNF: =ja. {‘yes’}
14 PF: okay::.
15 (2.7)
16 PF: gear down.
17 (0.5)
18 PNF: gear down.
42 Maurice Nevile
19 (4.4)
20 PF: landi:ng (0.2) checklist.=
21 PNF: =landi:ng (.) all green.
22 PF: landing all green.
23 PNF: landing checklist complet[ed::. ]
24 PF: [and we’ve] been cleared to land?
25 PNF: jawohl. {‘yes certainly’}
26 (24.0)
The PF asks if the fl ight has landing clearance, even though this matter was
covered earlier. It is evidence the PF is still uncertain, but why might he be so?
Perhaps the controller’s faint (non-transcribable) talk (line 4), and the PNF’s
delay in formally announcing the clearance (line 11), made the clearance less
salient for the PF. The PF did not himself speak to the controller. Also, the
earlier sequence of talk to establish crew awareness of the clearance was itself
atypical. An unremarkable PF reply would have been an acknowledgement
by repeating key wording, with fl at or falling intonation, as in ‘cleared to
land.’. However, the PF’s reply is produced with marked rising intonation,
as a checking question. We can see that the PNF treats it as a question
because he replies ‘ja.’, to which the PF responds ‘okay::.’. This four-turn
sequence is overly elaborate relative to routine untroublesome cockpit talk
(see Nevile 2004a). It can therefore be evidence that acting on an awareness
of the landing clearance is proving possibly problematic for the PF. So we
see overlapping talk occurring, talk out of order, when a matter is treated as
uncertain and troublesome for conducting tasks for the fl ight.
In the next segment, also concerning the landing phase of fl ight, talk
to establish shared awareness of a clearance is not delayed, in fact it does
not occur. The controller tells the Captain the fl ight is cleared for ‘a visual
approach’ (line 2), but the Captain does not initiate crew talk about this,
nor include the ‘visual’ clearance in his readback to the controller (which
is possibly hearable to the First Offi cer) (line 5). Instead, after talking to
the controller, the Captain prompts the First Offi cer to discuss speeds to be
used (line 7) (the First Offi cer is the Pilot-Flying on this fl ight). Much later
the First Offi cer treats the clearance as uncertain, by asking, ‘we’re cleared
for a visual approach?’ (line 13). Overlapping talk occurs when the Captain
1 ATC: >Airline One Two Three< ah (.) two miles east centreline and closing,
2 reduce to fi nal approach speed er (1.4) cleared a visual approach
3 caution: wake turbulence, (.) contact tower on fi nal.
4 (1.5)
5 C: reduce to (.) fi nal approach, tower on fi nal, (.) Airline One Two Three.
6 (0.9)
7 C: o::kay:: speeds,
8 FO: okay I might (0.3) decrease: ah: (0.7) to:: one hundred and eight,
Overlapping talk among airline pilots 43
10 FO: for the approach,
11 (1.8)
12 (1.1)
13 FO: we’re cleared for a visual approach?=
14 C: =ye:s [we’re] cleared.
15 FO: [okay]
16 FO: and (gi::ve) (.) spinner (.) standby (spinner out) thanks.
17 (1.8)
The overlap occurs when the Captain extends his answer beyond a simple
‘yes’ by adding ‘we’re cleared’ (line 14). The First Offi cer starts up after ‘yes’,
a possible point of completion for the Captain’s response, and so his ‘okay
(line 15) occurs in overlap with the Captain’s talk. So, treating a matter as
uncertain is again associated with a moment of overlapping talk, a break in
the typical ordered fl ow of cockpit talk.
This last segment is from a German airline fl ight and includes three
instances of overlapping talk. It is a complicated segment but it is particularly
revealing of what can go on behind the closed cockpit door. The crew treat
as uncertain the taxiway route to take as they prepare to taxi to the runway
for takeoff. The controller responsible for aircraft ground movements tells
the First Offi cer his aircraft is to use taxiway ‘delta three’ (line 4). The pilots
do not discuss this part of the taxi clearance. Some seconds later, the Captain
asks the First Offi cer about the taxiway (line 8). The Captain identifi es ‘delta
two’, which is incorrect (it is not the taxiway specifi ed by the controller),
but the First Offi cer incorrectly agrees (line 9). The crew could now make a
very serious error by using the wrong taxiway. However, a couple of seconds
later the Captain asks the First Offi cer to confi rm with the controller (line
12). The Captain treats the matter as still uncertain. The First Offi cer makes
the call (lines 15–19, 23) and both pilots acknowledge the controller’s reply
with ‘okay’ (lines 21–2). I will focus fi rst on the overlap occurring later at
line 27.
1 FO: and ground Airline (0.3) one two three request taxi.
2 (1.5)
3 ATC: Airline: one two three (0.2) taxi: (.) cross the bridge (0.5) right
4 turn (.) charlie one (0.4) delta three (0.2) to hold behind bay two.
5 (1.5)
6 FO: cross the bridge (.) charlie one delta three:: (0.5) fo:r (0.2) runway
7 one three Airline >one two three<.
8 C: via delta two hat er uns gegeben ne? {‘he gave us via Delta two yeah?’}
9 FO: via charlie one and delta two ja. {‘via charlie one and delta two yes’}
44 Maurice Nevile
10 (1.2)
11 (0.8)
12 C: confi rm Sie bitte noch[mal {‘please confi rm it again’}
13 FO: [ja {‘yes’}
14 (1.5)
15 FO: a [nd ah: (0.2) ground] Airline one two three confi rm via delta two
16 C: [con