The Pragmatics of Interaction

The Pragmatics of Interaction
Handbook of Pragmatics Highlights (HoPH)
The ten volumes of Handbook of Pragmatics Highlights focus on the most salient
topics in the field of pragmatics, thus dividing its wide interdisciplinary spectrum
in a transparent and manageable way. Each volume starts with an up-to-date
overview of its field of interest and brings together some 12–20 entries on its most
pertinent aspects.
Since 1995 the Handbook of Pragmatics (HoP) and the HoP Online (in conjunction
with the Bibliography of Pragmatics Online) have provided continuously updated
state-of-the-art information for students and researchers interested in the science
of language in use. Their value as a basic reference tool is now enhanced with the
publication of a topically organized series of paperbacks presenting HoP Highlights.
Whether your interests are predominantly philosophical, cognitive, grammatical,
social, cultural, variational, interactional, or discursive, the HoP Highlights volumes
make sure you always have the most relevant encyclopedic articles at your fingertips.
Editors
Jef Verschueren Jan-Ola Östman
University of Antwerp University of Helsinki
Volume 4
Edited by Sigurd D’hondt, Jan-Ola Östman and Jef Verschueren
Edited by
Sigurd D’hondt
Ghent University
Jan-Ola Östman
University of Helsinki
Jef Verschueren
University of Antwerp
John Benjamins Publishing Company

Amsterdam / Philadelphia
TM The paper used in this publication meets the minimum requirements of American
8
National Standard for Information Sciences — Permanence of Paper for Printed
Library Materials, ANSI Z39.48-1984.
Cover design: Françoise Berserik
Library of Congress Cataloging-in-Publication Data
The pragmatics of interaction / edited by Sigurd D’hondt, Jan-Ola Östman, Jef Verschueren.
p. cm. (Handbook of Pragmatics Highlights, issn 1877-654X ; v. 4)
Includes bibliographical references and index.
1. Pragmatics. 2. Social interaction. I. D’hondt, Sigurd. II. Östman, Jan-Ola. III. Verschueren,
Jef.
P99.4.P72P743 2009
306.44--dc22 2009022919
isbn 978 90 272 0781 4 (pb; alk. paper)
isbn 978 90 272 8919 3 (EB)
© 2009 – John Benjamins B.V.

No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means,
without written permission from the publisher.
John Benjamins Publishing Company • P.O. Box 36224 • 1020 me Amsterdam • The Netherlands
John Benjamins North America • P.O. Box 27519 • Philadelphia PA 19118-0519 • USA
Table of contents
Preface to the series xi
Acknowledgements xiii
The pragmatics of interaction: A survey 1

Sigurd D’hondt
1. Layers of interactional organization 2
2. Context, ethnography and categorization 3
3. Multimodality and mediation 8
4. Style and indexicality 11
5. This volume: Interaction as a topic 12
Communicative style 20
Margret Selting
1. Definition, delimitation, basic concepts 20
2. A few landmark reference works 21
3. Problems 25
4. Sample data and methodology of an interactional stylistic analysis 26
4.1 The first intuitive analysis of speech styles in the
given sequential context 29
4.2 Structural analysis: Decomposition/deconstruction 29
4.2.1 Recipient reaction after this first part of the story telling 31
4.3 Functional analysis 32
4.4 Warranting 32
4.5 Structural analysis: Decomposition/deconstruction 33
4.6 Functional analysis 33
4.7 Warranting 34
5. Perspectives for future research 37
Conversation analysis 40
Rebecca Clift, Paul Drew & Ian Hutchby
1. Introduction 40
2. Origins and overview 40
VI Handbook of Pragmatics Highlights
3. Data, transcription and analysis 43

4. Exhibiting an understanding in next turn 46
5. Conditional relevance of next position 48
6. Conclusions 50
Conversation types 55
Auli Hakulinen
1. Introduction 55
2. Three basic dimensions 56
2.1 The channel 56
2.2 Dyadic vs. multi-person 57
2.3 Everyday vs. institutional 58
3. Types of institutional talk 59
4. Symmetry and asymmetry in conversations 60
5. Conversation types and communicative genres 62
6. Conclusion 63
Ethnomethodology 66
Alan Firth
1. Introduction 66
2. Overview 67
3. Social action, social knowledge 69
3.1 Norms and rules 69
3.2 The contexted character of actions 70
3.2.1 Indexicality 70
3.2.2 Reflexivity 71
3.3 Rationality 73
4. Commonsense reasoning 73
5. Developments in ethnomethodology 75
6. Conclusion 76
Erving Goffman 79
Jim O’Driscoll
1. Introduction 79
2. The primacy of the situation 80
3. Ritual and the sacred self 83
4. Goffman’s working framework 85
5. Goffman’s influence and significance 89
Table of contents VII
Interactional linguistics 96
Jan Lindström
1. Background 96
2. Points of departure 97
3. Topics 99
4. Possibilities and challenges 100
Listener response 104

Deng Xudong
1. Introduction 104
2. Approaches to the study of listener responses 105
2.1 The lumping approach 105
2.1.1 Structural properties of listener response 106
2.1.2 Roles and functions of listener responses
in conversation 108
2.2 The splitting approach 111
3. Classification of listener response 113
4. Cross-cultural study of listener response 114
5. Gender-differentiated use of listener response 119
6. Future research 120
Participation 125
Jack Sidnell
1. “Phatic communion” and the practices of participation 125
2. Goffman: Attention, involvement and focused encounters 133
3. Goffman: Footing 139
4. Elaborations and critique of footing 149
5. Conclusion 153
Politeness 157
Gabriele Kasper
1. Historical overview 157
2. Approaches to politeness 157
2.1 Folk notion 157
2.2 Conversational maxim(s) 158
2.3 Redress to face-threat 160
2.4 Social marking 160
2.5 Conversational contract 161
VIII Handbook of Pragmatics Highlights
2.6 Politeness and politic behavior 162

2.7 Politeness and tact 162
3. Expression of politeness 162
3.1 Inherently polite speech acts? 162
3.2 Conventions of means 163
3.3 Conventions of form 164
4. Variables in politeness investment 165
5. Discourse perspective 168
6. Further reading 169
Prosody 174
Elizabeth Couper-Kuhlen
1. Prosody defined 174
2. Prosody as a pragmatic phenomenon 175
3. Prosody and early work on spoken discourse 176
4. Prosody in talk-in-interaction: Structural dimensions 178
4.1 Turn construction 178
4.2 Sequential organization 178
4.3 Floor management 179
5. Prosody in talk-in-interaction: Interactional dimensions 180
5.1 Prosodic routines for action 181
5.2 Prosodic cueing of stance and affect 182
6. Prosody in talk-in-interaction: A case study 182
7. Directions for future research 186
Reported speech 190

Elizabeth Holt
1. Introduction 190
2. Influential figures 191
2.1 Vološinov and Bakhtin 191
2.2 Goffman 191
3. Forms of reported speech 192
3.1 Categories and terminology 192
3.2 The reporting clause 194
4. Reported speech in discourse 195
4.1 The authenticity of reported speech 195
4.2 Reported speech in storytelling 198
4.3 The interactional environments of reported speech 199
5. Conclusion 201
Table of contents IX
Harvey Sacks 206

Rod Watson
Sequence 215
Jack Sidnell
1. Introduction 215
2. The adjacency pair 220
3. “A context of publicly displayed and continuously up-dated
intersubjective understandings” 223
4. Preference 227
5. Structural consequences of preference organization 229
6. Sequence organization 230
7. The power of sequential analysis 235
Transcription systems for spoken discourse 240

Daniel C. O’Connell & Sabine Kowal
1. Transcription: Basic terminology 240
2. Speaking: The behavior under consideration 242
2.1 The verbal component 242
2.2 The prosodic component 243
2.3 The paralinguistic component 244
2.4 The extralinguistic component 244
3. Current transcription systems 245
3.1 Du Bois’ discourse transcription (DT) 246
3.2 Ehlich’s heuristic interpretative auditory
transcription (HIAT) 247
3.3 The transcription system of Gumperz & Berenz 248
3.4 The Jeffersonian tradition 248
3.5 MacWhinney’s CHAT system for the CHILDES project 249
4. Conclusion: Basic principles for scientific use of transcription 250
Index 255
Preface to the series
In 1995, the first installments of the Handbook of Pragmatics (HoP) were published.
The HoP was to be one of the major tools of the International Pragmatics Association
(IPrA) to achieve its goals (i) of disseminating knowledge about pragmatic aspects of
language, (ii) of stimulating various fields of application by making this knowledge
accessible to an interdisciplinary community of scholars approaching the same gen-
eral subject area from different points of view and with different methodologies, and
(iii) of finding, in the process, a significant degree of theoretical coherence.
The HoP approaches pragmatics as the cognitive, social, and cultural science of lan-
guage and communication. Its ambition is to provide a practical and theoretical tool for
achieving coherence in the discipline, for achieving cross-disciplinary intelligibility in
a necessarily diversified field of scholarship. It was therefore designed to provide easy
access for scholars with widely divergent backgrounds but with converging interests
in the use and functioning of language, in the topics, traditions, and methods which,
together, make up the broadly conceived field of pragmatics. As it was also meant to
provide a state-of-the-art report, a flexible publishing format was needed. This is why
the print version took the form of a background manual followed by annual loose-
leaf installments, enabling the creation of a continuously updatable and expandable
reference work. The flexibility of this format vastly increased with the introduction
of an online version, the Handbook of Pragmatics Online (see www.benjamins.com/
online).
While the HoP and the HoP-online continue to provide state-of-the-art informa-
tion for students and researchers interested in the science of language use, this new
series of Handbook of Pragmatics Highlights focuses on the most salient topics in
the field of pragmatics, thus dividing its wide interdisciplinary spectrum in a trans-
parent and manageable way. The series contains a total of ten volumes around the
following themes:
–– Key notions for pragmatics

–– Philosophical perspectives
–– Grammar, meaning and pragmatics
–– Cognition and pragmatics
–– Society and language use
–– Culture and language use
–– The pragmatics of variation and change
–– The pragmatics of interaction
XII Handbook of Pragmatics Highlights
–– Discursive pragmatics
–– Pragmatics in practice
This topically organized series of paperbacks, each starting with an up-to-date over-
view of its field of interest, each brings together some 12–20 of the most pertinent
HoP entries in its respective field. They are intended to make sure that students and
researchers alike, whether their interests are predominantly philosophical, cognitive,
grammatical, social, cultural, variational, interactional, or discursive, can always have
the most relevant encyclopedic articles at their fingertips. Affordability, topical organi-
zation and selectivity also turn these books into practical teaching tools which can be
used as reading materials for a wide range of pragmatics-related linguistics courses.
With this endeavor, we hope to make a further contribution to the goals underly-
ing the HoP project when it was first conceived in the early 1990’s.
Jan-Ola Östman (University of Helsinki) &

Jef Verschueren (University of Antwerp)
Acknowledgements
A project of the HoP type cannot be successfully started, let alone completed, with-
out the help of dozens, even hundreds of scholars. First of all, there are the authors
themselves, who sometimes had to work under extreme conditions of time pressure.
Further, most members of the IPrA Consultation Board have occasionally, and some
repeatedly, been called upon to review contributions. Innumerable additional scholars
were thanked in the initial versions of handbook entries. All this makes the Handbook
of Pragmatics a truly joint endeavor by the pragmatics community world-wide. We are
greatly indebted to you all.
We do want to specifically mention the important contributions over the years
of three scholars: the co-editors of the Manual and the first eight annual installments,
Jan Blommaert and Chris Bulcaen were central to the realization of the project, and
so was our editorial collaborator over the last four years, Eline Versluys. Our sincerest
thanks to all of them.
The Handbook of Pragmatics project is being carried out in the framework of the
research program of the IPrA Research Center / Antwerp Center for Pragmatics at the
University of Antwerp. We are indebted to the university for providing an environ-
ment that facilitates and nurtures our work.
Jan-Ola Östman (University of Helsinki) &

Jef Verschueren (University of Antwerp)
The pragmatics of interaction
A survey
Sigurd D’hondt
Ghent University
Each one of the following articles outlines a tradition, presents a basic analytical con-
cept, or sketches the contribution of a particular author that, in one way or another,
enhances our understanding of naturally occurring interaction as a socially organized
activity. The domain that will be covered is defined rather loosely. A first requirement
for inclusion in the present volume is that the Handbook entry in question is con-
cerned with the empirical investigation of how human beings organize their exchanges
in natural settings. Second, each entry focuses on speech as a form of social action.
Third, the articles concentrate on how these actions are practically accomplished, thus
taking the analysis beyond the level of ‘what is said’ to that of the interactional organi-
zation of speech and action.
The intellectual origins and disciplinary foundations of this enterprise are mani-
fold. It bears the imprint, first of all, of a number of developments in sociology that
specifically target the analysis of interaction. Among these, Erving Goffman’s semi-
nal explorations of the interaction order stand out as some of the most prominent
(O’Driscoll in this volume, and also the chapters by Sidnell on participation and Kasper
on politeness). Equally influential is the tradition of conversation analysis (Clift, Drew
and Hutchby, this volume, and the chapter by Sidnell on sequence), which came to
fruition under the aegis of Garfinkel’s ethnomethodology (Garfinkel 1967, Firth in
this volume) and can be traced to Harvey Sacks’ decision to turn to conversation
as a privileged site for elucidating practical reasoning (Sacks 1992; Watson in this
volume). A second impetus came from anthropology. Against the background of the
anthropological study of the role language plays in the construction of the different set-
tings and events that make up the social life of particular communities (Hymes 1966,
1972), an ‘interactional’ approach to sociolinguistics (Gumperz 1982a, 1982b; Auer &
di Luzio 1992; Eerdmans, Prevignano & Thibault 2003) developed, which focuses on
the communicative basis of social processes in multi-ethnic industrialized societies,
on intercultural friction, ethnicity, and inequality. In the background, linguistic prag-
matics (in the narrow sense, advocated by Levinson 1983 and heavily influenced by
the philosophical accounts of language use by Austin, Grice and Searle) continues to
exert its influence over the field of interaction studies. Silverstein (1976, 1992, 2003),
2 Sigurd D’hondt
for example, introduced a Peircean perspective on the linguistic sign, while Gricean
notions of implicature and inference have informed interactional sociolinguistics (e.g.,
Gumperz 1982a) and politeness theory (Kasper, this volume). Speech act theory, and
in particular the foundational role it attributes to speaker intentions, played the part
of scapegoat for conversation analysts’ proceduralistic accounts of social action (e.g.,
Schegloff 1992a). In the meantime, the analysis of interaction also fertilized various
other domains, including psychology (Edwards & Potter 1992; Edwards 1997), edu-
cation studies (Varenne & McDermott 1998; Wortham 2008), medical sociology
(Heritage & Maynard 2006; Stivers 2006) and socio-legal studies (Travers & Manzo
1997; Matoesian 2001), and found a practical implementation in fields like the design
of human-computer interfaces (Arminen 2002).
1. Layers of interactional organization
Throughout his career (cf. the overview in O’Driscoll, this volume), Goffman insisted
that interaction embodies a structure in its own right, which can be reduced neither
to the psychological properties of the individual nor to the structural properties of
society. As it turns out, this interaction order is, in turn, composed of multiple lay-
ers or dimensions. Interaction is always multilayered, in the sense that the situated
accomplishment of an action always requires participants to attend to multiple orga-
nizational orders simultaneously. Conversation analysis (CA) contributed the fun-
damental insight that actions come in sequences. The notion of sequence refers to a
course of action in talk that is built collaboratively by different speakers in successive
turns (see Schegloff 2007 and the Sidnell chapter in this volume for state-of-the-art
overviews). The basic format of the sequence is the adjacency pair (Schegloff 1968).
Over the years, an extensive literature has developed on the way basic adjacency pairs
can be expanded by means of pre-, insertion-, and post-expansions (Schegloff 1980,
1990) and on the asymmetries that exist between different types of second pair parts
(‘preference organization’, Pomerantz 1984a). In addition to their detailed accounts
of the ‘sequence organization’ (Schegloff 2007) of interaction, conversation analysts
identified various other minute orders of interactional organization, sometimes met-
aphorically referred to as ‘machinery’, that are collectively implicated in making our
exchanges intelligible and guaranteeing intersubjectivity. These include the analysis of
turn-taking and the negotiation of rights to the floor (Sacks et al. 1974), the organiza-
tion of repair (Schegloff et al. 1977; Schegloff 1992b), the opening and closing of a con-
versation (Schegloff 1968; Schegloff & Sacks 1973) and a range of other phenomena,
including gaze (Goodwin 1981) and laughter (Jefferson 1984; Jefferson et al. 1987). CA
successfully demonstrated that (and how) these various phenomena are ‘sequentially
organized’ (to be distinguished from ‘sequence organization’ in the narrow sense,
The pragmatics of interaction 3
cf. Schegloff 2007) and thereby made other researchers aware of the overarching
force of relative positioning in producing and making sense of spoken discourse. The
ample evidence of the interactional significance of realizational detail supplied by
these analyses furthermore made clear that interaction can only be studied sensibly
on the basis of mechanical records (audio, video) and detailed transcripts of these
records (O’Connell & Kowal, this volume).
Interactional sociolinguistics contributed to the emerging picture by drawing
attention to the role of non-denotational aspects of talk in the inferential processes that
shape interaction. It complements sequential accounts of interaction, like those pro-
vided by conversation analysis, with a Peircean semiotic perspective (Silverstein 1992)
and recognizes that linguistic signs carry indexical meanings alongside referential ones.
These indexical meanings are based on a “direct linkage” between the sign and its con-
text that is “in no way mediated by any symbolic process” (Eerdmans 2003: 87). Thus,
Gumperz’ (1982a) account of contextualization describes how indexical meanings
associated with non-lexical aspects of the realization of an utterance (contextualization
cues, Couper-Kuhlen and Selting, both in this volume) trigger inferential processes on
the part of the hearer, inviting him/her to retrieve the contextual grounds of what is said
by situating the utterance into a complex of culturally based presuppositions. This infer-
ence process extends beyond the level of recognizing how subsequent utterances relate
to one another or detecting their action properties (the forms of ‘sequence organization’
that CA is interested in), but also includes presuppositions about the activity (Levinson
1992) carried out and about how the interaction is to be framed (Goffman 1974).
It should be kept in mind, furthermore, that various participants may be impli-
cated in the same interactional event in multiple ways. Advancing ‘co-presence’ as the
primary unit of analysis, Goffman (1981) contributed the fundamental insight that
dyadic models that describe interaction in terms of an exchange between a speaker
and a listener are insufficient for understanding the complexity of naturally occurring
speech events. Participants continually negotiate the participation framework of their
encounter as talk proceeds, and this idea has been fruitfully taken up by conversation
analysts (Goodwin & Goodwin 2004, and Sidnell, this volume).
This short overview of different organizational layers involved in the practical
accomplishment of interaction is by no means meant to be exhaustive, but only serves
to orient and prepare the reader for the overview of some recent developments that is
to follow.
2. Context, ethnography and categorization
Attempts to identify the various layers, units, and events that are pertinent to the anal-
ysis of face-to-face interaction inevitably lead to another fundamental debate: that of
4 Sigurd D’hondt
the relevant dimensions of context with which these various organizational layers and
units are associated. As far as the context issue is concerned, there only exists agree-
ment on the various interpretations of the concept that are to be rejected as inadequate.
Researchers of different stripes and persuasions agree that it is not some static entity
that determines interaction ‘from the outside’ (and must hence be specified before-
hand) but a dynamic and fluid phenomenon that is reflexively indexed in the details
of the talk, as participants negotiate the shape and course of the encounter in which
they are engaged (Duranti & Goodwin 1992; Auer & Di Luzio 1992; Tracy 1998; Auer
2009). It can, therefore, only sensibly be understood from ‘within’ the interaction. This
consensus quickly fades, however, the moment it comes to specifying in positive terms
what context amounts to. The main issue here is the question of how far one is exactly
allowed to depart from the details of the text when describing contextual dimensions
relevant to understanding an interactional phenomenon. Everyone agrees that context
must be specified ‘from within’ the exchange, but it is not always clear how far beyond
the exchange this contextualization of utterances should extend (which is perspicu-
ously summarized in the subtitle to Silverstein’s 1992 paper on the indeterminacy of
contextualization: “when is enough enough?”).
Usually the issue is framed as a practical epistemological question: how much
ethnographic description of the setting (or institution) is the analyst required (or better,
allowed) to utilize while making sense of naturally occurring interactions, in addition
to the mechanical records (audio, video) and transcripts documenting them?
CA furnished the most radically minimalist response to this query by embracing
an exclusively talk-internal, radically endogenous conception of context that conse-
quently became the primary point of reference for the debate that ensued. For CA, con-
text is up for grabs in what is made publicly available in the trajectory of the talk itself.
It coincides with the continually updated ‘architecture of intersubjectivity’ (Heritage
1984a: 254) that materializes in talk as each turn is superseded by a new one. Both con-
text and intersubjectivity, according to this account, are essentially public phenomena:
participants lay out their conduct in an orderly fashion so as to be understandable to
the other participants, and in doing so they disclose to one another how they under-
stand the activities they are engaged in and the contributions of the other partici-
pants. Context, it follows, is thus continually updated as interaction progresses, and as
a consequence context and interaction mutually constitute one another. In producing
new talk, participants orient to the framework of relevance that was established in the
preceding utterances. Their coparticipants will also interpret this new talk in relation-
ship to these previous utterances. On the other hand, this new talk, in turn, projects
further talk and hence it becomes part of the framework to which other participants
will orient as they add new talk to the conversation. Thus interactional conduct is, in
Heritage’s words (1984a: 242), at once ‘context-shaped’ and ‘context-renewing.’ This
radically endogenous conception of context, as the publicly available product of the
interactional activities in which the participants are engaged, goes hand in hand with
an equally radical rejection of all other sources of information that might shed light
on what is going on. Thus, vernacular descriptions of interaction (like participants’
self-characterizations of their conduct in narrative accounts or interviews) and prema-
ture ethnographic specifications of context in terms of social structural categories are
unacceptable because they entail the risk of ‘losing data’ (Maynard 2003) by glossing
over the orderliness and patterning of talk that is the ultimate repository of interac-
tional meaning (Schegloff 1991, 1992c).
The precise extent and nature of this public intersubjectivity is further curtailed
in actual research practice. A number of key publications (including seminal papers
like Schegloff (1968), Schegloff and Sacks (1973), and Sacks et al. (1974), influential
collections like Atkinson and Heritage (1984), and Schegloff ’s already mentioned
(2007) primer), seem to suggest that the one aspect of interactional orderliness that
is most central to reaching a common understanding of what exchanges amount to is
their sequential order, that is, the orderliness located in “the describable ways in which
turns are linked together into definite action sequences” (Hutchby & Wooffit 1998: 38).
One of the most eloquent critics of the ‘aridity’ of such CA accounts that explicate
interaction solely in terms of the “dry bones” (1988: xi) of talk is the ethnographer
Moerman (1988, 1990/1, 1993), who came to adopt CA while investigating everyday
interactions in a Thai village. For Moerman, “[a]ll actions are socially situated and all
situations are socially structured. Sequential analysis delineates the structure of social
interaction and thus provides the loci of actions. Ethnography can provide the mean-
ings and material conditions of the scenes in which the actions occur” (1988: 57).
CA can therefore be of great value to ethnographers. Since the latter depend on talk
for understanding how other people interpret their environment, they must come to
terms with “how the organization of talk influences what people say” (Moerman 1988: 9).
Analysts of talk-in-interaction in turn face the task of acquiring a grasp of the cultural
meanings and of the lay actors’ intentions and the projects they are engaged in, if they
want to appreciate fully how the latter make sense of subsequent turns. This poses few
problems as long as the analyst is content with analyzing records of everyday activities
collected in a society to which s/he is a native. This changes, however, when analysts
find themselves confronted with interactional materials documenting activities col-
lected in institutional environments or ‘exotic’ cultural settings with which they did
not previously acquire familiarity as a lay participant. Here, participant observation
and ethnographic fieldwork must come to the rescue. A similar critique of CA prac-
titioners’ naïve reliance on unexplicated forms of ‘membership knowledge’ underlies
much of the work of Aaron Cicourel. To appreciate the complexity of the tasks that are
carried out in medical settings and to comprehend how their execution is informed by
bureaucratic and interactional constraints, Cicourel (1992) argues that transcripts of
clinical interaction must be complemented with (1) ethnographic descriptions of the
6 Sigurd D’hondt
organizational layout of the clinic, of its working arrangements and the daily social
interactions that take place there, and with (2) the various textual materials (patients’
charts, medical records, students’ textbooks) that illuminate how knowledge is codified
(which is consistent with Cicourel’s long-standing interest in cognitive processing
and individuals’ cognitive coping mechanisms, cf. Cicourel 1974). Also, according to
Cicourel (1992), CA’s endogenous view of context requires analysts to manipulate their
data and prepare their transcripts in particular ways (cf. Bucholtz 2000; Blommaert 2005:
Chapter 3). It should also be noted, and this partially absolves CA, that Moerman and
Cicourel’s criticisms seem, to a certain extent, informed by different research agendas
(ten Have 2007: 75) because they are not so much interested in the technology of con-
versation as in explicating the ‘substance’ of particular scenes and settings (in the case
of Moerman, the texture of ‘human events’ (1988: 13), and for Cicourel, the nature of
institutional tasks).
One effect of this critique is that conversation analysts became more explicit
about their liaisons with ethnography. Maynard (2003), without yielding to the posi-
tion of Moerman and Cicourel, nonetheless concedes that ethnography can be of
‘limited affinity’ to conversation analysis, but only after one has first elucidated the
self-explicating patterning of verbal activities. More specifically, ethnographical tech-
niques are useful
to (1) refer to settings and participants according to institutional or other identities and
categories [in order to avoid having to demonstrate their relevance on each occasion
again and focus on focal activities instead], (2) describe courses of actions related
to a focal episode and familiar terms within it [a form of ‘cultural contextualization’
comparable to what Moerman called for, but in a strict data-controlled fashion], and
(3) explain curious sequential patterns. (2003: 77)
Approaching the issue from a slightly different angle, conversation analysts working
with video records of collaborative work in complex technological environments (cf.
infra) mention the necessity of carrying out fieldwork prior to data collection. Thus,
according to Heath (2004: 273), “[i]t is not unusual […] to delay gathering recorded
materials until researchers have a passing understanding of the activities in question
and the various tools and technologies which feature in the accomplishment of even the
more mundane activities in such settings.” In a similar vein, Mondada (2006a: 7) points
out that “fieldwork plays an essential role for the identification of expectable patterns
of action to which to adjust the video shot” (Mondada 2006a: 7). CA in turn represents
a useful tool for substantiating ethnographic findings as it enables one to explicate
in detail the interactionally negotiated nature of the social world or the exact ways in
which material artifacts are shaping encounters (Maynard 2003: 77; Heath & Hindmarsh
2002). On the whole, the idea that ethnography constitutes a kind of ‘supplement’ to
sequential analysis (criticized, e.g., in Blommaert 2004) is gradually being replaced by
a more refined conception of the relationship between the two and by a recognition
that CA constitutes in itself a specific form of fieldwork practice. According to Heath
and Hindmarsh (2002), it is frequently the case that initial analytic observations and
insights in turn trigger more focused observations and new video recordings, and, in
this sense, current workplace studies are characterized by the iterative cycle also found
in other field studies. Maynard (2003: 78) raises the possibility of developing a ‘longi-
tudinal’ form of CA that seeks to explicate extended social activities distributed across
time and space instead of concentrating on singular episodes (consider Wootton 1997
for an intriguing example of what this might look like). Furthermore, the practice of
data collection is itself transformed into a topic for analysis in its own right: as much
as videos are produced to preserve the relevant details and phenomenal fields of
action, they also reflexively contribute to the very interactional order they document
(Mondada 2006b).
Yet other authors responded that Moerman was mistaken to assume that CA is
solely concerned with sequential phenomena (Heritage 1990/1: 302). Presumably ‘men-
tal’ concepts like ‘intention’ (Heritage 1990/1), ‘orientation’ and ‘concern’ (Pomerantz
1990/1) and also cognitive states (Potter 1998), which Moerman considered to be the
object of ethnography, can themselves be analyzed as locally occasioned members’
constructs, and CA provides the requisite analytical tools for such an operation (cf.
Heritage & Raymond 2005 for another example). Proceeding in a similar direction,
Watson (1994, 1997) argued that Moerman’s complaint of ‘aridity’ can be avoided if
CA were to join forces with the forgotten tradition of membership categorization anal-
ysis (Sacks 1972a, 1972b; Jayyusi 1984; Hester & Eglin 1997; Fitzgerald & Housley 2002;
see also Watson, this volume) and would investigate sequential order and the situated
practices by which lay participants categorize one another ‘in a single take.’ Parallel with
this development, and also (at least implicitly) informed by the literature on member-
ship categorization, a flourishing literature emerged that specifically investigates how
‘discursive’ identities (related to the sequential organization of talk) and ‘situational’
identities (related to broader notions of social structure) come to be coarticulated in
talk (Wilson 1991; Zimmerman 1998), in response to Schegloff ’s injunction (1991,
1992c) that the relevance and procedural consequentiality should be demonstrated in the
details of the talk. A similar concern underlies the growing body of work that, in the
wake of Atkinson and Drew’s (1979) ground-breaking study of courtroom interaction,
seeks to identify the ‘unique fingerprint’ (Drew & Heritage 1992a: 26) of institution-
alized forms of interactions by systematically comparing how they differ (through
reductions, respecifications, or specializations) from ordinary, mundane conversation
(Drew & Heritage 1992b; McHoul & Rapley 2001; Arminen 2005; Heritage 2005, see
also Hakulinen this volume). What these different developments have in common is
their awareness of the fact that the sequential order of talk (Hutchby & Wooffit 1998: 38,
cf. supra) reflexively presupposes, and is simultaneously informed by, an inferential
8 Sigurd D’hondt
order (ibid.: 42) consisting of the various normative and inferential properties of talk
that participants draw upon in holding each other morally accountable for what they
are doing or saying.
In the late 1990s, the question of what counts as the relevant context of an exchange
briefly reasserted itself, but the focus of the debate was then no longer the appropriate-
ness of ethnographic ‘supplements’ but rather the ‘analytical relevancies’ the analyst
imposes on the data. Partially foreshadowed by Watson’s (1992) argument that Goffman’s
dramaturgical metaphor and CA’s attempt to describe the self-explicating properties
of talk constitute incompatible language games, the controversy erupted fully after
Schegloff (1997) accused critical discourse analysis of ‘theoretical imperialism.’ By not
grounding their analyses in the participants’ own demonstrable orientations to what
is contextually relevant, he argued, the latter are imposing their own political agenda
on a world that had already been interpreted by the lay actors concerned – an allega-
tion that was the basis of an animated debate with Wetherell and Billig on the pages of
Discourse & Society (Schegloff 1998, 1999a, 1999b; Wetherell 1998; Billig 1999a, 1999b).
The issue resurfaced a few years later in the context of interactionally oriented law and
society studies, when the work of critical scholars like Conley and O’Barr (1990, 1998)
was criticized from within ethnomethodology for its ‘theoretical’ preoccupation with
issues of power (Dingwall 2000; Travers 2006).
3. Multimodality and mediation
Recently, the debate over context seems superseded by a range of new analytical
concerns. One recent development that greatly reinvigorated the field of interac-
tion analysis is the fast growing attention for the multimodal character of interaction
(Norris 2004; Stivers & Sidnell 2005; Mondada 2006) and the fact that face-to-face
encounters are not only composed of ‘vocal-aural’ events and practices (as early CA
investigations of telephone calls somehow seemed to suggest) but also involve sign
systems situated at the ‘visuospatial’ modal plane (Enfield 2005). Pioneering studies
by the likes of Goodwin (1981), Heath (1986) and Kendon (1990) paved the way by
drawing attention to the way gaze and the comportment of the body are instrumen-
tal in aligning speakers, facilitating a common focus, selecting a recipient for one’s
utterance, etc. – in short, for the interactive organization of participation (Goodwin &
Goodwin 2004; Sidnell in this volume). Importantly, the task of this visuospatial
modality is not confined to elaborating the semantic content of talk or ‘providing the
necessary context.’ Rather, the different modalities actively work together in establish-
ing coherent courses of action (Stivers & Sidnell 2005) and in providing solutions for
interactional problems like the shifting of speakership (Mondada 2006a). Talk is no
longer considered the sole or primary component of meaningful action, and restricted
understandings of talk-in-interaction that focus exclusively on the vocal-aural modality

are complemented with a conception of situated conduct that also encompasses the
way the body is exploited as a resource for producing meaningful action (Goodwin
2000). Opening up the notion of action to include embodied conduct also enables us
to conceive how participants are able to engage in multiple actions simultaneously at
different levels of phenomenal awareness (Norris 2004).
Parallel to this first development, researchers increasingly became aware of the
mediating role tools and technologies play in the shaping of interaction. Suchman’s
(1987) pioneering study of how users interact with an intelligent copy machine put
the analysis of collaborative work in complex technological environments high on the
agenda, spurring a great amount of work on various kinds of coordination centers
like subway control rooms (Heath & Luff 2000) and airport ground operation control
rooms (Goodwin 1996; Suchman 1996, 1997), in addition to work on airliner cockpits
(Nevile 2004) and video-mediated interactions in a surgical team (Mondada 2007) (for
an overview see Luff et al. 2000). Mediating technologies also increasingly permeate
the sphere of everyday interaction (e.g., see Hutchby 2001 and Arminen & Leinonen
2006 on the widespread use of mobile phones). Importantly, this growing interest in
mediating technologies is closely connected to the turn to multimodality, and these
two developments may be considered, to a large extent, mutually elaborative. That
is, even if they are situated in a technologically mediated environment, shared work-
spaces are still constituted to a large extent through mutual attention (Suchman 1996),
while embodied actions (like, say, a move in a hopscotch game, cf. Goodwin 2000)
often derive their meaning from the way the body orients itself to meaningful semiotic
structures inscribed in the material environment or contained in an artifact. Paradoxi-
cally, technological mediation also resulted in the emergence of ‘disembodied’ forms
of action (Heath & Luff 2001: 183ff), of which computer-mediated communication
(CMC) and Internet Relay Chat (IRC) (Herring 2001; Hutchby 2001) are the most
radical representatives.
This recognition of the multimodal and mediated nature of interaction also
had a paradoxical effect on the subsequent trajectory of analytical notions of con-
text. Because participants rely on “a range of structurally different kinds of sign phe-
nomena in both the stream of speech and the body, graphic and socially sedimented
structure in the surround, sequential organization, encompassing activity systems,
etc.” (Goodwin 2000: 1490) in producing and understanding meaningful action, CA’s
endogenous conception of context comes to refer to a reflexively indexed, temporally
unfolding configuration of semiotic fields (ibid.), which mutually elaborate one another
and which may comprise multiple signs systems. Social actions cannot be understood
solely with reference to the details of language use or their sequential surround, and
reports of sequential organization must be complemented with detailed accounts of
“the way in which the social, cultural [and] material structures […] of the environment
10 Sigurd D’hondt
where action occurs, figure into its organization” (Goodwin 2000: 1491). In a way, then,
this opening up of action to fully include phenomena situated on other modal planes
further problematizes the dichotomous understandings of text and context up to a
point that the distinction between the two finally seems to collapse altogether. Space,
to name but one contextual parameter for a long time considered unaffected by talk,
is also coming to be treated as ‘configured by’ as much as ‘formative of ’ interactional
exchanges (Mondada 2005, see also the contributions in McIlvenny et al. forthcoming). In
the adjacent field of mediated discourse analysis (Scollon 1998; Norris & Jones 2005),
similar considerations lead to the notion that understanding social action requires
situating it in the dynamically (and reflexively) constituted nexus of the various tra-
jectories of the means and tools that mediate it – situated discourses (‘voices’, ‘ways of
speaking’), interaction orders (formats for organizing togetherness), non-verbal signs,
material objects, physical bodies and their habitus, etc. – each of which is subject to its
own cycle and timescale (Scollon & Scollon 2003, 2004).
The paradoxical effect of this demise of the text-context dichotomy, analytically
speaking, is that an increasingly wider array of phenomena is being subjected to CA’s
policy of ‘reverse engineering’ (i.e., the practice of ‘reasoning back’ from text to context
and from the details of interaction to the practices by which it is constituted, Arminen
2005). To the extent that this move entails a recognition of the role that historically
contingent material and semiotic resources play in sense-making, it provides part
of the answer to the much sought-after ‘contextualization’ of talk as advocated by
Moerman (1988). Sidnell (2008), for example, argues that talk-in-interaction must be
conceived of as “organized around a base of generic principles which are inflected
by a variety of local conditions. Local circumstances, including grammatical patterns,
onomastics, and demographic factors, can be seen to ‘torque’ the generic organiza-
tion in different directions” (2008: 481). In his view, “the generic problems of human
interaction (turn-taking and repair, for instance) are solved through the mobilization
of the eminently local semiotic resources of a particular language in a particular social
setting” (2008: 479). In fact, Sidnell is here sketching the outlines of a position that
elegantly reconciles orthodox CA claims about the universality of the machinery by
which talk is constituted (and, by extension, of the talk-internal, endogenous concep-
tions of context that is the corollary of this autonomy-claim) with a recognition of
historically constituted variability that is a longstanding ethnographic and sociolin-
guistic concern (cf. the comment on Sidnell in Rampton 2008). What is at stake in this
development is also nicely captured in the following quote from Jean Lave, initially
formulated in the context of practice-centered neo-Vygotskian accounts of cognition
(Lave 1988; Goodwin 1994; Hutchins 1995; Engeström & Middleton 1996), in which
she quotes Suchman as stating that “concern with historical mediation has become a
point of convergence between activity theorists and phenomenological analysts: the
latter are beginning to recognize that immediate situations include historical artefacts,
practices, and routines, and that historical artefacts provide resources, interactionally,
to be garnered and employed on next occasions” (Lave 1993: 21).
4. Style and indexicality
One conceivable text-context connection that we have not touched upon yet – probably
because it partially exposes the limits of CA’s policy of reverse engineering – centers
around the fact that speakers can convey meaning by selecting particular variables over
others and the interactional significance of particular communicative styles (Selting,
this volume). Recent work in stylistics (Eckert & Rickford 2001; Auer 2007) strongly
insists on the processual, practice-like nature of style and emphasizes the agency of the
speaker. The idea that linguistic variables passively mirror particular identity catego-
ries or that their social meaning is, in a straightforward sense, located in geographic
locations or class positions is no longer accepted. Rather, variables are assumed to
become socially meaningful when they are creatively reconfigured in the course of
linguistic performance to present oneself as a particular kind of person (Coupland
2001; Eckert 2004) or to perform a stylized version of a recognizable Other (Rampton
1999; Auer 2007). Because these reconfigurations-in-performance involve variables
situated at different levels of linguistic and interactional structuring (and also because
these reconfigurations are, in turn, part of broader semiotic systems that also encom-
pass particular types of clothing, musical tastes, etc., cf. Eckert 2004), style is ineluctably
also a “holistic and multilevel phenomenon” (Auer 2007: 11). Furthermore, these
performances cannot be understood solely within the context of the immediate
speech situation. To the extent that ‘doing’ a particular style is embedded in a sys-
tem of distinctions (Irvine 2001), and the notion of style mediates between socially
situated performance and heterogeneity within linguistic systems (or, in the case of
code-switching, across systems) (Auer 2007), styles minimally presuppose a form of
paradigmatic contrast that can only be understood in the context of the wider systems
of distinctions and arrangements of resources that foster them. As such, the meanings
of a particular stylistic choice are always indexical (in the Peircean sense, cf. supra) and
based on complex associations between linguistic signals and their contexts.
The attention accorded to an indexical phenomenon like style exemplifies the dif-
ference in analytical approach between interactional sociolinguistics and conversation
analysis. As indicated earlier, the former shares with CA an overarching concern with
situated meaning and with the practical reasoning that shapes interaction. Its treat-
ment of identity as a symbol produced in talk and not as an externally constraining
variable (Gumperz & Cook-Gumperz 1982, see Tannen 1981 for a classic analytical
demonstration) is furthermore consistent with the CA position that membership of
an identity category constitutes a practical achievement (Antaki & Widdicombe 1998).
12 Sigurd D’hondt
What sets these two approaches apart, however, is the way in which they conceptual-
ize the nature of conversational inferencing. While CA locates inferences primarily
in explicit identity ascriptions and in the capacity of particular ‘doings’ to invite cor-
responding categorizations of the actor (along the lines of the viewers’ and hearers’
maxims identified in Sacks 1972a, 1972b), interactional sociolinguistics includes in the
inferential process the various indexical meanings of the linguistic material speakers
use for producing their interactional contributions. Opening up the notion of infer-
ence to include indexical properties of talk requires analysts to take into consideration
aspects of context that lie beyond the immediate situation of speech. For one thing, the
indexicality of a particular linguistic feature is very often mediated by metalinguistic
awareness and linguistic ideology (Silverstein 2003): the ‘initial’ indexical values of a
particular variable (say, its association with a particular geographical region) may, in
turn, develop into an indexical ‘of the second order’ (say, it comes to be perceived as an
indication of a particular sociocultural milieu) on the basis of a language ideology that
distinguishes between correct and incorrect speech forms and projects these distinc-
tions onto a perceived class structure (Johnstone et al. 2006).
5. This volume: Interaction as a topic
While systematically investigating the organizational patterning of activity constitutes

a self-contained, autonomous topic of inquiry in its own right, it can also be used as a
method or resource for shedding light on a host of other phenomena in the social life of
human beings and in the fabric of the various societal arrangements they live in (con-
sider Goodwin 1994 for a powerful example). As a consequence, the field of interac-
tion analysis attracts many scholars with strong affiliations to various other intellectual
communities. As Tracy and Haspel recently noted in a review of the institutional and
disciplinary background of Language and Social Interaction (LSI) research, “the com-
munities LSI scholars argue with are quite diverse. Almost always LSI scholars have a
foot, albeit sometimes only a tentative one, in some other world besides the interdisci-
plinary one that comprises LSI research” (Tracy & Haspel 2004: 788; cf. Rampton 2007;
Bucholtz & Hall 2008 for similar statements). Dedicated to revealing the interactional
basis of the phenomena that make up the horizon of their native intellectual commu-
nity (be it anthropology, sociology, psychology, education studies, or another), for these
scholars, “it is mainstream of their discipline that is the chief target of their arguments”
(ibid.). One particularly visible example of such a collective of interaction scholars
occupying a dissident position within an adjacent disciplinary community is Discur-
sive Psychology (Edwards & Potter 1992; Edwards 1997), which concerns itself with an
interactional analysis of psychological phenomena in a way that is strongly indebted
to CA (pace Wetherell 2007, see also the discussion she had with Schegloff, referred
to earlier). In CA, the interactional patterning of participants’ displays of cognitive

states (and of related presumably mental phenomena) and the various tasks which
these displays may accomplish in conversation is a long-standing topic of inquiry (see,
e.g., Pomerantz 1980, 1984b; Heritage 1984b; Heritage & Raymond 2005; Stivers 2005;
Raymond & Heritage 2006). Discursive Psychology continues and extends this line of
work. Rooted in the disciplinary context of psychology (and also addressing cognitive
science, see te Molder & Potter 2005), it complements CA’s analytical respecification
of cognitive and mental phenomena as interactional displays with an explicit focus
on talk about mental phenomena, specifying the ways psychological categories and
constructs are used in everyday talk and the actions they accomplish in context. In
adopting such a distinctly anti-mentalist position, it has proven a potent alternative,
for example, for experimental work in laboratory settings, which often fails to come to
terms with the contextual variability of psychological phenomena.
For this volume, a principled choice was made to include only Handbook entries
that explicitly focus on interaction as the topic of inquiry. Thus, one finds in this volume
contributions that outline a particular tradition (ethnomethodology (Firth), conversa-
tion analysis (Clift, Hutchby and Drew)) or the intellectual contribution of a particular
author (Harvey Sacks (Watson), Erving Goffman (O’Driscoll)) next to others that
outline a particular level of interactional organization (prosody (Couper-Kuhlen),
communicative style (Selting), sequence organization (Sidnell), participation (Sidnell),
grammar (Lindstrom)) or an interactional phenomenon connected to a specific
episode (listener response (Deng Xudong), reported speech (Holt)). An overview
of various transcription systems for spoken discourse is supplied in O’Connel and
Kowal’s chapter.
It is hard to deny the overall influence of CA in the various Handbook articles that
were selected, although many of them also cover contributions from other scholars
and adjacent traditions, such as those of Goffman (see e.g., the chapters on participa-
tion (Sidnell), reported speech (Holt), politeness (Kasper)) and interactional socio-
linguistics (see e.g., the chapters on prosody (Couper-Kuhlen) and communicative
style (Selting)). It should be kept in mind, however, that this bias in favor of CA is not
so much an indication of its intrinsic superiority, but rather expresses the fact that its
account of the machinery of interaction represents a common core in which many will
recognize a point of departure.
As indicated, it was decided not to include forays into substantially defined
domains resulting from interaction analysts’ engagements with other disciplinary com-
munities. Priority was given to accounts of the interactional ‘machinery’ rather than
to work that focuses on the substance of particular settings. Thus, one finds accounts
of the indexical potential of speech styles as a contextualizing phenomenon, but no
surveys of the range of identities that may be indexed by them (e.g., through crossing
(Rampton 1995; Quis & Normann Jörgensen 2007) or stylization (Rampton 1999;
14 Sigurd D’hondt
Auer 2007)) or overviews of specific distributions of stylistic practices in particular

societal arrangements. For these, the reader is referred to other volumes in the Hand-
book of Pragmatics Highlights series, in particular Culture and language use and Society
and language use.
References
Antaki, C. & S. Widdicombe (eds) (1998). Identities in Talk. Sage.

Arminen, I. (2002). Design-oriented sociology. Acta Sociologica 45(4): 315–322.
Arminen, I. (2005). Institutional interaction: Studies of talk at work. Ashgate.
Arminen, I. & M. Leinonen (2006). Mobile phone call openings: Tailoring answers to personalized
summons. Discourse Studies 8(3): 339–368.
Atkinson, J.M. & P. Drew (1979). Order in court: The organisation of verbal interaction in judicial
settings. Macmillan.
Atkinson, J.M. & J. Heritage (eds) (1984). Structures of social action: Studies in conversation analysis.
Cambridge University Press.
Auer, P. (2009). Context and contextualization. In J. Verschueren & J.-O. Ostman (eds), Key Notions
for Pragmatics (Handbook of Pragmatics Highlights, vol. 1): 86–101. John Benjamins.
Auer, P. (ed.) (2007). Style and social identities: Alternative approaches to linguistic heterogeneity.
Mouton de Gruyter.
Auer, P. & A. di Luzio (eds) (1992). The contextualization of language. John Benjamins.
Billig, M. (1999a). Whose terms? Whose ordinariness? Rhetoric and ideology in conversation analy-
sis. Discourse & Society 10: 543–58.
Billig, M. (1999b). Conversation analysis and the claims of naivety. Discourse & Society 10: 572–76.
Blommaert, J. (2004). Workshopping: Professional vision, practices and critique in discourse analysis.
Academia Press.
Blommaert, J. (2005). Discourse: A critical introduction. Cambridge University Press.
Boden, D. & D.H. Zimmerman (eds) (1991). Talk and social structure: Studies in ethnomethodology
and conversation analysis. Polity Press.
Bucholtz, M. (2000). The politics of transcription. Journal of Pragmatics 32(8): 1439–1465.
Bucholtz, M. & K. Hall (2008). All of the above: New coalitions in sociocultural linguistics. Journal
of Sociolinguistics 12(4): 401–431.
Cicourel, A.V. (1974). Cognitive sociology: Language and meaning in social interaction. Free Press.
Cicourel, A.V. (1992). The interpenetration of communicative contexts: Examples from medical
encounters. In A. Duranti & C. Goodwin (eds): 291–310.
Conley, J.M. & W. O’Barr (1990). Rules versus relationships: The ethnography of legal discourse.
University of Chicago Press.
Conley, J.M. & W. O’Barr (1998). Just words: Law, language and power. University of Chicago Press.
Coupland, N. (2000). Language, situation and the relational self: Theorizing dialect-style in socio
linguistics. In P. Eckert & J.R. Rickford (eds), Style and sociolinguistic variation: 185–210.
Dingwall, Robert (2000). Language, law, and power: Ethnomethodology, conversation analysis, and
the politics of law and society studies. Law & Social Inquiry 25(3): 885–911.
Drew, P. & J. Heritage (1992a). Analyzing talk at work: An introduction. In P. Drew & J. Heritage
(eds): 3–55.
Drew, P. & J. Heritage (eds) (1992b). Talk at work: Interaction in institutional settings. Cambridge
University Press.
Duranti, A. & C. Goodwin (eds) (1992). Rethinking context: Language as an interactive phenomenon.
Eckert, P. (2004). The meaning of Style. Texas Linguistic Forum 47: 41–53.
Eckert, P. & J.R. Rickford (eds) (2001). Style and sociolinguistic variation. Cambridge University Press.
Edwards, D. (1997). Discourse and cognition. Sage.
Edwards, D. & J. Potter. (1992). Discursive psychology. Sage.
Eerdmans, S. (2003). A review of John J. Gumperz’ current contributions to interactional socio
linguistics. In S. Eerdmans, C. Prevignano & P.J. Thibault (eds): 85–104.
Eerdmans, S., C. Prevignano & P.J. Thibault (eds) (2003). Language and interaction: Discussions with
John J. Gumperz. John Benjamins.
Enfield, N.J. (2005). The body as a cognitive artifact in kinship representations: Hand gesture dia-
grams by speakers of Lao. Current Anthropology 46(1): 1–26.
Engeström, Y & D. Middleton (eds) (1996). Cognition and communication at work. Cambridge
University Press.
Fitzgerald, R. & W. Housley (2002). Identity, categorization and sequential organization: The sequen-
tial and categorial flow of identity in a radio phone-in. Discourse & Society 13(5): 579–602.
Garfinkel, H. (1967). Studies in ethnomethodology. Prentice-Hall.
Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Harper & Row.
Goffman, E. (1981). Footing. In Forms of talk: 124–159. University of Pennsylvania Press.
Goodwin, C. (1981). Conversational organization: Interaction between speakers and hearers. Academic
Press.
Goodwin, C. (1994). Professional vision. American Anthropologist 96: 606–33.
Goodwin, C. (1996). Transparent vision. In E. Ochs, E.A. Schegloff & S.A. Thompson (eds), Interaction
and Grammar: 370–404. Cambridge University Press.
Goodwin, C. (2000). Action and embodiment within situated human interaction. Journal of Pragmatics
32: 1489–1522.
Goodwin, C. & M.H. Goodwin (1996). Seeing as situated activity: Formulating planes. In Y. Engeström &
D. Middleton (eds): 61–95.
Goodwin, C. & M.H. Goodwin (2004). Participation. In A. Duranti (ed.), A companion to linguistic
anthropology: 222–243. Blackwell.
Gumperz, J.J. (1982a). Discourse Strategies. Cambridge University Press.
Gumperz, J.J. (ed.) (1982b). Language and social identity. Cambridge University Press.
Gumperz, J.J. & J. Cook-Gumperz (1982). Introduction: Language and the communication of
social identity. In J.J. Gumperz (ed.): 1–21.
Gumperz, J.J. & D. Hymes (eds) (1972). Directions in sociolinguistics. Holt, Rinehart & Winston.
Have, P. ten (2007). Doing conversation analysis: A practical guide (second edition). Sage.
Heath, C. (1986). Body movement and speech in medical interaction. Cambridge University Press.
Heath, C. (2004). Analysing face-to-face interaction: Video, the visual and material. In D. Silverman
(ed.), Qualitative research: Theory, method and practice (second edition): 266–282. Sage.
Heath, C. & J. Hindmarsh (2002). Analysing interaction: Video, ethnography and situated conduct.
In T. May (ed.), Qualitative Research in Action: 99–121. Sage.
Heath, C. & P. Luff (2000). Technology in action. Cambridge University Press.
Heritage, J. (1984a). Garfinkel and ethnomethodology. Polity Press.
Heritage, J. (1984b). A change‑of‑state token and aspects of its sequential placement. In J.M. Atkinson &
J. Heritage (eds): 299–345.
16 Sigurd D’hondt
Heritage, J. (1990/1). Intention, meaning and strategy: Observations on constraints on interaction

analysis. Research on Language and Social Interaction 24: 311–32.
Heritage, J. (2005). Conversation analysis and institutional talk. In K.L. Fitch & R.E. Sanders (eds),
Handbook of language and social interaction: 103–147. Lawrence Erlbaum.
Heritage, J. & G. Raymond (2005). The Terms of agreement: Indexing epistemic authority and sub-
ordination in talk-in-interaction. Social Psychology Quarterly 68: 15–38.
Heritage, J. & D. Maynard (eds) (2006). Communication in medical Care: Interaction between primary
care physicians and patients. Cambridge University Press.
Herring, S.C. (2001). Computer-mediated discourse. In D. Schiffrin, D. Tannen & H. Hamilton (eds),
The handbook of discourse analysis: 612–634. Blackwell Publishers.
Hester, S. & P. Eglin (eds) (1997). Culture in action: Studies in membership categorization analysis.
University Press of America.
Hutchby, I. (2001). Conversation and technology: From the telephone to the internet. Polity Press.
Hutchby, I. & R. Wooffitt (1998). Conversation analysis: Principles, practices and applications. Polity
Press.
Hutchins, E. (1995). Cognition in the wild. MIT Press.
Hymes, D. (1966). Two types of linguistic relativity (with examples from Amerindian ethnography).
In W. Bright (ed.), Sociolinguistics: Proceedings of the UCLA sociolinguistics conference, 1964: 114–
157. Mouton.
Hymes, D. (1972). Models for the interaction of language and social life. In J.J. Gumperz & D. Hymes
(eds): 35–71.
Irvine, J. (2001). ‘Style’ as distinctiveness: The culture and ideology of linguistic differentiation. In:
P. Eckert & J. Rickford (eds): 21–43.
Jayyusi, L. (1984). Categorization and the moral order. Routledge & Kegan Paul.
Jefferson, G. (1984). On the organization of laughter in talk about troubles. In J.M. Atkinson &
J. Heritage (eds): 346–369.
Jefferson, G., H. Sacks & E.A. Schegloff (1987). Notes on laughter in pursuit of intimacy. In G. Button &
J.R.E. Lee (eds), Talk and social organization: 152–205. Multilingual Matters.
Johnstone, B., J. Andrus & A. Danielson (2006). Mobility, indexicality, and the enregisterment of
‘Pittsburghese’. Journal of English Linguistics 34(2): 77–104.
Kendon, A. (1990). Conducting interaction: Patterns of behaviour in focussed interaction. Cambridge
University Press.
Lave, J. (1988). Cognition in practice. Cambridge University Press.
Lave, J. (1993). The practice of learning. In J. Lave & S. Chaiklin (eds), Understanding practice:
Perspectives on activity and context: 3–32. Cambridge University Press.
Levinson, S.C. (1983). Pragmatics. Cambridge University Press.
Levinson, S.C. (1992). Activity types and language. In P. Drew & J. Heritage (eds): 66–100.
Luff, P., J. Hindmarsh & C. Heath (eds) (2000). Workplace studies: Recovering work practice and
informing systems design. Cambridge University Press.
Matoesian, G.M. (2001). Law and the language of identity: Discourse in the William Kennedy Smith
rape trial. Oxford University Press.
Maynard, D.W. (2003). Bad news, good news: Conversational order in everyday talk and clinical
settings. University of Chicago Press.
McHoul, A. & M. Rapley (eds) (2001). How to analyse talk in institutional settings: A casebook of
methods. Continuum.
McIlvenny, P., M. Broth & P. Haddington (eds) (in press). Communicating place, space and mobility.
Special issue of Journal of Pragmatics.
Moerman, M. (1988). Talking culture: Ethnography and conversation analysis. University of

Pennsylvania Press.
Moerman, M. (1990/1). Exploring talk and interaction. Research on Language and Social Interaction
24: 173–187.
Moerman, M. (1993). Ariadne’s thread and Indra’s net: Reflections on ethnography, ethnicity,
identity, culture, and interaction. Research on Language and Social Interaction 26(1): 85–98.
Molder, H. te & J. Potter (eds) (2005). Conversation and cognition. Cambridge University Press.
Mondada, L. (2005). Espace, langage, interaction et cognition: une introduction. Intellectica
41/2: 7–23.
Mondada, L. (2006a). Participants’ online analysis and multimodal practices: Projecting the end of
the turn and the closing of the sequence. Discourse Studies 8: 117–129.
Mondada, L. (2006b). Video recording as the reflexive preservation-configuration of phenomenal
features for analysis. In H. Knoblauch, J. Raab, H.-G. Soeffner & B. Schnettler (eds), Video analysis:
51–67. Lang.
Mondada, L. (2007). Operating together through videoconference: members’ procedures for accom-
plishing a common space of action. In S. Hester & D. Francis (eds), Orders of ordinary action:
Respecifying sociological knowledge: 51–67. Ashgate.
Nevile, M. (2004). Beyond the black box: Talk-in-interaction in the airline cockpit. Ashgate.
Norris, S. (2004). Analyzing multimodal interaction: A methodological framework. Routledge.
Norris, S. & R.H. Jones (eds) (2005). Discourse in action: Introducing mediated discourse analysis.
Routledge.
Pomerantz, A. (1980). Telling my side: ‘limited access’ as a fishing device. Sociological Inquiry
50: 186–98.
Pomerantz, A. (1984a). Agreeing and disagreeing with assessments: Some features of preferred/
dispreferred turn shapes. In J.M. Atkinson & J. Heritage (eds): 57–101.
Pomerantz, A. (1984b). Giving a source or basis: The practice in conversation of telling ‘how I know’.
Journal of Pragmatics 8: 607–625.
Pomerantz, A. (1990/1). Mental concepts in the analysis of social action. Research on Language and
Social Interaction 24: 299–310 .
Potter, J. (1998). Cognition as context (whose cognition?). Research on Language and Social Inter
action 31: 29–44.
Quis, P. & J. Normann Jörgensen. (2007). Crossing: Negotiating social boundaries. In P. Auer & Li Wei
(eds), Handbook of multilingualism and multilingual communication: 371–389. Mouton de Gruyter.
Rampton, B. (1995). Crossing: Language and ethnicity among adolescents. Longman.
Rampton, B. (2007). Neo-Hymesian linguistic ethnography in the United Kingdom. Journal of Socio-
linguistics 11(5): 584–607.
Rampton, B. (2008). Disciplinary mixing: Types and cases. Journal of Sociolinguistics 12(4): 525–531.
Rampton, B. (ed.) (1999). Styling the other. Special issue of Journal of Sociolinguistics 3(4).
Raymond, G. & J. Heritage (2006). The epistemics of social relations: Owning grandchildren. Language
in Society 35: 677–705.
Sacks, H. (1972a). An initial investigation of the usability of conversational data for doing sociology.
In D. Sudnow (ed.), Studies in social interaction: 31–74. Free Press.
Sacks, H. (1972b). On the analyzability of stories by children. In J.J. Gumperz & D. Hymes (eds): 325–45.
Sacks, H. (1992). Lectures on conversation (two volumes). Basil Blackwell.
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn
taking for conversation. Language 50: 696–735.
Schegloff, E.A. (1968). Sequencing in conversational openings. American Anthropologist 70: 1075–95.
18 Sigurd D’hondt
Schegloff, E.A. (1980). Preliminaries to preliminaries: ‘Can I ask you a question?’ Sociological Inquiry
50: 104–52.
Schegloff, E.A. (1990). On the organization of sequences as a source of ‘coherence’ in talk-in-interaction.
In B. Dorval (ed.), Conversational organization and its development: 51–77. Ablex.
Schegloff, E.A. (1991). Reflections on talk and social structure. In D. Boden & D.H. Zimmerman
(eds): 44–70.
Schegloff, E.A. (1992a). To Searle on conversation: A note in return. In H. Parrett & J. Verschueren
(eds), (On) Searle on Conversation: 113–28. John Benjamins.
Schegloff, E.A. (1992b). Repair after next turn: The last structurally provided defense of intersubjectivity
in conversation. American Journal of Sociology 98: 1295–345.
Schegloff, E.A. (1992c). On talk and its institutional occasions. In P. Drew & J. Heritage (eds):
101–34.
Schegloff, E.A. (1997). Whose text? Whose context? Discourse & Society 8: 165–87.
Schegloff, E.A. (1998). Reply to Wetherell. Discourse & Society 9: 413–16.
Schegloff, E.A. (1999a). ‘Schegloff ’s texts’ as ‘Billig’s data’: A critical reply. Discourse & Society
10: 558–72.
Schegloff, E.A. (1999b). Naivety vs. sophistication or discipline vs. self-indulgence: A rejoinder to
Billig. Discourse & Society 10: 577–82.
Schegloff, E.A. (2007). Sequence organization in interaction: A primer in conversation analysis.
Schegloff, E.A., G. Jefferson & H. Sacks (1977). The preference for self-correction in the organization
of repair in conversation. Language 53: 361–82.
Schegloff, E.A. & H. Sacks (1973). Opening up closings. Semiotica 8: 289–327.
Scollon, R. (1998). Mediated discourse as social interaction: A study of news discourse. Longman.
Scollon, R. & Scollon S.W. (2003). Discourse in place: Language in the material world. Routledge.
Scollon, R. & Scollon S.W. (2004). Nexus analysis: Discourse and the emerging internet. Routledge.
Sidnell, J. (2008). Alternate and complementary perspectives on language and social life: The organi-
zation of repair in two Caribbean communities. Journal of Sociolinguistics 12(4): 477–503.
Silverstein, M. (1976). Shifters, linguistic categories, and cultural description. In K.H. Basso &
H.A. Selby (eds), Meaning in anthropology: 11–55. University of New Mexico Press.
Silverstein, M. (1992). The indeterminacy of contextualization: When is enough enough? In P. Auer &
A. di Luzio (eds): 55–76.
Silverstein, M. (2003). Indexical order and the dialectics of sociolinguistic life. Language and
Communication 23(3/4): 193–230.
Stivers, T. (2005). Modified repeats: One method for asserting primary rights from second position.
Research on Language & Social Interaction 38: 131–58.
Stivers, T. (2006). The interactional process of reaching a treatment decision in acute medical encoun-
ters. Cambridge University Press.
Stivers, T. & J. Sidnell (2005). Introduction: Multimodal interaction. Semiotica 156: 1–20.
Suchman, L. (1987). Plans and situated action: The problem of human-machine communication.
Suchman, L. (1996). Constituting shared workspaces. In Y. Engeström & D. Middleton (eds): 35–60.
Suchman, L. (1997). Centers of communication: A case and some themes. In L.B. Resnick, R. Säljö,
C. Pontecorvo & B. Burge (eds), Discourse, tools and reasoning: Essays in situated cognition:
41–62. Springer.
Tannen, D. (1981). New York Jewish conversational style. International Journal of the Sociology of
Language 30: 133–139.
Tracy, K. (ed.) (1998). Analyzing context. Special issue of Research on Language and Social Interaction
31(1).
Tracy, K. & K. Haspel (2004). Language and social interaction: Its institutional identity, intellectual
landscape, and discipline-shifting agenda. Journal of Communication 54: 788–816.
Travers, M. (2006). Understanding talk in legal settings: What law and society studies can learn from
a conversation analyst. Law & Social Inquiry 31(2): 447–465.
Travers, M. & J.F. Manzo (eds) (1997). Law in action: Ethnomethodological and conversation analytic
approaches to law. Dartmouth Publishing.
Varenne, H. & R. McDermott (1998). Successful failure: The school America builds. Westview.
Watson, G. & R.M. Seiler (eds) (1992). Text in context: Contributions to ethnomethodology. Sage.
Watson, R. (1992). The understanding of language use in everyday life: Is there a common ground?
In: G. Watson & R.M. Seiler (eds): 1–19.
Watson, R. (1994). Catégories, séquentialité et ordre social: Un nouveau regard sur l’oeuvre de Sacks.
In B. Fradin, L. Quéré & J. Widmer (eds), L’enquête sur les categories (= Raisons Pratiques 5):
151–84. École des Hautes Études en Sciences Sociales.
Watson, R. (1997). Some general reflections on ‘categorization’ and ‘sequence’ in the analysis of
conversation. In S. Hester & P. Eglin (eds): 49–76.
Wetherell, M. (1998). Positioning and interpretative repertoires: Conversation analysis and post-
structuralism in dialogue. Discourse and Society 9(3): 386–412.
Wetherell, M. (2007). A step too far: Discursive psychology, linguistic ethnography and questions of
identity. Journal of Sociolinguistics 11(5): 661–681.
Wilson, T.P. (1991). Social structure and the sequential organization of interaction. In D. Boden &
D.H. Zimmerman (eds): 22–43.
Wootton, A. (1997). Interaction and the development of mind. Cambridge University Press.
Wortham, S. (1998). Linguistic Anthropology of Education. Annual Review of Anthropology
37(1): 37–51.
Zimmerman, D.H. (1998). Identity, context and interaction. In C. Antaki & S. Widdicombe (eds):
87–106.
Communicative style
Margret Selting
Potsdam University
1. Definition, delimitation, basic concepts
‘Style’ refers to the meaningful deployment of language variation in written and

spoken discourse, to the particular way that discourse is formed and structured and
that it is interpreted by recipients as socially and interactionally relevant (Sandig &
Selting 1997).
In contrast to other kinds of language variation, such as dialects, sociolects, registers
etc., styles are not conceived of as uni-directionally determined by extra-linguistic and/
or contextual factors, but as meaningful choices made in order to achieve particular
effects or to suggest particular interpretations. While dialectal or sociolectal language
variation may be perceived as a symptom of the speaker’s biographical background
which is not interactionally relevant, stylistic variation is interpreted as a signal or
symbol to suggest particular meanings and interpretations, such as the definition of
the situation in terms of, say, a particular level of formality. This entails that, whereas
other kinds of language variation are mostly viewed as determined or at least influ-
enced by contextual factors in a uni-directional process, the relation of style and context
has been conceived of as a bi-directional process: “Context can influence choice of
style, but also choice of style can influence context” (Semino & Culpeper 1995: 518,
with particular reference to the work of Enkvist).
Stylistics has mostly dealt with the style(s) of written texts. The present entry will con-
centrate on communicative style(s) of spoken language in natural verbal interaction.
The conception of ‘style(s)’ has largely been dependent upon the approach adopted
(see below). Modern approaches prefer dynamic, processual views of style in which
style is looked upon as a production device or strategy which is both formed by and
adapted to the exigencies of the local situation for which it is produced.
Styles in spoken natural language have been shown as having the following properties:
Rather than simply applying pre-fabricated styles, speakers actively construct com-
municative styles as dynamic, flexible and alter(n)able linguistic structures, i.e. styles
of speaking and styles of conversation. These styles are constructed and interpreted
as contextualization cues in relation to contextual stylistic norms and expectations
as well as to fit the recipient design for the particular recipient(s) and their reactions
in the context of social interaction. Styles function as holistic structures which can
be reconstructed as being constituted by bundles of co-occurring style-constitutive
features. The goal of stylistic analysis is to decompose/deconstruct holistic styles and

reconstruct their inventory of more or less (proto)typical style features and their rules
of co-occurrence and alter(n)ation (Ervin-Tripp 1972). In relation to the activities
that are stylistically structured, ‘style’ refers to the particular way in which an activ-
ity is carried out. In relation to an activity type or genre that can be kept constant as
a tertium comparationis, meaningful different ways of constituting this activity type
or genre can be described as different communicative styles. Styles suggest additional
social or interactional meanings which often have to do with self-presentation, definition
of the situation, definition of the relationship between speaker and recipient, framing
of activities and situations, etc.
2. A few landmark reference works
The approaches that have been devised to describe communicative style(s) of spoken
language in natural verbal interaction naturally build on earlier conceptions of style in
linguistics. Most important landmarks are the following conceptions which developed
in orientation to development in general linguistics.
Traditional work in stylistics largely followed Jakobson’s (1960) giving birth to
modern stylistics by introducing structuralist methods for the analysis of literary texts
(for the further development of literary stylistics see Semino & Culpeper 1995). After
stylistics opened its subject matter and perspective by incorporating non-literary
texts, the notion of context began to play an important role. Enkvist (1973), dealing
mostly with style-markers of written texts, developed an approach that he later himself
summarized as follows: “I once suggested that styles come into being as aggregates of
probabilities of expressions in situational contexts, or, more briefly, as aggregates of
contextual probabilities” (Enkvist 1988: 129). Styles are thus conceived of as varieties
of language that correlate with particular contexts. In a somewhat similar conception,
Crystal & Davy (1969: 11) follow the hypothesis “that any use of language displays cer-
tain linguistic features which allow it to be identified with one or more extra-linguistic
contexts”. They set out to describe the stylistically significant or distinctive features
of varieties such as ‘the language of conversation, unscripted commentary, religion,
newspaper reporting and legal documents’. Differences between written and spoken
language are captured as structural differences that correlate with discourse medium
and situation or context of language use (cf. also Chafe 1994 for a more recent account).
This kind of structural description culminates in Crystal’s (1991) more recent proposal
of ‘stylistic profiling’, a method for analysing and classifying written styles in order to
overcome and remedy the criticized diversity and idiosyncracies of stylistic analyses.
In sociolinguistics, Labov (1972) devised a quantitative approach to the study of style,
using correlational methodology for the description of contextual styles as determined
22 Margret Selting
by contextual features of the situation and the degree of attention paid to speech. So,
in the speech of New York city speakers, the frequency of particular realizations of
the phonological variables ‘r’, ‘eh’, ‘oh’, ‘th’ and ‘dh’ was found to correlate with the fol-
lowing styles: careful speech in the interview situation, reading styles for continuous
passages of text, reading word lists, minimal pairs. This line of thinking was further
developed by Bell’s (1984) model of style as audience design. All these approaches
are based on the unidirectional model of the relationship between context and style,
treating extralinguistic factors as independent and linguistic variation as dependent
variables. The distribution of selected linguistic variables is calculated over selected
passages of interviews or other texts. Analysis mostly aims at a typology of styles from
an etic perspective. The turn-by-turn use and alter(n)ation of speech styles in talk as
an interactional resource in general slips through the analytic net.
R. Lakoff (1979) extends the model of transformational grammar for a conception
of styles: “when I speak of a grammar of style, I mean that we are to transfer the con-
cepts devised for linguistic theory — rules, co-occurrence constraints, ungrammati-
cality, and so forth — to the description of other forms of human behavior, a system
that if adequate will not only categorize what is actually extant, but will also indicate
what does not exist, in normal persons on the one hand, and in general on the other”
(Lakoff 1979: 59f.). She postulates that personal style can be distinguished with refer-
ence to four basic determinants: Clarity, Distance, Deference, Camaraderie (ibid.: 62).
Applying this to male and female speech, Lakoff suggests that male and female ideal
styles in the American culture differ with respect to their surface styles: “American
women’s traditional target is Deference as men’s is Distance/Clarity” (ibid.: 71). As
both genders produce and interpret the behavior of the other gender according to
their own styles, these stylistic differences may result in misunderstandings in cross-
gender-communication.
Textlinguistic and pragmatic approaches led the way to a modern conception of
style. Enkvist (1988: 146) outlines a dynamic, processual view of style, in which “styles
arise through a generative process steered by a set of heterarchic parameters” and in
which “(e)ach parameter is set at a value and it carries a certain weight in relation to
other parameters”. In spite of this dynamic conception of style, in which styles are
viewed as processes of text production, style is nevertheless viewed as determined by
parameter values and weightings which contrast with other values and weightings
(ibid.: 149). The proposed methodology relies “on the researcher’s stylistic competence
and on his intuitions” (ibid.: 150).
Pragmatic approaches in stylistics describe both the recurrent constitution of par-
ticular speech acts rather than others, and the particular features of the performance of
speech acts, for instance the particular wording of requests and their effects. Work in
this framework was presented by, for instance, Sandig (1978, 1986) and Hickey (1989).
Sandig (1986) developed a comprehensive conception of stylistic structures and functions.
Hickey (1989: 8) formulates the goal of his approach, which he calls ‘pragmastylistics’ as
follows: “Pragmastylistics will always attempt to show how the different possible ways
of saying ‘the same thing’ (style) depend on factors which compose the situation (prag-
matic factors)”. Most work in this approach, too, relies on intuitive analyses.
In sociolinguistics, Hymes (1974) built on Ervin-Tripp’s (1972) notions of rules of
co-occurrence and alternation and suggested the analysis of ‘styles of speaking’ as an
important goal of an ethnography of speaking. Within a speech community, “(p)ersons
are recognized to choose among styles themselves, and the choices to have social
meaning” (Hymes 1974: 434f.) Hymes in general distinguishes between etic and emic
accounts: ‘etic’ denotes a description and classification ‘from without’, without necessary
recurrence to the participants’ own notions and categorizations; ‘etic’ denotes a descrip-
tion ‘from within’, reconstructing participants’ own notions and categorizations. The
goal of ethnographic description is to show which etic components and functions of
speech are emically relevant in a particular speech community or speech event.
More recently, Gumperz (1982) suggested an interpretive-sociolinguistic concep-
tion of styles of speaking as contextualization cues. “(A) contextualization cue is any
feature of linguistic form that contributes to the signalling of contextual presupposi-
tions” (ibid.: 131). For the most part contextualization cues “are habitually used and
perceived but rarely consciously noted and almost never talked about directly” (ibid.);
their meaning is implicit. As contextualization cues, Gumperz describes code, dialect
and style switching processes, prosodic phenomena as well as choices among lexical and
syntactic options, formulaic expressions, conversational openings, closings and sequenc-
ing strategies (ibid.). In this approach, the notion of context is no longer conceived of
as an aggregate of material properties and factors, but as an interpretive construct, an
interpretive frame. Contextualization cues are thus cues “that speakers use to suggest the
interpretation of what is said within and in relation to particular interpretive frames for
dimensions such as politeness, degree of intimacy, formality, institutionalization, etc.”.
(Sandig & Selting 1997: 145; cf. also Gumperz 1992; Auer & di Luzio 1992).
An interactional approach to the analysis of communicative style(s), which com-
bines the analysis of style(s) as contextualization cues with conversation analytical
methodology, will be exemplified in Section 4 of this paper.
Within communicative styles, we can distinguish between styles of speaking and
styles of conversation. ‘Styles of speaking’ refers to the participants’ interactionally rel-
evant and meaningful ways of using co-occurrent cues from lexico-semantics, syntax,
morpho-phonology, prosody and rhetorics in their talk in interaction. In addition to
this, ‘styles of conversation’ also comprise the participants’ interactionally relevant
ways of organizing the routine tasks of (conversational) interaction in situational
contexts. This involves the organization of turn-taking, the constitution of activity
types and genres, the management of topical talk and interactional modality (Sandig &
Selting 1997: 5).
24 Margret Selting
In my sample analysis in Section 4, I will illustrate styles of speaking. A good

example of the analysis of styles of conversation is Tannen’s (1984) analysis of a thanks-
giving dinner at which participants met who brought along different conversational
styles. Tannen showed that participants with an American east coast Jewish background
and those with a west coast Californian (and one British) background used different
(sub-cultural) styles. These differed most saliently in terms of pacing, grossness of
humour, and story-telling (1984: 149). The east-coast Jewish raised participants per-
formed and preferred a “high-involvement style” with, e.g. machine-gun questions,
gross humour, lively story-telling, and personal statements, to get into contact with
their co-participants. The other participants, however, performed and preferred a
“high considerateness style” which avoids imposing on co-participants in any way.
While talking with a high-considerateness style co-participant, a high-involvement
style speaker lacked enthusiastic responses and therefore lost interest. At the same
time the high-considerateness style speaker had the feeling of being treated impolitely
and put down by their more active co-participants and so reacted with retreat. The
relevance of these styles was testified by showing that the participants encountered
difficulties in getting on with each other on easy terms in their cross-subcultural dis-
course. Nevertheless, style is not a matter of polar distinctions, and Tannen summarizes:
“each person used a unique mix of conversational devices that constituted individual
style. When their devices matched, communication between or among them was
smooth. When they differed, communication showed signs of disruption or outright
misunderstanding” (1984: 147).
All kinds of linguistic and non-linguistic cues can be used for stylistic purposes,
i.e. as a resource to signal interactional meaning. Work has been done on genre and
activity-type specific styles (cf. Selting 1992, 1994, 1997), regional and social styles
(cf. Dittmar; Schlobinski & Wachs 1988; Kallmeyer 1994; Hemphill 1989), gender
styles (e.g. Tannen 1990; Goodwin 1990), and cultural styles (e.g. Tannen 1980; Pavlidou
1994). Recently, a collection of papers edited by Selting & Sandig (1997) presented
work on speech styles as ways of contextualizing activities in verbal interaction, gender
styles, social styles, professional styles and cultural styles in German.
Bourdieu’s (1979) notion of ‘habitus’ has been adopted in order to model the fact
that our styles of speaking and conversation co-occur with other aspects of the life-
style of members of speech communities and other social entities: their housing, ways
of living, clothing, taste etc. Styles are socially evaluated, they are used and interpreted
according to the symbolic value attached to them in social and economic life.
Recently, analyses of style have been inspired by Bakhtin’s (1979) or Vološinov’s
(1929/1975) aesthetic semiotic theory, in which the utterance is taken as the starting
point of a dialogical conception of language and style, and in which the notion of
polyvocality plays an important role (cf. Günthner 1997; Kotthoff 1997). Many of the
insights which Bakhtin/Vološinov have originally developed with reference to literary
texts, mainly novels, have proved to be valuable also for students of spoken language
in natural interaction (cf. Bakhtin 1986; cf. also Linell’s 1998 exposition of a general
dialogical perspective for the analysis of talk in interaction).
3. Problems
As shown above, stylistics has incorporated ideas from other linguistic fields and has
seen the development of several distinct approaches: traditional, mostly structuralist,
stylistics was followed by pragmatic stylistics, text linguistic stylistics, sociolinguistic
stylistics, pragmatic stylistics and interactional stylistics (for more details see Sandig &
Selting 1997: 143ff.). Most often, stylistic analyses which orient to general linguistic
approaches share the theoretical and methodological merits and shortcomings of these
more general approaches. Some general issues are the following.
1. Stylistics has always been a field of study that many mainstream linguists have
been very critical of. Crystal (1991: 223) voiced the following criticism of stylistics:
“It is indeed the case that the goal of stylistics is the explication of linguistic
distinctiveness, but this leaves open the question of how this goal might be
achieved. […] we have too many procedures. Each stylistic article develops its
own approach, which is often as idiosyncratic as the characteristics of the style
it investigates.” He suggests his structural profiling procedure as a remedy in
order to arrive at comparable descriptions.
2. After stylistics has broadened its scope to incorporate the analysis of styles of spo-
ken language and discourse, ‘style’ has been criticised to have swallowed what
formerly was described as ‘text’. The concept has been criticised for having become
too broad, the boundaries of texts or activities and styles seem to have become
blurred. This attests the need to show why the concept of style and stylistic analy-
sis is a relevant one in linguistics, text and discourse analysis. In my opinion, the
concept of ‘style’ is relevant in order to describe the different ways of constructing
activity types and/or genres in interaction (see also below).
3. Should we adopt etic or emic standpoints for the description of styles? Do we wish
to adopt an etic perspective and thus follow largely Crystal’s (1991) plea for struc-
tural profiling ‘from without’ — or do we rather wish to reconstruct member’s own
categorizations and methods of constructing and interpreting styles as resources
for the construction and negotiation of meanings in local situated interactions
(see below)?
4. What relations between context and style should we adopt: style as unidirectionally
determined by contextual factors or style as a contextualization cue that suggests
contexts as interpretive frames?
26 Margret Selting
5. What methods should we use: On the one hand, many students of style use empirical
research methods and apply them to corpora of written and/or spoken discourse;
on the other hand, other students of style rely on intuitive analyses, often on the
basis of their own stylistic competence.
6. What kind of and how many categories do we need to identify in order to describe
styles? In general, students of style agree that style is a property of texts or discourses.
Style is a holistic structure that is constituted by bundles of features on several levels
simultaneously. How many such features, style-markers, do we need in order to
identify a distinct style or a style shift? Can single prototypical style-markers be
used on their own and still suggest a holistic style or a style shift?
7. How can we validate stylistic analyses? Stylistic meanings and functions are inter-
pretations. How can we be certain that our interpretations are not just our own
idiosyncratic ones? Most stylistic analyses lack evidence that their categories and
interpretations are really relevant to the participants in the discourse. Shouldn’t
we devise methods to validate our interpretations and analyses of styles? Validation
methods could involve, for instance, perception experiments and/or methods of
warranting discourse analytical interpretations, such as those shown in Section 4
of this paper.
4. Sample data and methodology of an interactional stylistic analysis
The following analysis focusses on styles of speaking in conversational interaction. On

the background of interactionist thinking, conversation and contextualization analysis
(Sacks, Schegloff & Jefferson 1974 and Gumperz 1982) are combined with stylistic
analysis for the study of language variation as an interactionally meaningful resource.
(This analysis is an abbreviated version of the analysis presented in Selting 1997.)
The constitution of style in spoken language in conversational interaction is
conceived of as a holistic contextualization device. Style is taken to be constructed by
the use of bundles of co-occurring linguistic cues from, e.g. lexico-semantics, syntax
and prosody. These make utterances interpretable within their conversational context
and constitute particular kinds of participant relations for the collaborative construction
of activity types in conversation.
Methodologically, I will follow the following steps:
1. Identification of styles as relevant resources in interaction. Start by (intuitively) identi-

fying holistically interpreted styles which seem to be relevant for the constitution
and recognition of activity types or the constitution of particular meanings for
given activity types in conversation.
2. Structural analysis: Decomposition or deconstruction of styles. Substantiate the intu-
itive interpretation of holistic styles by showing how style-constitutive features on
all linguistic levels, i.e. phonetic-phonological, prosodic, morphological, syntactic,

lexico-semantic, rhetoric, pragmatic, and strategies on the level of conversational
organization such as turn-taking and repair are used to constitute the style(s) and
their alter(n)ations.
3. Functional analysis. Reconstruct the social and interactive meaning, the interpretive
frames and inferences that are suggested by the use of styles in interaction.
4. Warranting the analysis and the interactional relevance of the category ‘style’. Use
details of the speakers’ use and the recipients’ reactions to the speakers’ use of
style, and in particular to alter(n)ations of style like style shifting and switching,
for the warranting of the analysis of style and for the demonstration that participants
indeed do orient to styles. Various methods of warranting the analysis will be
detailed below.
The example
The following transcript presents an extract from an informal conversation. The par-
ticipants are female. After talking about their bad consciences because of smoking,
Mia, a young university teacher willing to stop smoking, tells her recipient that a male
colleague, Friedrich, suggested she stop smoking on the occasion of her promotion.
(1) K0: 484–507 (Laufnr. 189ff.)1
1 Mia: =Friedrich sachte übrigens zu mir dann (.) ähm
Friedrich, by the way, then said to me uhm
2 ja m: dann mach das doch so=
well why don’t you do it like this
3 =ab erstn sechstn rauchs du nich mehr
the first of june you stop smoking
1. Transcription symbols
Segmental and sequential phenomena:

aber DA kam primary accented syllable(s) of a unit
aber dA kam secondary accented syllable(s) of a unit
si:cher lengthening of a sound
(.) brief pause of up to ca 0.5 secs
((lacht)) para- and/or non-linguistic events
ich [ gehe simultaneous talk, overlapping utterances
[ jaha
Pitch at the end of utterances:
. falling to low
; falling to mid
– level
, rising to mid
? rising to high
28 Margret Selting
4 Dor: JA
ya
5 Mia: neue stelle [ohne rauchen
new job without smoking
6 Dor: [ja
ya
7 Mia: (.) m: dat dat einzige was mir dazu eingefalln is
m: the the only thing that I could think of
8 is zu sagn (.)
was to say
9 dann hab ich angst daß ich meine seminare nich mehr schaff
then I am afraid that I won’t manage my teaching
10 Dor: ((gives a [snort and quietly laughs for ca 3 secs.))
11 Mia: (.) [is auch so
it is like that
12 Dor: ou(h) mi(h)i(h)ia
13 (.) ja ich mein: ausredn findse immer
well I mean one can always find an excuse
The extract presents a sequence of story telling. Work in conversation analysis has
shown that story telling is a sequentially organized process and accomplishment,
which is embedded into the local conversational context (e.g. Sacks 1992; Jefferson
1978; Goodwin 1984; for an introduction see also Hutchby & Wooffitt 1998: 131–142).
The basic sequential structure of story telling has been summarized as follows:
story telling can involve a story preface in which a teller projects a forthcoming story,
a next turn in which a coparticipant aligns himself as a story recipient, a next in
which teller produces the story, […] and a next in which the story recipient talks by
reference to the story. Further, the story preface can have consequences for the story’s
reception, and thus a rather extended series of turns at talk can be seen as a coherent
conversational unit. (Jefferson 1978: 219)
Pitch movements on and after accented syllables:

‘ falling
’ rising
– level
↑ little step-up to an accented syllable
↓ little step-down to an accented syllable
Other prosodic parameters which are used with local or global extension, the extension is
indicated by the position of the 〈 〉:
〈loud voice〉, 〈soft voice〉 etc.
4.1 The first intuitive analysis of speech styles in the given sequential context
In the present example, the story is announced as an aside into topical talk about
the participants’ bad consciences because of smoking (see übrigens ‘by the way’ in
line 1). The gap in line 1 can be analysed as leaving room for recipients’ ratification
(or rejection) of the story. After no rejection has come forth, Mia continues telling
her story.
In the telling of her story, Mia changes styles dramatically and thereby suggests
a particular interpretation of the problem which her story illustrates, and thus
makes relevant the very participant reactions Dor gives in lines 4, 6 and 10ff. Both
Friedrich’s proposal and Mia’s reaction are contextualized as different voices to sug-
gest opposite ways of thinking (cf. Bachtin’s (1979) and Vološinov’s (1976) notions
of ‘stylization’ and ‘voices’ in reported speech). Both proposal and reaction are pre-
sented in direct speech: Friedrich’s proposal as the preferred and Mia’s reaction as
the dispreferred way of thinking. The problem which Mia’s story illustrates is not
presented as dealing with a severe problem that the recipient is supposed to react
to seriously and helpfully, but as dealing with a ridiculous nuisance that needs to
be laughed about.
I want to show how, in addition to the sequential implicativeness of story telling
outlined by Jefferson (see above), the linguistic details of Mia’s story telling, her speech
styles constructed to represent the characters of her story, suggest a particular inter-
pretation and thus make the particular response to her story relevant that is provided
by the recipient.
4.2 Structural analysis: Decomposition/deconstruction

For the detailed analysis of spoken styles, we need a more detailed representation
of the extract. The following transcription conventions are most important in the
detailed transcript version (cf. Selting et al. 1998): If transcription lines do not rep-
resent turn-constructional units (Sacks, Schegloff & Jefferson 1974), they represent
rhythmic cadences (Couper-Kuhlen 1993), i.e. accented syllables (beats) plus their
following unaccented syllables up to the beginning of (but not including) the next
accented syllable. Capital letters are used to represent accented syllables. Cadences
which are heard as rhythmically structured, i.e. roughly similar in length of time,
are placed beneath each other in slashes. In order to verify the auditive interpreta-
tion of rhythm, the length of cadences was also measured with a stop watch; the
mean values are given in the right margin of the transcript. Beneath the text, relevant
intonation, loudness, tempo and other cues are indicated. They will be spelled out
in the analysis.
What are the styles that Mia constitutes and how are they constituted? I will
make explicit the various co-occurring style-constitutive cues for each of the two
30 Margret Selting
contrasting parts, dealing successively with cues from (a) lexico-semantics, (b) syntax,
and (c) prosody — here further differentiated on the basis of (ca) rhythm, (cb) into-
nation, (cc) loudness, and (cd) other articulatory features.
The first part of Mia’s story is represented in lines 1 through 13:
1 Mia: =↑ ‘FRIEDrich sachte übrigens zu mir dann. (.) ähm-

Friedrich, by the way, then said to me uhm
2 /↑‘JA; m: dann / /0.5/
〈loud voice
well then
3 /↑ ‘mAch das doch / /0.5/
loud voice
do it like
4 /↑ ‘SO;=ab / /0.3/
loud voice
this from
5 /↑ ‘Erstn / /0.4/
loud and tense voice, precise articulation
first of
6 /‘sEchstn / /0.4/
loud and tense voice, precise articulation
sixth (=june)
7 /‘RAUCHS du nich mehr. /1.2/
loud and tense voice, precise articulation〉
you don’t smoke any more
8 Dor: /‘JA / /0.3/
ya
9 Mia: /–nEue / /0.3/
〈loud, precise articulation
new
10 /↑–STELle– / /0.4/
loud, precise articulation
job
11 Mia: [ /–Ohne / /0.4/
loud, precise articulation
[ without
12 Dor: [ ‘JA;
[ ya
13 Mia: /↓ –RAUchen. / /0.5/
loud, precise articulation 〉
smoking
Friedrich’s proposal is presented as follows:

a. Lexico-semantics: After her very brief initiation of story telling, in which Mia
names the relevant characters in the episode, for the ‘complication’, Friedrich is
quoted in direct speech to have used positively evaluated vocabulary in presenting
a concrete proposal: MACH das doch SO, ab ERStn SECHStn RAUCHS du nich
mehr, and NEUe STELle OHne RAUchen.
b. Syntax: The telling of Friedrich’s proposal culminates, after some introductory
full sentences, in the phrasal units NEUe STELle OHne RAUchen. The full sen-
tences present the actual proposal, the culmination phrases formulate the pointed
positive result, if Mia were to follow this proposal.
c. Prosody: This is further contextualized with prosodic cues. These, however,
differentiate between the first part of the proposal and its culmination.
ca. Rhythm: After the initiation of story telling, which is not organized rhythmically
(line 1), for the presentation of the first part of Friedrich’s proposal, Mia adopts
a saliently rhythmic organisation of talk. Most of her accents in lines 2 through 13
are placed at rhythmic intervals in time, all these rhythmic cadences are com-
paratively short ones: Mia starts with two cadences of 0.5 secs length which are
then followed by a shorter cadence of 0.3 secs. This short cadence is the result
of the presentation of the beat on ERStn (line 5) early in comparison to the pre-
vious beats; this is why in the transcript the slash indicating the beginning of
the cadence starting with ERStn is shifted to the left. All the following rhythmic
cadences are now adjusted and timed with reference to the cadence starting
with ERStn; this is why the sequence of cadences from here on is presented as
beginning in line with the early cadence in line 5. Both the cadences for ERStn
and SECHStn are 0.4 secs short. The next cadence, however, which ends the first
part of Friedrich’s proposal, breaks the rhythm: this cadence is 1.2 secs long and
sounds like a tail to the prior cadences.
cb. Intonation: All the pitch movements in and after accented syllables are falling
ones. But while these normally end in a more or less mid pitch range, the last
pitch movement falls down to low pitch.
cc. Loudness: In contrast to normal loudness in the initiation, Friedrich’s direct
speech is quoted with increased loudness.
cd. Articulation is tense in lines 5–7, precise throughout the whole passage.
4.2.1 Recipient reaction after this first part of the story telling
The recipient of the story telling, Dor, arguably interprets the long cadence with
low falling pitch co-occurring with the possible end of Friedrich’s quoted words
as somehow segmenting the story telling. It is just after this that she gives her
recipient token JA in line 8 which displays rhythmically integrated agreement with
Friedrich’s proposal.
32 Margret Selting
After this, Mia continues her story telling by presenting the culmination of Friedrich’s
proposal once more in an even more pointed way, prosodically organized as follows:
ca. Rhythm: All the cadences are short: they slowly increase from 0.3 over 0.4 to
0.5 secs; all cadences are perceived as rhythmically structured, even if they
slowly become slightly longer.
cb. Intonation: The pitch accent movements in the four words NEUe STELle OHne
RAUchen are all stylized levels. While the first pair is stepping up from mid to
high level pitches, the second pair is stepping down from mid to low level pitches.
All four accents form an entire gestalt prosodically.
cc. Loudness: remains increased throughout.
4.3 Functional analysis

These cues in co-occurrence present Friedrich’s proposal as culminating in the sketching of
Mia’s positive situation if she were to follow his proposal. Within the gestalt that is config-
ured intonationally, the level pitches seem to signal something like the following meaning:
Metaphorically speaking, the positive situation of having a new job and having stopped
smoking is presented as a kind of complete gestalt. The short cadences and increased loud-
ness throughout the proposal present the quoted speaker Friedrich as a ‘positive-thinking’
type of personality who represents the preferred way of thinking about Mia’s problem.
4.4 Warranting
Mia’s presentation of a pointed culmination of Friedrich’s proposal is arguably under-
stood by Dor. She provides a recipient token JA in line 12, displaying her recognition
and collaborative constitution of a first culmination in Mia’s story-telling. This early
collaboration in making the point of Mia’s story seems to make up for the absence of
a(n) (audible) recipiency token after the entire culmination, e.g. in the brief pause in
line 14. This pause is not perceived as an awkward one but as a pause for dramatic
effect, as an increase of suspense.
See how Mia continues her story in lines 14–18:
14 (.) m: dat dat ↓ ‘EINzige was mir dazu ↓ ‘EINgefalln is?

〈decreasing loudness
the the only thing which I could think of
15 is zu sagn
decreasing loudness〉
was to say
16 (.) dann hab ich /0.7/
〈soft voice, breathy, lax articulation
then I am
17 /↑‘ANGST daß ich meine semi / /1.0/

soft voice, breathy, lax articulation
afraid that I won’t
18 /‘NAre nich mehr schaff; / /1.0/
soft voice, breathy, lax articulation 〉
manage my teaching
4.5 Structural analysis: Decomposition/deconstruction

As can be expected, after the brief pause, Mia presents her own reaction. At the very
beginning of this part, in line 14, the constitution of contrast between Friedrich’s
perspective, which she has talked about so far, and the new one, which she is going to
talk about now, is indicated by phonetic cues: in contrast to the standard pronunciation
das with a final [s] for the words das/daß in lines 3 and 17, she produces two items of
marked dialectal dat with final [t].
The second part of her story is then organized as follows:
a. Lexico-semantics: Her own words are presented as the only words she could think
of (dat EINzige was mir EINgefalln is), she talks about her anxiety (ANGST) that
she might not be able to manage her teaching (meine semiNAre nich mehr schaff).
Both the repetition of the syllables EIN in EINzige and EINgefalln and the repetition
of the sound /a/ in the words dann, hab, ANGST, daß, semiNAre, and schaff create
cohesion within the utterance and intensify its meaning. The items present negatively
evaluated matters.
b. Syntax: All the units in lines 14–18 are full sentences.
c. Prosody:
ca. Rhythm: After her first introductory sentence to her own direct speech, which is not
organized rhythmically, Mia again changes to an organization in rhythmic cadences.
Yet, this time, the cadences are notably longer than before, namely 1.0 secs (lines 17
and 18). This is heard as only thinly accented and markedly slower in tempo.
cb. Intonation: While the introductory sentence has rising pitch movements through-
out, the direct speech has falling pitch movements ending within a mid pitch
range, thus again constructing an entire gestalt with prosodic means.
cc. Loudness: After gradually decreasing loudness in the introductory sentence, all the
direct speech is given in decreased loudness.
cd. The entire quotation is produced with breathy and lax articulation.
4.6 Functional analysis

These cues in co-occurrence give the impression of Mia’s slow and played-down
reaction to Friedrich’s proposal. Mia suggests the interpretation of her own reaction
as an ‘unimaginative’, ‘fearful’, ‘negative-thinking’ one which is clearly dispreferred.
Her reaction is contextualized as in direct contrast to Friedrich’s proposal. At the same
34 Margret Selting
time, the contrast between the two ways of thinking is made very ostentatiously. It is
exaggerated and overdone. This exaggeration seems to be used as a cue to suggest and
make recognizable for the recipient that Mia’s modality in telling her story is a self-
mocking and humorous one.
4.7 Warranting
This is exactly what the recipient Dor reacts to in lines 19–22:
19 Dor: ((gives [ a snort and quietly laughs for ca 3 secs.))
20 Mia: (.) [ ↑ ‘IS auch so.
[ it is like that
21 Dor: OU(h) mi(h)i(h)ia.
22 (.) ja ich mein: ↑ ‘AUSredn findse ↑ ‘IMmer;
well I mean one can always find an excuse
Dor does not treat Mia’s problem as a serious problem which ought to be treated with
serious problem talk. Instead, Dor gives a snort and bursts out laughing. Simultaneously,
Mia changes to a different perspective: she confirms that it is like she said in reality, thus
implying that although she made fun of herself, the problem is a real one for her.
The fact that recipients orient to the styles shows that they are relevant to them
in their interaction. In her reaction after Mia’s story telling, Dor orients to both Mia’s
way of presenting her story and her confirmation that she has the problem as a real
one. First, Dor provides the interjection Ou(h) mi(h)i(h)ia in a laughing voice, giving a
more explicit reaction than her mere snorting and laughing, yet still remaining highly
implicit as to her exact evaluation. After a brief pause, Dor then goes on to tell Mia
that one can always find an excuse, openly implying that Mia is not willing enough to
stop smoking. In this, she agrees and aligns with Mia’s negative evaluation of her own
thinking as presented in the telling of her story. At the same time, however, she also
uses some items which save Mia’s face and thus seem to take Mia’s real problem as con-
firmed by IS auch so (line 20) into account: Firstly, the expression ja ich mein qualifies
Dor’s opinion as her own and softens the words projected to come. Secondly, in the
expression findse, which results from the reduction and assimilation of the finite verb
form findest plus the address term du, the form of address is not clearly pronounced; it
is ambiguous and may be a generalizing form of address, the English equivalent being
one rather than you. Similarly, the formulation with the adverb immer also supports
the interpretation of the utterance as a generalization. As a generalizing remark, how-
ever, it is less face threatening for Mia. Arguably, after Mia’s contextualization of her
story in the way outlined above, Dor’s reactions are perfectly expectable. Mia does not
raise any problems with them.
As a result, Dor’s reactions are just the kind of reactions that Mia suggested by
first using and altering her speech style for the presentation of her story in the way
she did, and by afterwards confirming that she has the problem in reality. Both Dor’s
reactions in succession orient to the immediately prior way in which Mia presented
her own problem.
This analysis shows that the co-occurrence of cues from lexico-semantics, syntax
and prosody is used by participants to constitute styles of speaking which are used
as actively deployable, flexibly and dynamically alterable resources in interaction.
Style is conceived of here as a dynamic resource which is altered according to
sequential context and recipient design. Styles are used as contextualization cues to
suggest interpretive frames for the interpretation of talk and make relevant particular
recipient responses. They are deployed to organize the construction of activities as
interactional accomplishments.
The type of language variation that I have described as ‘style’ is neither determined
by the context nor by the activity-types being constituted, nor is it a coincidental choice
by the speaker. It is a meaningful choice in that the speaker chooses a resource that
suggests an interpretive frame and is interpretable by the recipient. Such structures,
styles of speaking, can be adapted to the interaction at hand, they can be switched
and shifted, they can be used across activity types. Therefore, they must be conceived
of as autonomous and independent of the respective activity type. Styles of speaking
are deployed in order to suggest particular kinds of social and interactional mean-
ings such as definitions of the situation, definitions and negotiations of participant
relations, modality, contextualization of activity types and parts thereof in relation to
surrounding activities, etc. The categorization of ‘style’ presupposes that (a) cues that
are analysed as style-constitutive are not normally used coincidentally but recurrently,
and that (b) they are systematically used in order to suggest particular interpretive
frames that recipients orient to.
The notion and analysis of style can be justified by recourse to recipient reactions
and interpretations giving evidence that recipients indeed interpret styles as a mean-
ingful and interactionally relevant contextualization device. Therefore, an important
task of stylistic research is to develop methods of warranting analyses.
Some methods that can be used as search procedures and for warranting stylis-
tic analyses, i.e. methods detailing step (4) of my analysis, are the following (cf. also
Wootton 1989):
(4–1) Substantiation of interpretations with recourse to co-occurring evidence
within a turn or discourse. We can look for lexical items or other explicit formula-
tions co-occurring with more implicit style-constitutive features in order to find
more explicit evidence for our interpretation of implicit stylistic meanings. This was
used when I analysed the wording co-occurring with the prosodic contextualization
of Mia’s story as evidence confirming the meaning of the more implicit syntactic and
prosodic cues.
36 Margret Selting
(4–2) Reconstruction of suggested interpretive frames, presupposed stylistic expec-

tations and restrictions for the activities under investigation. Are there conventionally
expected stylistic realizations or restrictions for the given type of discourse which the
speaker in her/his style constitution conforms to or which s/he presupposes in her
style alter(n)ations or deviations from expected realizations? What interpretive frames
does the speaker suggest via the use of style? I showed how Mia in story telling used
style to present herself and Friedrich as different types of personalities and the prob-
lem at the heart of her story as a ridiculous nuisance.
(4–3) Comparison of juxtaposed styles. In the case of style-alter(n)ation, we can
compare the styles used for the constitution of a given activity with the styles used
(for other activities) previously in the discourse. Thus, the previously used style
functions as an empirical and dynamic tertium comparationis, which the partici-
pants in the interaction orient to in their successive style uses and alter(n)ations.
This procedure was employed when the styles of Mia’s presenting the two parts of
her story were compared.
(4–4) Reconstruction of the interactional organization of style choices and alter(n)ation
in the interaction at hand. In-depth analysis of interaction sequences can show how
participants use style to locally organize the process of interaction, how they negotiate
styles and thus collaboratively constitute the style of an interaction as an interactional
achievement. For instance, style alter(n)ations for culminations in story telling have
been analysed as devices to secure the recipients’ collaborative constitution of the pro-
cess of story telling (Selting 1994).
(4–5) Analysis of the interpretation and treatment of styles, or the activities thus
contextualized, in subsequent interaction. This procedure is intended to demonstrate
that and how recipients in their reaction to a device display that they indeed orient to
it, i.e. that the device is interactionally relevant, and that participants’ orientation to it
corroborates the researcher’s analysis of the device. This was made use of when Dor’s
reactions were shown to orient both to Mia’s style(s) of presenting her story in two
parts and then to Mia’s confirmation that she has the problem as a real one.
(4–6) Analysis of the interpretation and treatment of “deviant cases”. The strongest
evidence we can find for our analyses are “deviant cases”, in which other than expected
styles are used, and in which speakers give explicit explanations and accounts for it, or
recipients complain and react with initiations of repair or the like. In such cases, the
treatment of the deviant case is, on a deeper level, evidence that the participants orient
to the device under investigation as the one to be expected.
Of course, in actual analyses, not all of these and other possible warranting procedures
need to be relevant simultaneously. Furthermore, not all of the procedures can be kept
separate; some interact and overlap with each other. Yet, each of the procedures points
to an aspect that might be relevant.
As an alternative way to analyse this example one might think of, for instance,
Hymes’ (1998) line or verse analysis of spoken narratives. While his textual and pro-
sodic analysis of the utterances constituting oral narratives is partly similar to my
structural analysis, Hymes’ analysis is aimed at exposing the poetic skills with which
members of (in his case native American) speech communities tell their stories. As
his data are not interactional ones — he analyses the telling of traditional myth stories
by narrators in their cultures, he does not deal with any interactional matters. His
approach is much more promising for the analysis of myth stories as told to anthro-
pological or ethnographic field workers by semi-professional narrators than for the
analysis of story telling in everyday conversational interaction.
5. Perspectives for future research
The construction and alter(n)ation of communicative style(s) in everyday conversa-

tional interaction is one of the most subtle ways of suggesting interactional meaning
and thereby of managing (and manipulating) interaction. Stylistic analyses should
therefore be integrated into analyses of social interaction, in particular into analyses of
narratives or story telling in interaction (for an overview see also Gülich & Quasthoff
1986; Ochs 1997), argumentation and disputes, and into the analysis of larger ‘packages’
and ‘communicative genres’ in general (see Bergmann & Luckmann 1995; Günthner &
Knoblauch 1995), wherever relevant.
The management of communicative style in interaction is part of our communicative
competence that needs to be described for theoretical and practical reasons. A lot of
work is still needed which describes the principles of constructing and manipulating
styles in interaction in different activity types and genres and in different languages.
Analyses of communicative style are aimed to reconstruct our stylistic competence,
i.e. our competence to effectively and flexibly use language variation as a situationally
adapted and recipient-designed resource in social interaction.
The analysis of communicative styles can be useful as part and prerequisite of
consciousness raising for a practical and descriptive rhetoric of talk-in-interaction
(cf. Kallmeyer 1996). Such consciousness raising with respect to the effects of stylistic
variation in talk will be useful both for mothertongue and foreign language teaching
as well as for professional rhetorical training.
References
Auer, P. & A. Di Luzio (eds.) (1992). The Contextualization of Language. John Benjamins.
Bachtin, M. (1979). Die Ästhetik des Wortes. Suhrkamp.
38 Margret Selting
——— (1986). Speech genres and other late essays. University of Texas Press.
Bell, A. (1984). Language Style as Audience Design. Language in Society 13: 145–204.
Bergmann, J. & T. Luckmann (1995). Reconstructive genres of everyday communication. In U. Quasthoff
(ed.), Aspects of Oral Communication: 289–304. de Gruyter.
Bourdieu, P. (1979). La Distinction: critique sociale du jugement. Éd. de Minuit.
Chafe, W. (1994). Discourse, Consciousness, and Time. University of Chicago Press.
Couper-Kuhlen, E. (1993). English Speech Rhythm. John Benjamins.
Crystal, D. & D. Davy (1969). Investigating English Style. Longman.
Crystal, D. (1991). Stylistic profiling. In K. Aijmer & B. Altenberg (eds.), English Corpus Linguistics:
221–238. Longman.
Dittmar, N., P. Schlobinski & I. Wachs (1988). Berlin Style and Register. In N. Dittmar & P. Schlobinski
(eds.), The Sociolinguistics of Urban Vernaculars: 44–113. de Gruyter.
Enkvist, N.E. (1973). Linguistic stylistics. Mouton.
——— (1988). Styles as parameters in text strategy. In W. van Peer (ed.), The Taming of the Text:
125–151. Routledge.
Ervin-Tripp, S. (1972). On sociolinguistic rules: alternation and co-occurrence. In J.J. Gumperz &
D. Hymes (eds.), Directions in Sociolinguistics: 213–250. Holt, Rinehart & Winston.
Goodwin, C. (1984). Notes on story structure and the organization of participation. In J.M. Atkinson &
J. Heritage (eds.), Structures of Social Action: 225–246. Cambridge University Press.
Goodwin, M. (1990). He-said-she-said: Talk as Social Organization among Black Children. Indiana
University Press.
Gülich, E. & U. Quasthoff (1986). Story-telling in conversation. Cognitive and interactive aspects.
Poetics 15: 217–241.
Gumperz, J.J. (1982). Discourse Strategies. Cambridge: University Press.
——— (1992). Contextualization and understanding. In A. Duranti & C. Goodwin (eds.), Rethinking
context: 229–252. Cambridge University Press.
Günthner, S. (1997). Stilisierungsverfahren in der Redewiedergabe. In M. Selting & B. Sandig (eds.),
Sprech- und Gesprächsstile: 94–122. de Gruyter.
Günthner, S. & H. Knoblauch (1995). Culturally patterned speaking practices — the analysis of
communicative genres. Pragmatics 5(1): 1–32.
Hemphill, L. (1989). Topic development, syntax, and social class. Discourse Processes 12(3): 267–286.
Hickey, L. (1989). Introduction. In L. Hickey (ed.), The Pragmatics of Style: 1–12. Routledge.
Hutchby, I. & R. Wooffitt (1998). Conversation Analysis. Polity Press.
Hymes, D.H. (1974). Ways of Speaking. In R. Bauman & J. Sherzer (eds.), Explorations in the ethnography
of speaking: 433–451. Cambridge University Press.
——— (1998). When is oral narrative poetry? Generative form and its pragmatic conditions. Pragmatics
8(4): 475–500.
Jakobson, R. (1960). Closing Statement: Linguistics and poetics. In T.A. Sebeok (ed.), Style in Language:
350–377. MIT Press.
Jefferson, G. (1978). Sequential aspects of storytelling in conversation. In J. Schenkein (ed.), Studies
in the organization of conversational interaction: 219–248. Academic Press.
Kallmeyer, W. (1994). Kommunikation in der Stadt. Band 1: Exemplarische Analysen des Sprachverhaltens
in Mannheim. de Gruyter.
Kallmeyer, W. (ed.) (1996). Gesprächsrhetorik. Narr.
Kotthoff, H. (1997). Erzählstile von mündlichen Witzen. In M. Selting & B. Sandig (eds.), Sprech- und
Gesprächsstile: 123–169. de Gruyter.
Labov, W. (1972). Sociolinguistic Patterns. University of Pennsylvania Press.

Lakoff, R.T. (1979). Stylistic Strategies within a Grammar of Style. In J. Orsanu, M.K. Slater & L.L. Adler
(eds.), Language, Sex and Gender: 53–80. Annals of the New York Academy of Sciences.
Linell, P. (1998). Approaching Dialogue. John Benjamins.
Ochs, E. (1997). Narrative. In T. van Dijk (ed.): 185–207.
Pavlidou, T. (1994). Contrasting German-Greek politeness and the consequences. Journal of Pragmatics
21: 487–511.
Sacks, H. (1992). Lectures on conversation. Blackwell.
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A Simplest Systematics for the Organisation of
Turn-taking for Conversation. Language 50: 696–735.
Sandig, B. (1978). Stilistik. de Gruyter.
——— (1986). Stilistik der deutschen Sprache. de Gruyter.
Sandig, B. & M. Selting (1997). Discourse Styles. In T.A. van Dijk (ed.): 138–165.
Selting, M. (1992). Intonation as a contextualization device. Case studies on the role of prosody,
especially intonation, in contextualizing story telling in conversation. In P. Auer & A. di Luzio
(eds.): 233–258.
——— (1994). Emphatic speech style. Journal of Pragmatics 22: 375–408.
——— (1997). Interaktionale Stilistik. In M. Selting & B. Sandig (eds.): 9–43.
Selting, M. et al. (1998). Gesprächsanalytisches Transkriptionssystem (GAT). Linguistische Berichte
173: 91–122.
Selting, M. & V. Hinnenkamp (1989). Einleitung: Stil und Stilisierung in der interpretativen
Soziolinguistik. In V. Hinnenkamp & M. Selting (eds.), Stil und Stilisierung: 1–23. Niemeyer.
Selting, M. & B. Sandig (eds.) (1997). Sprech- und Gesprächsstile. de Gruyter.
Semino, E. & J. Culpeper (1995). Stylistics. In J. Verschueren & J.-O. Östman & J. Blommaert (eds.),
Handbook of Pragmatics. Manual: 513–520. John Benjamins.
Tannen, E. (1980). A comparative analysis of oral narrative strategies: Athenian Greek and American
English. In W. Chafe (ed.), The pear stories: Cognitive, cultural, and linguistic aspects of narrative
production: 51–87. Ablex.
——— (1984). Conversational Style. Ablex.
——— (1990). You just don’t understand. Women and man in conversation. William Morrow.
Van Dijk, T. (ed.) (1997). Discourse as Structure and Process. Discourse Studies: A multidisciplinary
introduction. Vol. 1. Sage.
Vološinov, V.N. (1929/1976). Freudianism: A Marxist Critique. Academic Press.
Wootton, A. (1989). Remarks on the methodology of conversation analysis. In D. Roger & P. Bull (eds.),
Conversation. An Interdisciplinary Perspective: 238–258. Multilingual Matters.
Conversation analysis
Rebecca Clift, Paul Drew & Ian Hutchby

University of Essex, York University & University of Leicester
1. Introduction
Conversation analysis (CA) has come to be the recognised term for what is in fact the
study of talk-in-interaction (henceforth ‘talk’) in general; while it is not restricted to
the study of conversation, it recognises that ordinary conversation is the basic environ-
ment for language use. In studying conversation and its adaptations, CA contributes to
the development of a naturalistic, observation-based empirical science of actual verbal
behavior alongside work in related fields within pragmatics (e.g. sociolinguistics and
discourse analysis). What makes this approach distinctive is both its analytical focus
and its treatment of interactional data. The focus of CA research is talk as a vehicle for
action and its concern with how participants collaborate in constructing recognisable
and coherent courses of action. To that end, recordings of naturally-occurring inter
action are transcribed in such a way as to capture the temporal production of utter-
ances in turns-at-talk and thus make available for analysis how participants understand
and respond to one another. We shall see in due course a sample of data transcribed
according to conversation analytic conventions and establish what this transcription
makes possible, but we first turn to its emergence as a distinct field of inquiry.
2. Origins and overview
Conversation analysis emerged in the pioneering researches of Harvey Sacks into

the structural organization of everyday language use, at the University of California
in the early 1960s (Sacks 1992). Influenced both by ethnomethodological concerns
with members’ methods of practical reasoning (through an early association with
Harold Garfinkel; see Garfinkel 1967), and by Erving Goffman’s explorations of the
structural properties of face-to-face interaction (Goffman 1963, 1964), Sacks initi-
ated a radical research programme designed to investigate the levels of social order
which could be revealed in the everyday practice of talking.
The hypothesis with which this programme was begun — that ordinary conver-
sation may be a deeply ordered, structurally organized phenomenon — could best
be explored, Sacks reasoned, through the use of recorded naturally-occurring data
which could be looked at repeatedly. Initially, Sacks worked on whatever data were
available to him: principally recordings of calls to a Los Angeles Suicide Prevention
Center. While retaining his sensitivity to the troubles of the persons whose talk he was
studying, Sacks began to develop a unique approach to the study of ordinary language,
which focused on the ‘machinery’ of conversational turn-taking — the methods by
which persons concertedly manage the routine exchange of turns while minimiz-
ing gap and overlap between them — and on the sequential patterns and structures
associated with the management of social activities in conversation (see Sacks, 1992).
Because he had access to a whole corpus of phone calls, Sacks was able to show that the
activities he was investigating were accomplished in systematic and methodical ways
across several calls, irrespective of the personal characteristics or individual histories
of the participants involved. As these researches progressed, and through his collabo-
ration with colleagues Emanuel Schegloff and Gail Jefferson, the available data were
supplemented with expanding corpora of more ‘mundane’ telephone calls; and the
exploratory research was refined and developed. Their landmark paper on turn-taking
in conversation (Sacks, Schegloff & Jefferson 1974) is notable for being the first study
of language to put participants’ own displayed understandings in interaction at the
centre of the analysis; the units of analysis are taken to be those that the participants
themselves are observed to be using, rather than any abstraction or grammatically
defined entity, such as the sentence or clause. This focus on turns and the management
of turn exchange has laid the groundwork for two distinct but related lines of investi-
gation, firstly into the structural organization of talk, and secondly, investigation of the
distinctive methods of turn-taking and activity organization found in the specialized
settings of institutional talk. We now examine these two lines of investigation in turn.
Research into the structural organization of talk encompasses both the mecha-
nisms through which talk is accomplished and the actions and activities prosecuted by
means of the talk itself. We now have compelling accounts of sequence organization
(Schegloff, 1990, 2006), structural organizations in overlapping talk (Jefferson, 1986;
Schegloff, 2000), repair (Schegloff, Jefferson & Sacks, 1977; Schegloff, 1979, 1992,
1997, Drew, 1997), and preference organization (Pomerantz, 1984, Schegloff, 1988a).
The actions analyzed in talk have included the familiar, such as greetings (Schegloff,
1986), invitations (Drew, 1984), complaints (Drew & Holt, 1988; Drew, 1998), teases
(Drew, 1987), questions (Koshik, 2005), answers (Raymond, 2003), anticipatory
completions (Lerner, 1996) and agreements and disagreements (Pomerantz, 1984;
Heritage, 2002), as well as the less so, such as preliminaries of various kinds (Terasaki
1976 [2004]; Schegloff 1980, 1988b), and even the hitherto unidentified, such as con-
firming allusions (Schegloff, 1996) or asserting epistemic priority (Heritage & Raymond,
2005; Stivers, 2005). Alongside these has been research into activities, or a series of
connected actions managed over a sequence, such as storytelling (Jefferson 1978; Lerner
1992) and topic shift (Jefferson, 1984a, 1993; Drew & Holt, 1998, Holt & Drew, 2005).
42 Rebecca Clift, Paul Drew & Ian Hutchby
This structural investigation of talk has imported a rigorously empirical methodology

to traditional areas of linguistic inquiry (see Clift, 2005 for an overview). So CA work
on such items as oh (Heritage, 1984, 1998, 2002), well (Pomerantz, 1984; Schegloff &
Lerner, 2009), actually (Clift, 2001) and so (Raymond, 2004) addresses issues relevant
to lexical semantics; and research on turn-taking has been integrated with work on
prosody (see Local & Kelly, 1986; Couper-Kuhlen, 2001; Local & Walker, 2004; Local,
2005, and the collection in Couper-Kuhlen & Selting, 1996, and Couper-Kuhlen &
Ford, 2004) as well as syntax (see, e.g. Lerner, 1991; Ono & Thompson, 1996, and
the collections in Ochs et al. 1996, and Ford et al. 2002). Yet it is also evident that the
study of naturally-occurring interaction necessitates reference to domains broader
than those of traditional linguistics. There has, for example, been extensive work on
non-lexical response tokens such as Uh huh (Schegloff 1981), laughter (Jefferson,
1984b; Jefferson, Sacks & Schegloff 1987), Mm (Gardner 1997) and the expression
of surprise (Kitzinger & Wilkinson, 2005). And work on membership categorization
(see, e.g. Kitzinger, 2005, Zimmerman, 1998), beyond the realm of the solely linguistic,
returns to one of Sacks’s earliest preoccupations (Sacks, 1992). In addition, conversa-
tion analytic work has expanded beyond the domain of telephone interactions, which
represented a propitious starting point for the analysis of talk-in-interaction precisely
because of the lack of possibly ‘complicating’ visual and gestural contact between par-
ticipants, to study video recordings of interactions with the aim of analyzing the inte-
gration of speech with non-vocal activities. This has revealed further ways in which
participants may mobilize a range of resources – both linguistic and non-linguistic
– in the service of interaction (Charles Goodwin 1981, 2000). Indeed, in cases where
language is impaired, as in cases of aphasia (Charles Goodwin, 1995, the collection
in Goodwin, 2003), or yet to develop, as in young children (Lerner & Zimmerman,
2003), the study of video materials is essential.
That the preponderance of conversation analytic work has been with English
materials should not go unremarked. This is, of course, a function of its origins. However,
there is an increasing diversity of research on languages as varied as Thai (Moerman,
1988), Swedish (Lindström, 1994), Dutch (Mazeland & Huiskes, 2001) Finnish (Sorjonen,
1996, 2002; Ogden, 2004), German (Egbert, 1997a, b), Mandarin (Wu, 2004), Japanese
(Hayashi, 1999; Hayashi et al. 2002; Lerner & Takagi, 1999; Tanaka 1999), and Korean
(Kim, 1999; Park, 1998).
The central focus of conversation analytic work – to identify the generic practices
of talk and develop the tools for analyzing them – has emerged from its engage-
ment with the materials of mundane conversation. Yet, as Sacks’s early work on
helpline calls indicates, conversation analysts have, from the earliest days, worked
on interactional materials from institutional settings, and it is research on institu-
tional talk which forms the other main line of investigation in CA research. We now
have accounts of interactions in courts of law (Atkinson & Drew, 1979, Maynard,
1984), classrooms (Mehan, 1979; Lerner, 1995), public speeches (Atkinson, 1984),
broadcast news interviews (Clayman & Heritage, 2002), calls to the emergency ser-
vices (Zimmerman, 1992), and medical interaction (Stivers & Heritage, 2001, see
the collection in Heritage & Maynard, 2006) amongst others; for an overview, see the
introduction in Drew and Heritage (1992) and the collection therein. CA has also been
used within a broader ethnographic framework by anthropologists such as Moerman
(1988) and Marjorie H. Goodwin (1990, 2002).
In all these applications, CA research aims to reveal how the technical aspects of
speech exchange are the structured, socially organized resources by which participants
perform and co-ordinate activities through talking-in-interaction. Talk is treated as a
vehicle for social action; and also as a principal means by which social organization
in person-to-person interaction is mutually constructed and sustained. Hence it is a
strategic site in which social agents’ orientation to and evocation of the social contexts
of their interaction can empirically and rigorously be investigated.
3. Data, transcription and analysis
Conversation analysis focuses its attention on recordings of actual spates of talk-in-

interaction. These are transcribed using a system which is intended to capture in fine
detail the temporal production of talk; in other words, the characteristics of the sequenc-
ing of turns, including gaps, pauses and overlaps; and elements of speech delivery such as
audible breath and laughter, stress, enunciation, intonation and pitch, all of which have
been shown to have interactional import. The transcription system of course is a selective
system: it does not capture — nor does it claim to capture — all the possible distinctions
that any recording of talk might yield. Rather, transcripts aim to provide a detailed but
accessible rendering of those features that, for CA researchers, prove to be the most rele-
vant for analyzing the methods by which participants concertedly accomplish orderly and
intelligible social interaction. An important methodological upshot is of course transpar-
ency: transcripts are available for inspection alongside the analyses themselves.
The following extract, taken from a recording of a telephone call between two
middle-aged, middle class American women, is representative of the kind of tran-
scripts with which conversation analysts work.
(1) NB: II:2:1–2
1. Nancy: Hello:,
2. Edna: .hh HI::.
3. (.)
4. Nancy: Oh: ‘i:::=‘ow a:re you Edna:,
5. Edna: FI:NE yer LINE’S BEEN BUSY.
6. Nancy: Yea:h (.) my u-fuhh! h- .hhhh my fa:ther’s wife
7. ca:lled me,h .hhh So when she ca:lls me::,h I

8. always talk fer a lo:ng ti:me cuz she c’n afford it
9. en I ca:n’t.hhhhhhhhuh
10. Edna: ÔH:::: : my go:sh=Ah ↑thaht=
11. Nancy: ↑↑AOO:::::hh!
12. Edna: =my phone wuz outta order:

13. (0.2)
14. Nancy: n:No::?
15. Edna: I called my sister en I get this busy en then I’d
16. hang up en I’d lift it up again id be: busy.
17. (0.9)
18. Edna: .hh How you doin’.
19. Nancy: .t hhh Pretty good I gutta rai:se.h .hhhh
20. Edna: Goo:u d.

21. Nancy: Yeh
22. two dollars a week.h

23. (.)
24. Edna: Oh wo:w.
25. Nancy: ↑Ih:::huh hu:h huh,
26. Edna: Wudee gun: do with it a:ll.
27. Nancy: Gol’ I rilly I jis’ don’t know Ah’m gunnuh

28. spend all that money
29. (0.2)
30. Edna: Y’oughta go sho:pping,
31. Nancy: .hhhh Well I should but (.) yihknow et eight
32. dollars a mo:n:th:, anything I’d  buy’d, be using=
33. Edna: hm hmm hm-mm-hm. 
34. Nancy: =up my raise fer ‘alf a YEA:R: ((smile voice))

35. Edna: Ye:a:h. 
36. Edna: .hhhhh Bud j’s lef ’ t’ play go:lf he’s gotta go tuh
37. Riverside
This transcript shows a number of relevant features of the socially organized nature of
talk-in-interaction. At the most basic level, it is designed to display how the talk is orga-
nized into a series of turns. For conversation analysis, however, turns are not just serially
organized but are sequentially organized (Sacks 1987). That is to say, there are describable
loci of order relating one turn to a next; and coordinating them into patterned sequences
through which particular activities are accomplished. Those loci of order are found by
treating the transitions between turns as revealing two kinds of things. First, ‘next turn’ is
the place in which speakers display their understanding of a prior turn’s possible comple-
tion; and second, next turns are places where speakers display their understanding of a
prior turn’s ‘content’, or more specifically, the action it has been designed to do.
In terms of the first issue, empirical observation shows that overwhelmingly,

turn-transitions are coordinated by participants with no or minimal gap and overlap
between turns. Note in the transcript above, for instance, that only four between-turn
gaps occur, and the longest of these is no more than two-tenths of a second, showing
participants’ orientations to the projectability of turns. (We return to the issue of over-
laps presently.) Sacks et al. (1974) proposed that turns are made up of turn-constructional
units (TCUs) — examples are: a sentence, a clause, a phrase, or a single word such as
Hello — which are recognizable by members as meaningful units. The end of any TCU
represents a point at which a next speaker may legitimately make a bid for the floor:
Sacks et al. captured this feature by referring to TCUs as presenting transition-relevance
places (TRPs) at their completion. At a TRP a candidate next speaker may, but need
not, attempt to take a turn; while a current speaker may, but need not, attempt to
produce a next TCU. Current speakers may also select a next speaker, in which case
the one selected is obliged to take a turn at that point. These rules for turn-taking are
context-free: that is, they allow for such local contextual variations as the identities and
number of speakers, length and content of turns, and so on. But they are also context-
sensitive in that they apply to the local circumstances of particular turns in particular
conversations.
The crucial point about these rules is that they are observably oriented to by mem-
bers. An orientation to the possible completion of a turn at TCU completion, and the
legitimate relevance of turn-transition at that point, can be illustrated using the extract
above if we focus on the occurrence of overlap. On the face of it overlapping talk may
be considered evidence of an incoming speaker’s failure adequately to attend to the sta-
tus of a current speaker’s turn. However, we can notice that most instances of overlap
in Extract 1 (marked with left brackets [ for onset and right brackets ] for cessation)
clearly occur in the environment of TRPs (see Jefferson 1986). For instance, at line 9
what Edna’s OH::::: overlaps is a quiet laugh particle, hhh hhh huh, which Nancy fits
onto the end of a TCU: So when she calls me::, h I always talk fer a lo:ng ti:me cuz she c’n
afford it en I ca:n’t. At line 21, what Nancy overlaps is the last phoneme of a recogniz-
able TCU: Edna’s assessment, Goo:ud, to Nancy’s announcement I gutta rai:se.
Other instances appear more complex, but can still be accounted for as orderly.
For instance at line 11, Nancy’s AOO:::::hh! is a high-pitched laugh (or hoot), which
seems to be produced in overlap with — but prior to the recognizable completion
of — Edna’s remark that she thought her phone was out of order (since she had tried
numerous times to get through). However, notice that in her immediately prior turn
Nancy had offered a joke of sorts about talking on the phone for a long time when
her fa:ther’s wife calls, cuz she c’n afford it en I ca:n’t. She then begins quietly to laugh.
Edna’s turn is begun with a loud, and high-pitched OH::::: my gosh, to which Nancy
responds with her similarly-pitched AOO:::::hh! We can say then that Nancy hears
the action that Edna is doing as that of responding to her joke, and starts to laugh by
reference to OH::::: my gosh as a recognizable, and possibly complete, joke-response.

The overlap is ‘complicated’ by Edna’s carrying on her turn, following up Nancy’s quip
by herself quipping about the length of time Nancy’s phone has been engaged: OH:::::
my gosh=Ah thaht my phone wuz outta order.
Focusing on this instance of overlap, and wondering why it occurs in the place it
does, allows us to illustrate a principal policy of conversation analysis, and to suggest
the analytical pay-off from that policy. The policy is to treat anything that occurs in
talk-in-interaction as possibly orderly — to dismiss no detail a priori as disorderly,
trivial, or irrelevant. The pay-off is that we thereby gain an insight into the nature of
participants’ own understandings of what is going on at any moment in interaction, as
displayed in the ways their turns address the turns they are sequentially ‘next’ to.
4. Exhibiting an understanding in next turn
CA’s interest in how the sequential organization of talk can be used to reveal the ways
participants exhibit understanding of one another’s utterances can be illustrated
further with the use of data from the second half of Extract (1). In line 19, Nancy
announces that she gutta rai:se. By the end of the extract, it is evident that the raise has
been presented as, and understood to be, a lousy raise: that Nancy is dissatisfied with
the raise; the raise, in brief, was hardly worth getting. However, none of these things
are said outright. The presentation of the raise as a lousy raise is achieved entirely
indirectly. A central resource used by Nancy here is that of irony. What is interest-
ing to note is the way that ironical complaint, and its uptake, emerge in the course of
a sequence of talk in which Edna’s understanding of the meaning of Nancy’s talk is
observably modified.
The sequence begins at line 18, when Edna inquires, How you doin’. This inquiry
reciprocates Nancy’s earlier ‘ow a:re you Edna: (line 4); the intervening 13 lines hav-
ing been taken up with the talk occasioned by Edna’s remark yer LINE’s BEEN BUSY,
which we focused on above.
In line 21, Nancy’s response to Edna’s inquiry begins: Pretty good. A first thing
to note is that Pretty good is a different kind of response from a How are you-type
inquiry to the response that Edna had given earlier — i.e., FI:NE. Fine represents the
conventional response to How are you (Sacks 1975); it is a no problem response. Pretty
good, on the other hand, represents a ‘downgraded conventional response’: although it
appears very similar to Fine, one kind of work which Pretty good does that Fine does
not do is to adumbrate ‘bad news’ (Jefferson 1980). Basically, if a speaker has some
bad news to report or some trouble to tell, it appears they will use Pretty good in this
sequential environment in order to set up a trajectory in which the trouble might be
elaborated on; by contrast, use of Fine in this position, although it may be followed by
news of some sort, is specifically not followed by bad news.
Adumbrating bad news, then, is a potential property of a Pretty good response to
How are you: potential in that bad news may or may not follow, and may or may not
be told (for instance, Jefferson 1980 analyzes cases in which the troubles talk adum-
brated by a Pretty good response at the beginning of a conversation in fact does not
emerge until some minutes into that conversation). This potentiality makes it a perfect
kind of resource for Nancy to engage in complaining about her raise ironically, and
hence indirectly. The first mention of the raise immediately follows the Pretty good
response; and itself takes the form of a straightforward, unelaborated announcement:
I gutta rai:se. At this stage, then, the news that is being offered is, it appears, good news.
And indeed, it appears that Edna understands that to be the case, as exhibited in her
response in line 20: Goo:ud.
It is only in the next two turns (lines 21–24) that the sense of Nancy’s news being
‘not so good’ in fact emerges. But notice that there is nothing in Nancy’s next turn
itself — Yeh two dollars a week — which overtly suggests that Edna may need to revise
her initial understanding of the news. She does not contest Edna’s congratulatory reac-
tion, for instance by saying, It’s not that good — it’s only two dollars a week’. Rather, her
turn begins with an affirmation, Yeh, and then goes on simply to name the amount.
In other words, the turn does the work of ironicizing the news implicitly: it is left up
to Edna to recognize the significance of two dollars a week, and so to detect the irony
in Nancy’s talk. Edna’s reinterpretation of the announcement appears in the next turn,
line 24. Notice that while her initial reaction, Goo:ud, was fitted to the form of the
announcement as ‘good news’, this second reaction, a downward-intoned Oh wo:w.
(the period marking the downward inflection), equally is fitted to the revised status
of the news following Nancy’s naming of the tiny sum involved. The fact that the turn
begins with Oh is significant here. The marker Oh routinely performs the interactional
work of displaying that its producer has undergone some ‘change of state’ in their
knowledge (Heritage 1984). Thus Edna’s use of the item here connects with the way
she is exhibiting a new understanding of her coparticipant’s talk. More importantly,
the particular kind of new understanding being exhibited is marked in the enuncia-
tion of the wo:w. itself. The downward inflection on wo:w marks the ‘bad news’ — or
perhaps more accurately, ‘no news’ — status to be accorded the raise, just as an alterna-
tive, upward and animated inflection (Oh wow!) would mark the news as something
quite different.
Following that, and Nancy’s burst of laughter in line 25, Edna works to sustain the
joke about the paltriness of the raise by asking, ironically, Wudee gun: do with it a:ll
(line 26) and suggesting that Nancy oughta go sho:pping (line 30). Nancy’s responses to
these turns — especially the heavily ironic Gol- I rilly I jis don’t know how Ah’m gunnuh
spend all that money (lines 27–28), in sustaining the irony, work to display to Edna that
her revised understanding in fact is the correct one.
These brief remarks on the interactional accomplishment of irony illustrate how ‘next
position’ can be treated as a systematic locus in which participants in talk-in-interaction
use essentially local interpretive resources to establish and maintain a shared orientation
on salient aspects of social reality. Furthermore, by focusing on the sequential emergence
of irony in this instance, we have illustrated another central issue in CA: that of the
relationship between particular social actions and the sequential resources by which
they are accomplished. As Schegloff has noted, ‘…both position and composition are
ordinarily constitutive of the sense and import of an element of conduct that embodies
some phenomenon or practice’ (1993:121). Our observations on the ironical form and
ironical uptake of Nancy’s complaint show how indirect actions such as ironical reference
are not simply properties of individual speech acts but are situated features of interaction,
achieved in local space and real time (Schegloff 1988b).
5. Conditional relevance of next position
Next position can also be a place in which specific interactional constraints are opera-
tive. Certain categories of utterance make relevant a circumscribed class of responses
in next position. Canonical examples are: a question, which makes an answer relevant
as the next move; a greeting, which makes a return greeting relevant in next turn; an
invitation, which makes an acceptance or declination relevant in next position; or an
accusation, which makes a rebuttal or justification relevant next. These are all repre-
sentative of types of adjacency pairs: one of the central concepts in CA research.
The concept of the adjacency pair illustrates the way in which particular types of
utterance can be made conditionally relevant by prior turns. The production of a first
pair-part, such as a greeting, sets up a constraint that a next selected speaker should
follow directly by producing the relevant second pair-part — in this case, a return
greeting. Moreover, whatever does follow a first pair-part will be monitored for exactly
how it works as a response to that move. By saying that a second pair-part is condition-
ally relevant given a first, conversation analysts are pointing to the normative character
of the adjacency pair relationship. The normative constraint is strong on two levels.
First, motivational inferences can be drawn from the non-occurrence of a second part
following the production of a first. For instance, not returning a greeting may be taken
as a sign of rudeness; not providing an answer to a question may be taken as indicative
of evasiveness; while not proffering a defence to an accusation may be taken as a tacit
admission of guilt.
Second, the oriented-to relevance of second parts following the production of a
first can remain in play across time: it is not limited to cases of literal adjacency. Thus,
instances in which, say, a question is followed by another question, rather than an

answer, may seem to militate against the force of the adjacency pair concept. But such
cases in fact can quite strongly display the temporally extendible relevance of the adja-
cency pair framework, once we see that the second question routinely represents a first
move in an insertion sequence (Levinson 1983: 304–306; Schegloff 1990). Insertion
sequences defer a second pair-part’s production, but they do not negate its relevance.
A speaker may respond to a question such as, Can I borrow the car? with another
question: How long do you need it? The response to that inserted question — say, Only
a couple of hours — provides a next slot in which a response to the first question is
relevant and can be monitored for.
Another aspect of the normative properties of adjacency pairs lies in the system-
atically different ways that recipients of first parts design the alternative actions to
be done in second position. Invitations, for instance, can be accepted or declined;
requests can be granted or rejected. These alternatives are non-equivalent. That non-
equivalence is traced in the features of turn design through which alternative second
parts are proffered. Broadly, responses which agree or are congruent with the expecta-
tion projected by a first pair-part are produced contiguously and without mitigation.
Responses which diverge from that expectation — which in some way disagree — tend
to be prefaced by hesitations, discourse markers such as Well…, and, unlike congruent
responses, are accompanied by accounts for why the speaker is responding in this way
(Pomerantz 1984; Sacks 1987; Schegloff 1988b).
These different response types are termed preferred and dispreferred respectively.
The concept of preference in CA is not used to refer to the psychological dispositions
or motives of individuals; but to point to just this structural feature of the sequen-
tial organization of some types of adjacency pair. Research has shown that the design
features of dispreferred responses can be used as a resource for the maintenance of
social solidarity in talk-in-interaction. This is so not only in the way that dispreferred
responses may be accompanied by accounts or explanations; but also in the way that
hesitations and other means of ‘marking’ a dispreferred response can provide a source
for a first speaker to revise the original first pair-part in such a way as to try and avoid
disagreement or rejection (Davidson 1984).
These points bring out again the centrality, for CA, of the inferential properties
that attach to speakers’ moves in interaction sequences. They also address the ways
that those inferences have a distinctly moral, or evaluative, dimension. Speakers can
be seen not only to be establishing and maintaining mutual understanding of one
another’s actions in sequences of talk, but also to be holding each other accountable
for those actions. In this sense the adjacency pair framework, and the preference orga-
nization that operates for some types of adjacency pair, constitute an important site in
which to observe the relationships between patterns of language use and structures of
social action.
6. Conclusions
We have stressed two principal areas of analytic concern in conversation analysis:

‘position and composition’, or the relationship between turns at talk in specific
sequences, and the design features of individual turns within sequences. Of course,
other subfields within or related to the pragmatics tradition have been interested in
these ‘syntagmatic’ and ‘paradigmatic’ dimensions of language use. Central to the
originality of CA is the fact that it is not primarily interested in one or other of these
elements, but specifically in the interrelationships between them. This is what has
made conversation analytic insights so valuable to linguists, sociologists, psycholo-
gists, anthropologists, and indeed all those interested in social interaction; through
exploring these relationships, CA reveals an institutionalized substratum of rules
and procedures by reference to which conversationalists engage in recognizable,
coherent and accountable interaction. And this is a form of social organization which
is at the heart of the pragmatic enterprise: the description of language as a vehicle of
social action.
References
Atkinson, J.M. (1984). Our masters’ voices. Methuen.

Atkinson, J.M. & P. Drew (1979). Order in court. Macmillan.
Atkinson, J.M. & J. Heritage (eds.) (1984). Structures of social action. Cambridge University
Press.
Button, G. & J.R.E. Lee (eds.) (1987). Talk and social organization. Multilingual Matters.
Clayman, S. & J. Heritage (2002). The News Interview: Journalists and Public Figures on the Air. Cam-
bridge University Press.
Clift, R. (2001). Meaning in interaction: The case of “actually”. Language 77: 245–291.
——— (2005). Discovering order. Lingua 115: 1641–1665.
Couper-Kuhlen, E. (2001). Interactional prosody: High onsets in reason-for-the-call turns. Language
in Society 30: 29–53.
Couper-Kuhlen, E. & M. Selting (eds.) (1996). Prosody in conversation: Interactional Studies.
Couper-Kuhlen, E. & C.E. Ford (eds.) (2004). Sound Patterns in Interaction. Benjamins.
Davidson, J. (1984). Subsequent versions of invitations, offers, requests and proposals dealing with
potential or actual rejection. In J.M. Atkinson & J. Heritage (eds.): 102–128.
Drew, P. (1984). Speakers’ reportings in invitation sequences. In J.M. Atkinson & J. Heritage (eds.):
129–151.
——— (1987). Po-faced receipts of teases. Linguistics 25: 219–253.
——— (1997). ‘Open’ class repair initiators in response to sequential sources of troubles in conversa-
tion. Journal of Pragmatics 28: 69–101.
——— (1998). Complaints about transgressions and misconduct. Research on Language and Social
Interaction 31: 295–325.
Drew, P. & J. Heritage (1992). Talk at work. Cambridge University Press.

Drew, P. & E. Holt (1988). Complainable matters: The use of idiomatic expressions in making com-
plaints. Social Problems 35: 398–417.
——— (1998). Figures of speech: Figurative expressions and the management of topic transition in
conversation. Language in Society 27: 495–522.
Egbert, M. (1997a). Schisming: The collaborative transformation from a single conversation to
multiple conversations. Language in Society 25: 587–612.
——— (1997b). Some interactional achievements of other-initiated repair in multi-person conversa-
tions. Journal of Pragmatics 27: 611–634.
Ford, C.E., B.A. Fox & S.A. Thompson (eds.) (2002). The Language of Turn and Sequence. Oxford
University Press.
Gardner, R. (1997). The conversation object Mm: a weak and variable acknowledging token. Research
on Language and Social Interaction 30: 131–156.
Garfinkel, H. (1967). Studies in ethnomethodology. Prentice-Hall.
Goffman, E. (1963). Behavior in public places. Free Press.
——— (1964). The neglected situation. American Anthropologist 66: 133–136.
Goodwin, C. (1981). Conversational organization. Academic Press.
——— (1995). Co-constructing meaning in conversations with an aphasic man. Research on Lan-
guage and Social Interaction 28: 233–260.
——— (2000). Action and embodiment within situated human interaction. Journal of Pragmatics 32:
1489–1522.
—— (ed.) (2003). Conversation and Brain Damage. Oxford University Press.
Goodwin, M. H. (1990) He-said-she-said. Indiana University Press.
——— (2002). Building power asymmetries in girls’ interaction. Discourse & Society 13: 715–30.
Greatbatch, D. (1988) A turn-taking system for British news interviews. Language In Society 17:
401–430.
Hayashi, M. (1999). Where grammar and interaction meet: A study of co-participant completion in
Japanese conversation. Human Studies 22: 475–499.
Hayashi, M., J. Mori & T. Takagi (2002) Contingent achievement of co-tellership in a Japanese con-
versation. In C.E. Ford, B.A. Fox & S.A. Thompson (eds.): 81–122.
Heritage, J. (1984). A change-of-state token and aspects of its sequential placement. In J.M. Atkinson &
J. Heritage (eds.): 299–345.
——— (1998). Oh-prefaced responses to inquiry. Language in Society 27: 291–334.
———(2002). “Oh”-prefaced responses to assessments: A method of modifying agreement/disagreement.
In C.E. Ford, B.A. Fox & S.A. Thompson (eds.): 196–224.
Heritage, J. & G. Raymond (2005). The terms of agreement: Indexing epistemic authority and
subordination in talk-in-interaction. Social Psychology Quarterly 68: 15–38.
Heritage, J. & D. Maynard (eds.) (2006). Communication in Medical Care: Interaction between Pri-
mary Care Physicians and Patients. Cambridge University Press.
Holt, E. & P. Drew (2005). Figurative pivots: The use of figurative expressions in pivotal topic transi-
tions. Research on Language and Social Interaction 38: 35–61.
Jefferson, G. (1978). Sequential aspects of storytelling in conversation. In J. Schenkein (ed.), Studies
in the organization of conversational interaction: 219–248. Academic Press.
———(1980). On trouble-premonitory response to inquiry. Sociological Inquiry 50: 153–185.
———(1984a). On stepwise transition from talk about a trouble to inappropriately next-positioned
matters. In J.M. Atkinson & J. Heritage (eds.): 191–222.
——— (1984b). On the organization of laughter in talk about troubles. In J.M. Atkinson & J. Heritage
(eds): 346–369. Cambridge University Press.
——— (1986). Notes on latency in overlap onset. Human Studies 9: 153–183.
——— (1993). Caveat speaker: Preliminary notes on recipient topic-shift implicature. Research on
Language and Social Interaction 26: 1–30.
Jefferson, G., H. Sacks & E.A. Schegloff (1987). Notes on laughter in pursuit of intimacy. In
G. Button & J.R.E. Lee (eds.): 152–205.
Kim, K. (1999). Other-initiated repair sequences in Korean conversation as interactional resources.
Japanese/Korean Linguistics 3: 3–18.
Kitzinger, C. (2005) Heteronormativity in action: Reproducing the heterosexual nuclear family in
after-hours medical calls. Social Problems 52: 477–498.
Kitzinger, C. & S. Wilkinson (2006) Surprise as an interactional achievement: Reaction tokens in
conversation. Social Psychology Quarterly 69: 150–182.
Koshik, I. (2005). Beyond Rhetorical Questions: Assertive Questions in Everyday Interaction. John
Benjamins.
Lerner, G.H. (1991). On the syntax of sentences-in-progress. Language in Society 20: 441–458.
——— (1992). Assisted storytelling: deploying shared knowledge as a practical matter. Qualitative
Sociology 15: 247–271.
——— (1995). Turn design and the organization of participation in instructional activities. Discourse
Processes 19: 111–131.
——— (1996). On the “semi-permeable” character of grammatical units in conversation: conditional
entry into the turn space of another speaker. In Ochs et al. (eds.): 238–276.
Lerner, G.H. & T. Takagi (1999). On the place of linguistic resources in the organization of talk-
in-interaction: A co-investigation of English and Japanese grammatical practices. Journal of
Pragmatics 30: 49–75.
Lerner, G.H. & D.H. Zimmerman (2003). Action and the appearance of action in the conduct of very
young children. In P. Glenn, C.D. LeBaron & J. Mandelbaum (eds.), Studies in Language and
Social Interaction: In honor of Robert Hopper: 441–457.Lawrence Erlbaum.
Lindström, A. (1994). Identification and recognition in Swedish telephone conversation openings.
Language in Society 23: 231–252.
Local, J. (2005). On the interactional and phonetic design of collaborative completions. In W. Hardcastle &
J. Beck (eds.), A Figure of Speech: a Festschrift for John Laver: 263–282. Lawrence Erbaum.
Local, J. & J. Kelly (1986). Projection and “silences”: Notes on phonetic and conversational structure.
Human Studies 9: 185–204.
Local, J. & G. Walker (2004). Abrupt-joins as a resource for the production of multi-unit, multi-
action turns. Journal of Pragmatics 36: 1375–1403.
Maynard, D. (1984) Inside plea bargaining. Plenum.
Mazeland, H. & M. Huiskes (2001) Dutch ‘But’ as a sequential conjunction: Its use as a resumption
marker. In M. Selting & E. Couper-Kuhlen (eds.), Studies in Interactional Linguistics: 141–169.
John Benjamins.
Mehan, H. (1979) Learning lessons. Harvard University Press.
Moerman, M. (1988) Talking culture. University of Pennsylvania Press.
Ochs, E., E.A. Schegloff & S.A. Thompson (eds.) (1996) Interaction and Grammar. Cambridge
University Press.
Ogden, R. (2004). Non-modal voice quality and turn-taking in Finnish. In E. Couper-Kuhlen &
C.E. Ford (eds.): 29–62. Benjamins.
Ono, T. & S. Thompson (1996). Interaction and syntax in the structure of conversational discourse. In E.
Hovy & D. Scott (eds.), Discourse processing: An interdisciplinary perspective: 67–96. Springer Verlag.
Park, Y-Y. (1998). A discourse analysis of contrasting connectives in English, Korean and Japanese
conversation: With special reference to the context of dispreferred responses. In A. Jucker &
Y. Ziv (eds.) Discourse markers: Descriptions and theory: 277–300. John Benjamins.
Pomerantz, A. (1984). Agreeing and disagreeing with assessments. In J.M. Atkinson & J. Heritage
(eds.): 57–101.
Raymond, G. (2003). Grammar and social organization: Yes/no type interrogatives and the structure
of responding. American Sociological Review 68: 939–967.
——— (2004) Prompting action: The stand-alone “so” in ordinary conversation. Research on Lan-
guage and Social Interaction 37: 185–218.
Sacks, H. (1975). ‘Everyone has to lie’. In M. Sanches & B. Blount (eds.) Sociocultural dimensions of
language use: 57–80. Academic Press.
——— (1987). On the preferences for agreement and contiguity in sequences in conversation. In
G. Button & J.R.E. Lee (eds.): 54–69.
——— (1992). Lectures on conversation, 2 vols. Blackwell.
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn-
Schegloff, E.A. (1979). The relevance of repair to syntax-for-conversation. In T. Givón (ed.), Syntax
and semantics, volume 12: Discourse and Syntax: 261–286. Academic Press.
——— (1980). Preliminaries to preliminaries. Sociological Inquiry 50: 104–152.
——— (1981). Discourse as an interactional achievement. In D. Tannen (ed.), Georgetown University
roundtable on languages and linguistics: 71–93. Georgetown University Press.
——— (1986). The routine as achievement. Human Studies 9: 111–151.
——— (1988a). On an actual virtual servo-mechanism for guessing bad news. Social Problems
35: 442–457.
——— (1988b). Presequences and indirection. Journal of Pragmatics 12: 55–62.
——— (1990). On the organization of sequences as a source of ‘coherence’ in talk-in-interaction.
In B. Dorval (ed.), Conversational organization and its development: 51–77. Ablex.
——— (1992). Repair after next turn: The last structurally provided defense of intersubjectivity in
conversation. American Journal of Sociology 97: 1295–1345.
——— (1993). Reflections on quantification in the study of conversation. Research on Language and
Social Interaction 26: 99–128.
——— (1996). Confirming allusions: Toward an empirical account of social action. American Journal
of Sociology, 102: 161–216.
——— (1997). Practices and actions: Boundary cases of other-initiated repair. Discourse Processes
23: 499–545.
——— (2000). Overlapping talk and the organization of turn-taking for conversation. Language in
Society 29: 1–63.
——— (2006). Sequence Organization in interaction: A primer in conversation analysis. Cambridge
University Press.
Schegloff, E.A., G. Jefferson & H. Sacks (1977) The preference for self-correction in the organization
of repair in conversation. Language 53: 361–382.
Schegloff, E.A. & G. Lerner (2009). Beginning to respond: Well-prefaced responses to Wh-questions.
Research on Language and Social Interaction 42(2): 91–115.
Sorjonen, M-L. (1996) On repeats and responses in Finnish conversations. In E. Ochs, E.A. Schegloff &
S.A. Thompson (eds.): 277–327.
——— (2002). Recipient activities: The particle no as a go-ahead response in Finnish conversations.
In C.E. Ford, B.A. Fox & S.A. Thompson (eds.): 165–195.
Stivers, T. (2005). Modified repeats: One method for asserting primary rights from second position.
Research on Language and Social Interaction 38: 131–158.
Stivers, T. & J. Heritage (2001). Breaking the sequential mould: Narrative and other methods of
answering “more than the question” during medical history taking. Text 21: 151–185.
Tanaka, H. (1999). Turn-taking in Japanese Conversation. A Study in Grammar and Interaction.
John Benjamins.
Terasaki, A. (1976) [2004]. Pre-announcement sequences in conversation. In Gene Lerner (ed.),
Conversation Analysis: Studies from the First Generation: 174–224. John Benjamins.
Wu, R-J. R. (2004). Stance in Talk: A Conversation Analysis of Mandarin Final Particles. John Benjamins.
Zimmerman, D.H. (1992) Achieving context: openings in emergency calls. In G.Watson & R.M. Seiler
(eds.), Text in Context: Contributions to Ethnomethodology: 35–51. Sage.
——— (1998). Identity, context and interaction. In C. Antaki & S. Widdicombe (eds.), Identity in
Talk: 87–106. Sage.
Conversation types
Auli Hakulinen
1. Introduction
One way of defining conversation is to see it as something that people engage in off-
duty. According to Levinson’s influential text book (1983: 284), conversation is “the
predominant kind of talk in which two or more participants freely alternate in speak-
ing, which generally occurs outside specific institutional settings”. On the basis of
more recent research, one can view types of conversation as forming a continuum
with mundane talk at one end and carefully pre-planned interviewing or some other
strictly role and status dependent form of institutional interaction at the other end.
What all forms of conversation share, however, is the fact that it is through them that
we, as human beings, manage our daily affairs and construct and make sense of our
life and activities.
The classificatory dimension for kinds of conversation that easiest comes to mind
is one of formality. Informal talk is unplanned, and takes place between peers, among
family and friends; formal talk is typical of exchanges in public, between strangers
and in the broadcast media. When talking about a phenomenon he calls ‘conversa-
tionalization’ that is taking place both in the broadcast media and in printed media
Fairclough (1992: 204) is using conversation as a synonym to ‘informal talk’. Another
factor that is often seen as influencing the interactional outcome of talk is setting or
context, both physical and social. As a matter of fact, however, no one-dimensional
typology of conversations will correctly capture the variety and richness of conversa-
tional activity, nor does the notion of context prove particularly helpful as a decisive
factor in defining the linguistic characteristics or the sequential patterns of talk in
specific situations.
In this entry, conversations are grouped along several dimensions, in order to
attempt to capture the multi-faceted character of talk-in-interaction. Accordingly,
context is taken to be a complex and multi-dimensional phenomenon. (For a similar
view, see the chapter on Context and Contextualization in volume 1 of this series).
Conversational activity has been approached from a range of different research
traditions and disciplines from sociology and social psychology to more linguistically
oriented discourse and text analysis. These approaches vary with respect to their views
on the relationship between theory and empirical analysis, the position of ‘ordinary
conversation’ vis-a-vis institutional talk, and the relevance of ethnographic data on the
56 Auli Hakulinen
interpretation of what is going on in a conversation. Moreover, they draw on different

conceptual frameworks, sharing the interest in the dialogical nature of language use,
however. For social-psychologically oriented discourse analysts, discourse is a means
of building identities and of constructing social reality (Potter & Wetherell 1989); dis-
course analysts of the critical linguistic tradition (Fairclough 1992) combine a sys-
temic linguistic analysis developed by Halliday (1985) with a political and ideological
criticism. Two pioneering examples of more strictly linguistically oriented discourse
analytical studies were Labov & Fanshel’s (1977) exhaustive description of therapeutic
discourse, and Sinclair & Coulthard’s (1975) classification of teacher’s and pupil’s turns
of talk in the classroom; the latter one inspired by the ideas of speech act theory, origi-
nally developed within the philosophy of language (Austin 1962; Searle 1969). The
method of interactional sociolinguistics as developed by Gumperz (1982) has specifi-
cally influenced research on intercultural communication by showing the relevance of
cues other than verbal for the interpretation of a message.
Within sociology, two important innovations that have influenced not only socio-
logical work on human co-operation but also the development of interactionally oriented
pragmatics were Goffman’s (1981, 1986) ideas of the interaction order and, especially,
the research tradition of conversation analysis that was developed on the ideas of
Garfinkel’s ethnomethodology by Harvey Sacks (1992; Sacks, Schegloff & Jefferson
1974; cf. also Heritage 1984). It is the latter methodology that has most consistently
focussed on and pursued the analysis of conversation as a phenomenon in itself, and it
has been greatly influencing empirical work not only in sociology but in anthropology,
education, linguistic pragmatics and other neighbouring fields of study. Therefore, in
what follows we will mostly make use of conversation analytic references.
2. Three basic dimensions
There are three basic dimensions that cut across all kinds of talk-in-interaction:
Firstly, the channel through which conversation is carried out — whether auditory,
visual or visual-cum-auditory (multi-modal). Secondly, conversation may be dyadic or
multi-party, and thirdly, conversation may be mundane/everyday or institutional by
nature. Let us look at each of these dimensions at a time.
2.1 The channel

Although an enormous amount of information in face-to-face conversation is passed
through the visual channel, interaction is perfectly fluent without it: between people
who are blind, or talk in the dark or on the telephone. Early conversation analytic
work concentrated on telephone conversations, partly in order to isolate the purely
auditory aspects of interaction. Much of the sequential organization of talk follows the
same patterning irrespective of the channel, but e.g. the beginnings and closings are
managed in a way dependent on the channel (Schegloff 1979; Schegloff & Sacks 1973),
manifesting different cultural patterns in the ways the participants are identified
(cf. e.g. Houtkoop-Steenstra 1991; Hopper 1992). To what extent other verbal means
of communication depend on the accompanying visual cues has yet to be attested, in
particular the gestural behaviour vis-a-vis the use of response tokens like mm, yeah,
or no, which also have visual counterparts like nodding and head-shaking, need to be
further investigated. (See Gardner 1995 and Sorjonen 2001 for extensive analyses of
verbal response tokens in conversation.)
The post-war period has witnessed a quick expansion of the broadcast media,
and both radio and TV have become increasingly interactive. Much research has been
devoted to political interviews in particular (cf. e.g. Heritage & Roth 1995). Interviews
and debates resemble ‘genuine’ conversations outside the media, but they also show
certain features that are not there in a conversation carried out without an audience. For
example, as the interviewer is not the primary recipient of the interviewee’s answers,
(s)he does not use the kinds of responses that mark the acknowledging of new infor-
mation (Heritage & Greatbatch 1991). Broadcast language does not merely represent
one-way communication from performers to audience but involves, in increasing
numbers, phone-in programmes where the audience is expected to engage in public
exchange with hosts who are professional broadcasters (Hutchby 1996). In this genre
that is a mixture of telephone conversation and broadcast interaction, then, lay persons
may be called ‘show-relevant callers’ whose identities are locally created and may vary
according to the nature of the show they participate in (Hester & Fitzgerald 1999).
In the 1990’s, a fast growing type of interaction has been computer mediated com-
munication (Herring 1996) through the internet, which has given rise to new forms
of written exchange that closely resemble both dyadic (e-mail) and multi-person
(chat-boxes, etc.) conversations. Due to these new means, the notion of conversation
is extended to encompass purely visual, i.e. written exchanges that, in many ways, do
resemble talk in that they increasingly take place in real time.
Sign languages, as used by the deaf and the deaf-blind, are dependent on visual
and tactile channels in interaction. Most of the research on these types on language has
focussed on their grammar and semantics. Recently, some work has been done on the
conversational aspects of signed languages. McIlvenny (1995) has conducted pioneer-
ing work on the visual conversations in the Finnish sign language. In Sweden, Mesch
(1998) is a dissertation on the tactile aspects of questioning in deaf-mute interaction.
2.2 Dyadic vs. multi-person

Dyadic conversations differ from those between many speakers first of all with respect
to the ways in which turn allocation works. In a dyad, it is basically a question of
speaker and recipient incessantly taking turns, with the recipient always being the
58 Auli Hakulinen
addressee as well, the recurring pattern being AB AB AB, and the relevant variability,
when does speaker change occur rather than who speaks next. In multi-party talk,
turn allocation tends to follow more complex rules (cf. the classic paper by Sacks,
Schegloff & Jefferson 1974), which are to a high extent contex-independent. With three
speakers or more, the issue of who is to speak next is perpetually relevant. A speaker
may self-select, steal the turn or be appointed by the current speaker as the addressee
by gaze, naming or by topic selection. Within a conversation, participants may align
with certain others to form a team — something that often happens in disputes
particularly. The more there are speakers in a group, the easier it will get split into two
or more separate conversations (see Egbert 1997).
In addition, the positions and orientations of the participants in multi-party
conversations can be illuminatingly described in terms of the so called participation
framework, to replace the received dichotomies of speaker and listener(s), or speaker
vs. addressee. This Goffmanian (1981) concept was introduced to the analysis of deic-
tic phenomena by Levinson (1988), and first applied to the analysis of conversation by
C. Goodwin (1981; 1987). In multi-party conversations in particular, the participants
create and re-create discourse identities for each other, thereby forming an ever chang-
ing context for the talk. Goodwin shows how the current speaker can, by way of eye
contact, by turn design, and even by hesitation at turn beginnings, gain the attention
of a particular participant, and make her/him the addressed recipient; while the others
remain at that point non-addressed participants.
2.3 Everyday vs. institutional

Classifying conversations into everyday and institutional is partly methodological,
partly hypothetical. The difference between them is not clear cut, and it can with rea-
son be claimed that all talk is institutional in so far as language itself is an institution,
and so is the family, one of the typical settings for mundane conversation. However,
as was pointed out at the beginning, it has proved useful to discover and explicate
the basic kinds of rules and practices that any unplanned conversation might exhibit
before taking a stand as to how these resources are made use of in more restricted,
specific kinds of contexts, such as institutional settings.
A major difference between everyday and institutional conversations is that the
latter have as a rule an overall structure and are organized into phases: they have a
recognizable beginning and an end, and the conversationalists — or at least one of
them — know what to expect from the talk and what the topics to be dealt with are.
In everyday conversation, there are normally no recognizable phases, and no formal
procedures (other than the local rules underlying turn taking) to guide the speakers
through the event. As Bergmann (1990) pointed out, one of the consequences is that
in mundane talk, the topic may easily ebb out, which may lead to speakers resorting to
anything at hand for a new topic. Another, obvious difference in multi-party conversa-
tions is that in the institutional ones, the turns are pre-allocated as there is normally
someone chairing or otherwise responsible whereas in mundane conversations, the
turns are being allocated on a local basis.
In institutional conversations, the participants have typically different rights
and responsibilities so that it is most likely that their institutional roles have a bear-
ing on their discourse identities. In interviews for example, it is the right and the
duty of the interviewer to present the questions and the interviewee is obliged to
answer. Thus, institutional conversations are often asymmetrical. (For more about
asymmetry, see Section 4.) Moreover, although institutional conversations can be
shown to play upon the resources that are there in ordinary interaction, there are
also clear differences. If the doctor asks the patient in the beginning of a consulta-
tion, How are you today, it is not expected that the patient answer back, Fine, how are
you? In opening interviews with the elderly (Coupland et al. 1993), older people are
likely to answer the same question with expressions such as Not too bad considering
or Mustn’t grumble.
3. Types of institutional talk
The institutionality of an encounter is not crucially dependent on the setting. Rather, it

is the conversational rights and responsibilities of the participants, and especially their
orientation to each other’s roles and to specific tasks that constitute the institutional
nature of a conversation; as Heritage (1984) has put it, institutions are “talked into
being”. (For detailed examples, see Drew & Heritage 1992; Drew & Sorjonen 1997.)
Also, institutional interaction has typically a goal or purpose, which is not necessarily
a constitutive element of mundane conversations.
Much empirical research has been carried out on dyadic conversations between
lay/client and professional speakers, most of it on face-to-face encounters. These types
of interaction share one feature: the orientation to a question — answer format of the
exchange. The range and variation as to the turn design of both the questions and
the answers within this overall format can be detected by comparing research on e.g.
courtroom interaction (Atkinson & Drew 1979), police interrogation (Linell & Jönsson
1991), and news interviews (Clayman, Heritage & Greatbatch 2002) on one hand, with
counselling in career guidance (Vehviläinen 1999), intensive therapy talk (Labov &
Fanshel 1977), and medical interaction in general (Heritage & Maynard 2004) on the
other hand. As the goals of these activities are quite different, and the agenda of the
professional who proposes the questions as well, the sequential outcome may differ,
despite the fact that it is the professional who is doing the questioning. In classroom
interaction, the question — answer format is manifested in a one — many arrangement
60 Auli Hakulinen
(McHoul 1978; 1990), and typically consist of a three-part sequence that includes a
question, an answer, and a follow-up (see also Sinclair & Coulthard 1975).
Certain institutionalized multi-party conversations have systematically exploited
ways of conversing that may be found occurring in a more marginal way in ordinary
talk. For example, AA-talk follows very strict (written down) rules of interacting with
one speaker talking at a time, allowing no interruptions or questions, thereby prescrib-
ing extended turns to each speaker at a time. In addition, each turn has an end that will
inform others of the imminent change of speakers. The aim of the strictly regulated
way of interacting is to lead the members into articulating their shared experience of
being an alcoholic by connecting their stories to those in previous turns: to produce
‘second stories’ (Arminen 1998). Counselling work, with delicate issues like AIDS
in particular, has led to specific patterns of turn taking called ‘circular questioning’,
as shown by Peräkylä (1995). Although meetings tend to have negotiations as their
generic constitution, negotiations form a conversational genre that exists practically
everywhere (cf. Firth 1995; see also Section 5 below). Some types have become insti-
tutionalized activities, with clearly definable profiles and aims. It is possible to make
further classifications into business negotiations, political negotiations between inimical
parties, and negotiations carried out in the labour market. However, it is not yet fully
clear to what extent these may systematically differ among themselves with respect to
the kind of linguistic means that the participants make use of.
Having an institutional role, say that of a scientist, does not automatically make the
talk produced as representing institutional talk. For example, talk among colleagues
over a specimen in the microscope, or over a piece of linguistic data on a tape, may
come close to everyday talk (Goodwin 1998).
4. Symmetry and asymmetry in conversations
Understanding asymmetries in interaction presupposes knowledge of what is typical of

ordinary talk between people of similar status and shared outlook (Heritage 1984: 240).
Methodologically, it has proved useful to try and discover the recurrent typicalities of
talk in interaction in their own right before launching into explanations that resort to
social attributes like gender, (professional) status or social class, or to more diffuse
interpersonal phenomena like power or politeness.1 With respect to the discourse-
external roles and such as social and professional identities that co-conversationalists
embody, conversation is, on the face of it, seldom symmetrical. In many types of
1. This approach is not subscribed to by all researchers. Traditions that are closer to sociolinguis-
tics will prefer to use ethnographic and cultural information in the analysis from the outset.
conversation, the roles — with the (linguistic) rights and responsibilities that they
imply — are institutionally given. However, it is an empirical question to what extent
and under what kind of circumstances these roles are displayed in the actual discourse
identities of the participants, i.e. whether and how the identities are reflected in or
constituted by the linguistic choices made during an encounter.
As Linell & Luckmann (1991) point out, ‘asymmetry’ can mean a range of
different things. By it, one can refer to the inequality — in terms of rights and/or
abilities — inherent in child–adult, novice–expert, lay–professional, foreign–native,
aphasic (~ autistic, Alzheimer, etc.)–non-aphasic relationships, presumably manifest-
ing in globally asymmetrical types of conversation. However, it is not necessarily so
that the less able would not excel in certain areas where the ‘normal’ participant prove
less successful. Additionally, asymmetry can be used to characterize the less easily
attestable relationship of imbalance in male — female interaction. First of all, there is
the tricky issue of whether gender proves to be an omni-relevant category, or one that
is locally made relevant in certain kinds of context. It is equally debatable and a serious
empirical question to what extent gender is inextricably entangled with attributes like
power, status, and age in any kinds of context.2
On the other hand, ‘asymmetry’ may simply mean a temporary lack of reciproc-
ity, caused by e.g. a speaker’s unfamiliarity with the topic at hand, by a temporal lapse
of one of the parties’ attention, or by some features of the setting that may be felt as
discriminatory by one or more of the participants. The latter kind of asymmetry will
of course cut across any type of conversation.
Linell & Luckmann (1991: 9) take the view that asymmetry is an ‘intrinsic fea-
ture of dialogue’. However, the cross-categorial asymmetry is easier to detect than the
intra-categorial one, and it has until now received a lot more attention in research;
see, however, e.g. Drew (1991) on the exploitation, by the participants in ordinary
conversation, of asymmetries in their respective knowledge. Harness Goodwin (1990)
is an important work as it addresses the issue of peer group interaction, looking into
adolescent interaction in separate gender groups.
For obvious reasons, much more work has been done on male-female conver-
sation than on communication between men, or among women, and hypotheses
about conversation among women being inherently different from that among men
still await proof. Within accommodation theory (cf. Giles, Coupland & Coupland
1991), the interest lies more in finding out how old people are talked to than how
they talk among themselves. Goodwin (1995) gives insights to the way successful
even if highly asymmetric conversation is carried out with a severely aphasic man;
2. For an overview, consult the articles in Tannen (ed., 1993); for a problematization and explica-
tion of the gender imbalance see Crawford (1995).
62 Auli Hakulinen
Klippi (1996) is a pioneering report on the gestural and other compensating abilities
of the aphasic in group conversation. Of growing relevance is research on native–
non-native conversation not only with respect to its possible applications to second
language teaching but also as a means of coming to grips with the notions of shared
knowledge and breakdown of communication between ‘equals’, respectively (Kalin
1995; Wagner 1996).
5. Conversation types and communicative genres
By virtue of the way in which types of conversation have been delineated above
they could also be called genres. However, ‘communicative genre’, or ‘communica-
tive activity type’ is a term that is particularly suitable for a further sorting out of
different kinds of ordinary conversation such as intimate talk, family dinner-table
conversation, troubles telling, small talk, etc. Genre can be viewed as the fabric of
discourse as it were, consisting of much practised patterns of language use. Some
genres are linguistically more easily characterizable, and germane to certain contexts
(cf. Linell 1998: 238–241) whereas others may cut across different types of conversa-
tion. For example, the question — answer patterns are constitutive of a number of
institutional settings, as was mentioned earlier, but they may well be resorted to in
family situations between parents and children, in quarrels between spouses, etc.
Troubles talk is a genre that occasionally emerges in talk between friends that trust
each other (Jefferson 1988). This type of talk in everyday setting shares characteristics
with some aspects of counselling talk (Vehviläinen 1999: 127) as well as certain
sequences in medical settings (Haakana 1999) even though professionals will most
likely behave in a ‘neutral’ way vis-à-vis the client’s talk in comparison with the affiliation
and/or second stories expected of friends.
Less conspicuous with respect to the linguistic means employed, and as yet much
less known is negotiation talk, which could also be seen as a genre of its own. Negotia-
tions have been intensively studied within economics and political theory, but much
less within discourse and conversational studies (Firth 1995). Their generic constitu-
tion has probably more to do with the kinds of turns used than the structure of turn
taking as such. Any bargaining tends to have proposals and counter proposals, sug-
gestions and refusals, arguments and counter arguments (Maynard 1984) but none
of these are perhaps obligatory elements in a negotiation. Negotiations, furthermore,
often glide into and out of disagreements and conflict talk, which create alliances and
alignments among the speakers (Kangasharju 1996), but negotiations need not involve
conflicts. In fact, drawing a clear line between genres is not very fruitful; quarrelling
and conflict talk come close to negotiations but they could be viewed as genres of their
own, as well.
6. Conclusion
In the past three or four decades, one of the important changes within the study of
language has been the rise of various traditions of use-oriented research to comple-
ment the earlier, almost exclusively structure-oriented, to a large extent unempirical
research that is still taken to be the mainstream within linguistics in many parts of the
world. We now understand better than say twenty years ago ‘the situationally relevant
meanings of utterances’ — but by far not enough. Results are cumulating of work on
how we as members of a linguistic community manage our daily routines in mundane
encounters and make use of the practices learned within them3 to cope with the ever
expanding jungle of different types of institutional situations that we encounter. In
particular, the investigations on interaction between ‘normal’ and less able (children,
aphasic, non-native etc.) speakers help us understand the importance of co-operation
in conversation for making sense of and verbalising individual experience, but also,
hopefully, deepen our understanding of the variability within ‘normality’. The fast
development of highly sophisticated techniques for analysing video recorded interac-
tion will, in the course of time, also increase our informedness of the status of non-
verbal means such as gesture as constitutive elements in various types of conversation.
Linguists are not alone in studying different forms of discourse. Within conversa-
tion analysis, there is more and more collaboration between sociologists, linguists,
and anthropologists in particular when investigating institutional talk. Various forms
of discourse analysis, on the other hand, are the fashion of the day in social studies.
Negotiation, which is a recurring genre of interaction in many if not most types of
conversation, has received much attention within economics, political science, and
social psychology. Consultations and advice giving is a genre that is growing in impor-
tance in many different kinds of work place.
References
Arminen, I. (1998). Therapeutic interaction. The Finnish Foundation for Alcohol Studies, Vol. 45.
Atkinson, J.M. & P. Drew (1979). Order in Court: the Organization of Verbal Interaction in Judicial
Settings. MacMillan.
Austin, J.L. (1962). How to do Things with Words. Harvard University Press.
3. Not much is known yet about how children acquire different kinds of conversation and become
aware of differences between genres. Wootton (1997) has made an in-depth analysis of the devel-
opment of a child’s sequential reasoning parallel with the learning of different kinds of requests.
This is a line of research that has great promise in enriching out understanding of the importance
of a rich conversational environment for the child in the early formative years.
64 Auli Hakulinen
Bergmann, J. (1990). On the local sensitivity in conversation. In I. Markova & K. Foppa (eds.), The
Dynamics of Dialogue: 201–226. Harvester Wheatsheaf.
Boden, D. & D. Zimmerman (eds.) (1991). Talk and Social Structure. Polity Press.
Clayman, S., J. Heritage & D. Greatbatch (2002). The News Interview: Studies in the History and
Dynamics of a Social Form. Cambridge University Press.
Crawford, M. (1995). Talking difference. On Gender and Language. Sage.
Coupland, N. & J.F. Nussbaum (eds.) (1993). Discourse and Lifespan Identity. Sage.
Drew, P. (1991). Asymmetries of knowledge in conversational interaction. In I. Markova & K. Foppa
(eds.): 21–48.
Drew, P. & M.-L. Sorjonen (1997). Institutional dialogue. In T. van Dijk (ed.), Discourse as social
interaction, 92–118. Sage.
Egbert, M. (1997). Schisming: The collaborative transformation from a single conversation to
multiple conversations. ROLSI, 30 (1): 1–59.
Fairclough, N. (1992). Discourse and Social Change. Polity Press.
Firth, A. (ed.) (1995). The Discourse of Negotiation. Pergamon.
Gardner, R. (1995). On some uses of the conversational token mm. Ph.D. diss., University of
Melbourne.
Giles, H., J. Coupland & N. Coupland (eds.) (1991). Contexts of Accommodation. Cambridge University
Press.
Goffman, E. (1981). Forms of Talk. Blackwell.
——— (1986). Frame Analysis. Northeastern University Press.
Goodwin, C. (1981). Conversational Organization. Interaction between Speakers and Hearers. Academic
Press.
——— (1987). Forgetfulness as an interactional resource. Social Psychology Quarterly 50(2): 115–131.
——— (1995). Co-constructing meaning in conversation with an aphasic man. ROLSI 28, 233–260.
——— (1998). Transparent vision. In E. Ochs, E. Schegloff & S. Thompson (eds.), Interaction and
Grammar: 370–404. Cambridge University Press.
Haakana, M. (1999). Laughing matters. A conversation analytical study of laughter in doctor-patient
interaction. Unpublished Ph.D. diss., University of Helsinki.
Halliday, M.A.K. (1985). Introduction to Functional Grammar. Edward Arnold.
Harness Goodwin, M. (1990). He-Said-She-Said. Talk as Social Organization among Black Children.
Indiana.
Gumperz, J.J. (1982). Discourse Strategies. Cambridge University Press.
Heritage, J. (1984). Garfinkel and Ethnomethodology. Polity Press.
Heritage, J. & D. Greatbatch (1991). On the institutional character of institutional talk. In D. Boden &
D. Zimmerman (eds.): 93–137.
Heritage, J. & D. Maynard (eds.) (2004). Practising Medicine: Talk and action in primary care encounters.
Heritage, J. & A.L. Roth (1995). Grammar and institution: questions and questioning in broadcast
media. Research on Language and Social Interaction 28(1): 1–60.
Herring, S. (1996). Computer-mediated Communication: Linguistic, Social and Cross-cultural Perspec-
tives. John Benjamins.
Hester, S. & R. Fitzgerald (1999). Category, predicate and contrast: Some organizational features in a
radio talk show. In P.L. Jalbert (ed.), Media Studies: Ethnomethodological Approaches: 171–193.
University of America Press.
Hopper, R. (1992). Telephone Conversation. Indiana University Press.
Houtkoop-Steenstra, H. (1991). Opening sequences in Dutch telephone conversation. In D. Boden &
D. Zimmerman (eds.): 232–250.
Hutchby, I. (1998). Confrontation Talk: Argument, asymmetries and Power on Talk Radio. Lawrence
Erlbaum.
Jefferson, G. (1988). On the sequential organization of laughter in talk about troubles. Social Problems
35(4): 418–441.
Kalin, M. (1995). Coping with problems of Understanding. Studia Philologica Jyväskyläensia 36.
Kangasharju, H. (1996). Aligning as a team in multiparty conversation. Journal of Pragmatics
26: 291–319.
Klippi, A. (1996). Conversation as an achievement in aphasics. Finnish Literature Society.
Levinson, S. (1983). Pragmatics. Cambridge University Press.
——— (1988). Putting linguistics on a proper footing: Explorations in Goffman’s concepts of partici-
pation. In P. Drew & A. Wootton (eds.), Erving Goffman: 161–227. Polity.
Labov, W. & D. Fanshel (1977). Therapeutic Discourse: psychotherapy as conversation. Academic
Press.
Linell, P. (1998). Approaching Dialogue. John Benjamins.
Linell, P. & L. Jönsson (1991). Suspect stories: perspective-setting in an symmetrical situation.
I. Markova & K. Foppa (eds.): 75–100.
Linell, P. & T. Luckmann (1991). Asymmetries in dialogue: some conceptual preliminaries. In
I. Marková & K. Foppa (eds.): 1–20.
Marková, I. & K. Foppa (eds.). Asymmetries in Dialogue. Harvester Wheatsheaf.
Maynard, D.W. (1984). Inside Plea Bargaining. The language of Negotiation. New York: Plenum Press.
McHoul, A.W. (1978). The organization of turns at formal talk in the classroom. Language in Society
7: 182–213.
McIlvenny, P. (1995). Seeing Conversations: Analysing Sign LanguageTalk. In P. ten Have &
G. Psathas (eds.), Situated Order: Studies in the Social Organisation of Talk and Embodied Action:
129–50. University Press of America.
Mesch, J. (1998). Teckenspråk i taktil form. [Sign language in tactile form; Turn taking and questions
in conversations between the deaf-blind.] University of Stockholm.
Peräkylä, A. (1995). AIDS counselling: Institutional Interaction and clinical Practice. Cambridge
University Press.
Potter, J. & M. Wetherell (1989). Discourse and Social Psychology. Sage.
Sacks, H. (1992). Lectures on Conversation. Blackwell.
Sacks, H., E. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn-
taking for conversation. Language 50(4): 696–735.
Schegloff, E. (1979). Identification and recognition in telephone conversation openings. In G. Psathas
(ed.), Everyday Language: Studies in Ethnomethodology: 23–78. Irvington Publishers.
Schegloff, E. & H. Sacks (1973). Opening up closings. Semiotica 7: 289–327.
Searle, J.A. (1969). Speech Acts. An essay in the Philosophy of Language. Cambridge University press.
Sinclair, J. & M. Coulthard (1975). Towards an Analysis of Discourse: the English used by teachers
and pupils. Oxford University Press.
Sorjonen, M.-L. (2001). Responding in Conversation. John Benjamins.
Tannen, D. (ed.) (1993). Gender and Conversational Interaction. Oxford University Press.
Vehviläinen, S. (1999). Structures of Counselling Interaction. A Conversation Analytic Study of
Counselling Encounters in Career Guidance Training. Helsinki University Press.
Wagner, J. (ed.) (1996). Special issue of Journal of Pragmatics, Vol. 26.
Wootton, A.J. (1997). Interaction and the development of mind. Cambridge University Press.
Ethnomethodology
Alan Firth
Newcastle University
1. Introduction
Ethnomethodology is a radical form of sociology that offers a distinctive perspec-

tive on the nature and researchability of social order. Ethnomethodologists study
the taken-for-granted, commonsense reasoning practices through which members
of society coordinate, structure and understand their everyday social activities.
Through practical, situated actions, people are seen to create and sustain social
order. At core, ethnomethodology is concerned with social action, with inter-
subjectivity, and with linguistic communication — topics that lie at the heart of
pragmatics. In addition to producing a diversified range of percipient and ground-
breaking studies in its own right, ethnomethodology in large measure underpins
conversation analysis. The insights and foundational ideas of ethnomethodology
can be increasingly detected in ethnography, social psychology, cognitive science,
and in researches into language and social interaction. Yet despite a growing influ-
ence on such areas, ethnomethodological writings can be notoriously difficult to
penetrate, particularly for the uninitiated. Due to a general resistance to theorize
upon its own unconventional research procedures and findings, the perceived lack
of ‘theory’ or ‘method’ (in the accepted senses of the word), and a predilection
for a compressed and sometimes jargonistic prose style,1 ethnomethodology has
acquired and maintained a position as a marginalized, and commonly misunder-
stood, sociological enterprise.
1. Garfinkel’s seminal Studies in ethnomethodology (1967) is a prime example (and in many ways
the genesis) of the compressed and jargonistic argot that has become one of ethnomethodology’s
hallmarks. In recent years several monograph-length ‘Introductions’ to ethnomethodology have
been produced, and many of these are admirable attempts to render the enterprise more easily
accessible, though without diminishing the importance of reading Garfinkel’s Studies (see, e.g.,
Mehan & Wood 1975; Leiter 1980). A particularly lucid description and appraisal of Garfinkel’s
work, and of the field of study in general, is Heritage (1984).
Ethnomethodology 67
2. Overview
The term ‘ethnomethodology’ was coined in the 1950s by the American sociologist, and
ethnomethodology’s leading intellectual figure, Harold Garfinkel.2 Researching on jurors’
deliberations, Garfinkel became interested in the jurors’ reasoning procedures — which
he entitled ‘methods’ — through which collective decisions were made and verdicts
reached. Such ‘methods’ were not based upon specialized knowledge or particular
types of logical or legalistic thinking; rather they were predicated upon everyday, com-
monsense knowledge.
Garfinkel’s invention of the term ‘ethnomethodology’ marked the initiation of a
program of studies that was to undermine prevailing structural functionalist preoccupa-
tions with ‘scientific’ explanations of how social order is constituted and maintained.
Rather than seeking corrective ‘scientific’ or evaluative explanations of social order,
studies in ethnomethodology were to be conducted with ‘ethnomethodological indif-
ference’ (see Garfinkel & Sacks 1970). This was to be achieved by suspending (or ‘brack-
eting’) corrective and evaluative aspirations, and focusing instead on how members
of society themselves, through their own taken-for-granted, commonsense practices
(‘methods’) accomplish the social order. To ethnomethodologists, social order, sense-
making, and rationality are socially accomplished phenomena, ceaselessly ‘worked at’,
and locally and contingently achieved. ‘Ethno-methods’ are thus people’s own (and not
scientists’) methods for both accomplishing and exhibiting such phenomena; they are,
as Turner (1974: 83) phrases it, members’ “production practices”.
Ethnomethodology holds that it is the study of how ‘practices’ or ‘methods’ are used
by members3 acting contingently in particular and concrete circumstances to create and
sustain social order, that provides for a fundamental reappraisal and a detailed under-
standing of the nature of that order. Such an ‘order’ is seen to be created and known,
not on the basis of an external or ‘objective’ matrix, but endogenously — from within
temporally ordered activities. Ethnomethodology, as Garfinkel (1967: 185) puts it, is
concerned with the question of how, over the temporal course of their actual
engagements, and ‘knowing’ the society only from within, members produce stable,
accountable practical activities, i.e., social structures of everyday activities.
The program of ethnomethodology holds that no social activity or event — no matter how
apparently trivial or insignificant — is immune from or unworthy of study. In a substantial
and variegated body of work spanning thirty years, ethnomethodologists’ descriptions
2. On the origins of the term ‘ethnomethodology’, see Garfinkel (1974).

3. The term used by ethnomethodologists to refer to the ‘mastery of natural language’; see
Garfinkel & Sacks (1970: 342).
68 Alan Firth
of ‘production practices’ include the way clinicians compile records (Garfinkel 1967:
Ch. 6), the way ‘gender’ (Garfinkel 1967: Ch. 5) and ‘ethnicity’ (Moerman 1968) is
accomplished, the methods by which conversationalists categorize persons (Sacks 1972),
how jazz piano playing is learned (Sudnow 1978), the way scientists construct reportable
findings and scientific objects (Garfinkel et al. 1981), how social relations are textually
mediated (Smith 1984), and the methods by which software engineers follow techni-
cal guidelines (Button & Sharrock 1994). (For an extensive bibliography of eth-
nomethodological studies, see Coulter 1990: 475–559.)
In effect, Garfinkel launched a full-frontal assault on the assumption of an
objective, external reality whose existence can be known and described independently
of human agency.4 He proposed that members are not simply in ‘objective’ social set-
tings; rather, that they in fact do those settings (cf. McDermott & Wertz 1976: 166).
And it is through the doing — and only there — that the order, sense, rationality, and
stability of social activities is made possible and achieved. To study this achievement,
ethnomethodology focuses on “a member’s knowledge of his ordinary affairs, of his
own organized enterprises, where that knowledge is treated by us as part of the same
setting that it also makes observable” (Garfinkel 1974: 17). That is, the observable-
reportable quality of actions — i.e., the accountability of actions — cannot be divorced
from the way members within a setting organize and coordinate their actions. Neither, by
implication, can social settings be accounted for by recourse to the scientist’s a priori
analytic constructs. As Garfinkel (1967: 33) inimitably puts it:
a leading policy [for ethnomethodology] is to refuse serious consideration to the
prevailing proposal that efficiency, efficacy, effectiveness, intelligibility, consistency,
planfulness, typicality, uniformity, reproducibility of activities — i.e., that rational
properties of practical activities — be assessed, recognized, categorized, described by
using a rule or standard obtained outside actual settings within which such properties
are recognized, used, produced and talked about by settings’ members.
In sum, the purpose of the program of study that came to be known as ethnomethod-
ology was to analyze social organization solely from members’ ‘experience structures’,
rather than through ‘objectively’ or ‘scientifically’ deduced categories, constructs or
schemes. For it is uniquely through such ‘experience structures’, ethnomethodolo-
gists maintain, that activities and events are recognizably produced as orderly and
rational phenomena.
4. This is not to assert that an ‘objective’, ‘external’ reality has no relevance in people’s everyday
assumptions and reasoning. Indeed, Pollner’s (1987) ethnomethodological thesis on mundane
reason is predicated on the belief that perceptions of reality, self, and the social world are under-
pinned by a presupposed ‘external’ and ‘objective’ world.
Ethnomethodology 69
3. Social action, social knowledge
In forwarding ethnomethodology as a program of study, Garfinkel was reacting critically

to Talcott Parsons’s hugely influential sociological work, The structure of social action
(1937). In this work, Parsons propounded what he called the ‘voluntaristic theory of
action’. This theory in essence contends that social order is made possible and maintained
as a result of individual agents’ internalization of a given culture’s shared social norms.
It is these norms, argued Parsons, that are internalized as ‘need dispositions’, such
that each agent is socialized into wanting to conform to a community’s norms. Agents
thus voluntarily choose to conform to the shared norms, and, Parsons argued, it is this
norm-driven choosing that renders society stable and orderly.
However, although proclaiming his work as a theory of action, Parsons omitted
or at least seriously distorted issues of central importance. In particular, Garfinkel saw
that the voluntaristic theory gave rise to three major problems which center on actors’
knowledge of their practical, concrete circumstances. These problems relate to norms
and rules, the contexted character of actions, and rationality (cf. Heritage 1984: Ch. 2).
3.1 Norms and rules

To Parsons, and the majority of a generation of sociologists following him, socially
learned and internalized norms exert a constraining and determining influence on
action. Garfinkel saw this as an erroneous portrayal of the individual agent as a ‘judg-
mental dope’, acting passively and unreflectingly on his or her surroundings (Garfinkel
1967: 68). In a series of case studies, Garfinkel showed that, in actual situations of occur-
rence, ‘norms’ and ‘rules’ are flexible, interpretative resources that participants (most
often tacitly) ‘orient to’ in order to understand and recognize meaningful (and deviant)
behavior. Moreover, rules cannot be mechanically followed, regardless of their elaborate-
ness and specificity. Rule-following necessitates ceaseless, contingent, judgmental ‘work’
— requiring ad hocing practices. Whenever a rule is applied, it is done so “for another
first time” (Garfinkel 1967: 9). Furthermore, rules in themselves are insufficient to deter-
mine or explain action: they must be adapted for each and every occasion of their use.
Focusing on the activity of coding in a psychiatric clinic, Garfinkel (1967: Ch. 6) found
that participants interpretatively ‘filled in’ relevant (though ‘missing’) ‘background’
information where required, and allowed perceived anomalies to pass, in order to make
sense of the rules and make the coding categories ‘fit’ the data (see also Cicourel 1968;
Zimmerman 1971). Generalizing on the study, Garfinkel (1967: 21) concluded that
No matter how definitely and elaborately instructions had been written, and despite
the fact that strict actuarial rules could be formulated […] in every instance, for every
item, ‘et cetera’, ‘unless’, ‘let it pass’ and ‘factum valet’ accompanied the coder’s grasp of
the coding instructions as ways of analyzing actual folder contents.
70 Alan Firth
In order to code adequately and reliably, it was found that the coders were assuming
contextual knowledge of the clinic’s procedures — the very phenomena that the coding
was intended to determine (Heritage 1987: 236). Such ad hocing practices were invari-
ably treated as a ‘nuisance’ by the coders and coding designers, resulting in attempts
to remedy the ‘nuisance’ by stipulating increasingly more elaborate and explicit rules
of procedure. However, complaining about ad hocing getting in the way of rules and
instructions is, wrote Garfinkel (1967: 22), “very much like complaining that if the
walls of a building were only gotten out of the way one could see better what was
keeping the roof up”. Garfinkel’s aim here was not to demean coding procedures in
particular, or, indeed, question the existence of rules and norms in general; rather it
was first to reject the Parsonian idea that rules ‘cause’ or ‘explain’ behavior; second to
underscore the inherent ‘looseness’ and resourcefulness of rules; and third to dem-
onstrate that their use in actual settings was empirically researchable as a topic in its
own right.
3.2 The contexted character of actions

The ‘voluntaristic theory of action’ was oblivious to participants’ actual actions carried
out in concrete settings. By way of contrast, ethnomethodologists take participants’
actions as their point of departure, and attend to the way those actions are, in ‘seen
but unnoticed’ ways, made observable and orderly in concrete and particular situa-
tions. The concern with local occasions of action has two major manifestations in
ethnomethodology; these are (1) the indexicality of actions, and (2) the contextually
reflexive qualities of actions.
3.2.1 Indexicality
Parsons’s insistence that shared, internalized norms underpin behavior extended
to linguistic communication. Interpersonal meaning is thus implicitly explained by
recourse to a theory of semiology that claims a correspondence between ‘signs’ and
‘referents’; socialized individuals are viewed as being cognizant of the correspondences.
In a paper co-authored with Harvey Sacks (Garfinkel & Sacks 1970), Garfinkel rejected
this notion of intersubjectivity, proposing instead a procedural basis for interpersonal
meaning. The case against Parsons’s (Augustinian) conception of meaning was
predicated on the linguistic notion of ‘indexicality’. Drawing upon Schütz’s phenom-
enological writings (see below), Wittgenstein’s (1958) and Peirce’s (1932) work on natural
language, and Bar-Hillel’s (1954) philosophy, Garfinkel & Sacks (1970) expanded the
notion in two ways: first by showing that it is not only pro-terms (I, her, you, etc.) and
deictics (here, this, that, etc.) that are indexical — the concept, they argued, could
usefully be extended to all linguistic forms, and second by showing that the indexicality
of actions is both ubiquitous and irremediable. Hence, logicians’ (and, in particular,
Ethnomethodology 71
sociologists’) efforts to somehow scientifically ‘remedy’ indexicality are, in the final

analysis, doomed to failure. Any linguistic form (utterance, description, expression,
etc.) is irremediably and inescapably indexical. That is, the sense of an utterance is not
fixed, but motile, yet contextually and procedurally ‘discoverable’ by virtue of members
contingently invoking an array of commonsense-based interpretative ‘methods’ and
contextual information. No linguistic actions — including utterances that purport to
‘step outside’ events and ‘gloss’ or ‘formulate’ them — can be performed ‘context-free’
or ‘non-indexically’. Indeed, even a remark’s characterization as, for example, ‘out of
context’, is dependent upon that very utterance’s indexicality.
In light of the inherent indexicality of natural language, Garfinkel (1967: 28–29)
recommends that we “drop the assumption” of a sign-referent correspondence. Once
this assumption is dropped, “then what the parties talked about could not be distin-
guished from how the parties were speaking” (Garfinkel, op. cit.: 28). In other words,
understanding is accomplished not on the basis of pre-established shared meanings,
but procedurally and contextually — in that what is said is invariably assessed in a
particular, local context, by particular persons, at a particular moment.
Importantly, Garfinkel and Sacks point out that the indexicality of natural language
is a resource, rather than a handicap, for conversationalists. Following Mannheim,
Garfinkel contends that the inherent vagueness of language is overcome in the way
members tacitly invoke the ‘documentary method’ of interpretation. This involves the
search for “an identical homologous pattern underlying a vast variety of totally dif-
ferent realizations of meaning” (Mannheim, in Garfinkel 1967: 78). Garfinkel (ibid.)
continues: “The method is recognizable for the everyday necessities of recognizing
what a person is ‘talking about’ given that he does not say exactly what he means”
(see 4.1. below). Thus, in the sense that the search for ‘homologous patterns’ engages
interactants’ considerations of the way language use and interpretation is embedded
in particular activities in specific circumstances, the ethnomethodological position on
intersubjectivity displays strong affinities with the Wittgensteinian notion of ‘language
games’ (see Wittgenstein 1958, also Heritage 1984: 310).
3.2.2 Reflexivity
According to ‘mainstream’ (Parsonian) thinking in sociology, actors share common
perceptions about normative requirements of social settings; the argument runs that
it is these requirements that serve to direct behavior in different contexts. Ethnometh-
odology proposes an alternative conception. Rather than situation-specific norms
being determinative of behavior, they enter a matrix of contextual (tacitly known)
information about context. This information is centrally underpinned by the observed
behavioral patterns (actions), which in turn continually — procedurally — feed back
into the matrix of contextual information. To ethnomethodologists, then, actions are
not causally related to context; they are reflexively related: They help to construct and
72 Alan Firth
elaborate the very context in which they are an intelligible — accountable — part. An
abiding task for ethnomethodological studies has been to detail this reflexive character
of actions. Heritage (1984: 109) claims that the concern with the ‘reflexive account-
ability’ of actions is the central pillar of Garfinkel’s work. The notion of ‘reflexivity’
emphasizes the way observable-reportable actions — ‘accounts’ — and context, mutually
and ‘endlessly’ elaborate one another. Behavior and talk are thus simultaneously in
and about the settings they describe (Leiter 1980: 139). Thus Garfinkel (in Hill &
Crittenden 1968: 208):
One might talk of a kind of endless reflexivity that accounts have in the same way that
‘talk’ folds back on the setting in which it happens to illuminate the features of that
setting and thereby illuminate the talk’s own features as well.
While an account is produced and recognized as intelligible and orderly on the basis
of its embeddedness in a particular (temporal, sequential, spatial, etc.) context, the
account itself simultaneously — and hence reflexively — permits participants to con-
struct, observe and elaborate the order and intelligibility in the selfsame context (cf.
Mehan & Wood 1975: 13; Heritage 1984: 106–110).
One of the most famous ethnomethodological studies that focuses specifically on
the ‘reflexivity’ of ‘accounts’ is Wieder’s (1974a) study of the way the “convict code”
was used in a halfway house for paroled narcotics addicts. The ‘convict code’ identi-
fied a set of (unwritten) ‘maxims’ of conduct (e.g., ‘don’t snitch’, ‘don’t take advantage
of other residents’, ‘share what you have’). Wieder (1974b) observed that on several
occasions, while he was engaged in conversation with a resident, the resident would
respond to certain questions by retorting ‘you know I won’t snitch’. This utterance,
Wieder showed, was multifunctional, and reflexively related to its context of use. It
formulated the preceding action (a request for the resident to snitch), provided the
resident with a reason for not complying with the request, and indexed the relationship
of the interactants as ‘outsider’ and ‘resident’. Talk which invoked the code was thus
not simply a description of life in a halfway house; it was also an ‘embedded method’
for seeing and describing ‘routine’, ‘typical’, ‘normal’ and ‘deviant’ behavior in the
setting. Talk that ‘tells the code’ thus simultaneously exhibits and accomplishes the
observed order:
this [convict code] talk was at the same time part of life in the halfway house, and it
was a part that was itself included within the scope of things over which the code had
jurisdiction. It is in this sense that talk involving the code was reflexive within the
setting of its occurrence. (Wieder 1974b: 152, emphasis added)
Wieder’s work on the reflexivity of accounts is a vivid illustration of the inherent

‘looseness’ of rules (see 3.1. above). In showing that accounts (here invocations of the
‘convict code’) and settings mutually configure, Wieder demonstrates that rules are
Ethnomethodology 73
not invoked by residents for merely abstract or descriptive purposes; rather, rules do
contextual work. They are a means for particular members to do particular things at
particular times, and to ‘index’ those doings — as, e.g., a rebuttal or an assurance — as
they are being done (cf. Mehan & Wood 1975: 141).
3.3 Rationality
Rather than ignoring or (as was usually the case among sociologists) subsuming
commonsense knowledge under the canons of ‘scientific rationality’, ethnomethodol-
ogy asserts that the situated application of commonsense knowledge can be studied
directly in its own right. After all, Garfinkel (1967: 35–37, 96–103) pointed out, while
asserting the superiority of scientific knowledge and rationality over the layman’s
commonsense knowledge, social scientists’ theories and observations are ineluctably
predicated upon commonsense knowledge; such knowledge is both the scientist’s and
the layman’s taken-for-granted, ‘uninteresting’ and ‘unseen’, resource. For Garfinkel,
the unavoidably ‘local’ nature of this ‘resource’ is ethnomethodology’s ‘topic’ of inquiry
(cf. Zimmerman & Pollner 1971).
4. Commonsense reasoning
In order to document the empirical character of the actor’s ‘seen but unnoticed’ com-
monsense knowledge and reasoning, Garfinkel exploited the exceptional fecundity of
the conceptual framework developed by the German phenomenological sociologist,
Alfred Schütz (see Garfinkel 1967: 36ff.). Schütz (1962, 1964) had worked extensively
on Husserl’s notion of the ‘natural attitude’ of the Lebenswelt (‘life world’). This notion
emphasizes the actor’s ability to draw upon past experience, and to ‘suspend doubt’
about the stability of the ‘life world’ in the face of apparent anomalies. Schütz champi-
oned the pragmatically-infused verstehende processes of interpretation, the processes
through which actors (social scientists and laymen alike) ascribe meaning and ‘rational-
ity’ to social events and activities. This is done by contingently applying commonsense ‘type
constructs’ (e.g., goals, sign systems, motives, and typifications) in order to understand
‘what is going on here and now’. Moreover, while these constructs are assumed to be
shared ‘for all practical purposes’, they are also seen to be fundamentally ‘elastic’, ‘rules of
thumb’, and thus eminently capable of revision in accordance with local circumstances.
To Schütz, ‘rationality’ cannot be divided into ‘scientific’ and ‘non-scientific’ categories,
since ‘rationality’ is irredeemably located in contingent commonsense knowledge and
actual social practices. Hence it is on the basis of a ceaseless ‘fixed-contingent’ dialec-
tic that everyday intersubjective understanding — and, by extension, social order — is
made possible. As Schütz (1962: 55) puts it: “commonsense knowledge of everyday life
74 Alan Firth
is sufficient for coming to terms with fellow-men, cultural objects, social institutions —
in brief, with social reality”.
One of ethnomethodology’s major achievements was to demonstrate that Schütz’s
conceptual framework for social action is an empirically-researchable domain. This
was memorably achieved in Garfinkel’s breaching experiments, the purpose of which
was to provide insight into how ‘ordinary’, ‘routine’ social activities are constructed
through a reliance on a set of ‘seen but unnoticed’ assumptions underlying practical
actions. Garfinkel (1967: 36) asserted that the aim of the breaching experiments was
to ‘make commonplace scenes visible’. In order to do so, the experiments started with
familiar, life-as-usual scenes, and then ‘made for trouble’. ‘Trouble’ entailed the experi-
menters acting markedly ‘out of the ordinary’. The reason for doing so being that
[t]he operations that one would have to perform in order to multiply the senseless features
of perceived environments; to produce and sustain bewilderment, consternation, and
confusion … and to produce disorganized interaction should tell us something about
how the structures of everyday activities are ordinarily and routinely produced and
maintained. (Garfinkel 1967: 37–38)
Garfinkel showed that experimenters’ unexpected, discrepant behavior resulted in the

subjects’ immediate attempts to restore normality. For example, when playing the game
of ticktacktoe (‘noughts and crosses’), the experimenter would first erase the other
player’s mark, relocate it in a different square, and then make their own mark. All this
was carried out without indication of anything being unusual. In half the 247 trials,
the unknowing subjects treated the move as a gesture with hidden but definite signifi-
cance, such as making a sexual pass, commenting on the subject’s stupidity, or making
an impudent gesture. Garfinkel (1967: 72) reports that identical effects occurred when
experimenters joined without invitation an unacquainted group of conversationalists;
or when the experimenters asked their unsuspecting co-conversationalists to clarify
commonplace remarks (e.g., A: How are you? B: Fine. A: How do you mean, ‘fine’;
in what way?); or when the experimenters allowed their gaze to wander randomly
during a conversation; or, in the course of a conversation, and without indication of
anything unusual happening, the experimenters brought their faces up to the subject’s
until their noses were almost touching.
Garfinkel (1967: 94) reports that the ‘normality’ perceived by the subjects arose
from the subjects’ documenting work. In essence, such ‘work’ is the substance and sus-
tenance of social order, consisting of a complex array of contingent commonsense
knowledge, background assumptions, and tacitly known skills. “Searching for and
determining pattern” (1967: 94) — i.e., the ‘documentary method of interpretation’ — is
one interpretative resource that enabled the subjects to make sense of otherwise inco-
herent actions. This ‘method’, and the Husserlian notion of the ‘natural attitude’, are key
elements in ethnomethodology’s conceptual apparatus, and are claimed to permeate all
Ethnomethodology 75
acts of mundane perception and cognition. Following Schütz, Garfinkel maintains that
the ‘natural attitude’ is part of a member’s ‘commonsense assumptions’ that underpin
social interaction and the ‘accommodative work’ which the interaction necessitates.
The assumptions include:
1. Searching for a normal form. That is, when discrepancies and/or ambiguities
appear, people suspend doubt and search for an assumed ‘normal form’ that would
account for the discrepancies.
2. Doing the reciprocity of perspectives. This entails sustaining the assumption that
each participant would have the same experiences if they were to change places.
3. Employing the et cetera principle. This assumption entails participants ‘filling in’
(observedly ‘missing’) information during social activities; in this way, partici-
pants let unclarities or anomalies ‘pass’ in the belief that they will be subsequently
clarified (cf. Mehan & Wood 1975: 101–2; Cicourel 1973: 84–88).
The breaching experiments were meant to cast light upon a tenebrous terrain: the
taken-for-granted resources through which order, sense, routineness and normality
are accomplished. What is remarkable for ethnomethodologists is that such accom-
plishments remain, for members, unremarkable.
5. Developments in ethnomethodology
Ethnomethodology can be seen to have been forged by three separate though contigu-
ous aims that originated in Garfinkel’s writings. These aims are manifest in past and
current work in the field. While each centers around ethnomethodological concerns
with the accomplished nature of social order, taken together they account for the var-
iegated research undertakings of the enterprise in its present form. The first aim was
to challenge prevailing sociological theory on the nature of social order. In particular,
by highlighting practical, commonsense knowledge rather than ‘objective’, ‘scientific’
knowledge as the basis upon which the social world can be known and described,
ethnomethodology has seriously brought into question sociology’s conception of
the social actor and, more fundamentally, raised the issue of the ontological status of
sociology as a scientific discipline. Consequently, a major scholarly preoccupation in
contemporary ethnomethodology is the enterprise’s position in and implications for
social theory. Recent publications dealing specifically with this topic include Wilson &
Zimmerman (1980), Button (1991), Hilbert (1992), and Garfinkel & Wieder (1992).
The second aim of ethnomethodology was to emphasize the constitutive role of
cognition in the organization of social activities. This aim was manifest most clearly in
Garfinkel’s reliance on Schütz’s phenomenological framework in order to account for
how ‘rationality’ was constructed upon the cognitive foundations of commonsense
76 Alan Firth
assumptions. A prominent strain of ethnomethodological research reflects this aim,

notably Cicourel’s Cognitive sociology and its focus on ‘interpretive procedures’ (1973),
and Coulter’s The social construction of mind (1979) and Rethinking cognitive theory
(1983); a useful review of this work is provided by McHoul (1988). The cognitive
dimension of ethnomethodology has also extended to research into human-machine
interaction, as illustrated by Suchman’s Plans and situated actions (1987).
The third aim of ethnomethodology was to describe the local, contingent and
reflexive ‘work’ through which concrete social settings, identities and activities are
rendered recognizable and meaningful. This third aim has undoubtedly produced the
most substantial amount of ethnomethodological research. Garfinkel’s (1967) studies
on the ‘accomplished’ character of gender (i.e., the study of the transsexual, ‘Agnes’),
and the methods through which clinical records are compiled, were groundbreaking
studies which reflect this third aim. Other prominent researches include Sudnow’s
Passing on (1967), and Wieder’s Language and social reality (1974a). During the last
decade, important ethnomethodological work has been carried out into the ‘practices’
and discourses of natural scientists. This work attempts to show how, in the laboratory
worksetting, scientific objects and reports are constructed; see, e.g., Gilbert & Mulkay
(1984) and Lynch (1985).
It is also in terms of this third aim that ethnomethodology has exerted significant,
indeed formative, influence on the enterprise known as ‘conversation analysis’ (CA).
Although there is some debate about contemporary CA’s precise relationship to eth-
nomethodology (for discussions, see Bilmes 1986: 165–166 and Clayman & Maynard
1994), there is general agreement that CA’s intellectual heritage is firmly grounded in
ethnomethodology. Whereas most CA work is directed to the sequential analysis of
interactive talk, in which case the ethnomethodological heritage is at best alluded to,
others (e.g., Boden 1994) are attempting to unify CA methodologies and sensibilities
with ethnomethodology’s aim of describing the accomplished character of particular
activities carried out in concrete social settings.
6. Conclusion
Ethnomethodology is a diversified sociological enterprise that is concerned with the

nature of an ‘accomplished’ social order and people’s engagement in that order. With
its insistence on the primacy of empirical, situated investigations of social actions,
its concern with intersubjectivity and interpersonal meaning, its exhortation that
social activities are locally accomplished achievements, and with its abiding inter-
ests in linguistic communication, ethnomethodology has important affinities with
the encompassing concerns of pragmatics. It is only recently that scholars have begun
addressing directly the nature of the affinities and the relevance of ethnomethodology
Ethnomethodology 77
for studies in pragmatics. Bilmes (1993), in his study of conversational implicature,

proposes that the empirical foundations of pragmatics will be strengthened if scholars
are able to address natural language from an ethnomethodological viewpoint. This
will necessitate studying actual, situated instances of language use, and will awaken us
to the artful, contingent, and socially-accomplished character of language-in-action.
Discovering and explicating the methods and practices of people using language is
thus an ethnomethodological enterprise.
Throughout its development, ethnomethodology has established and maintained
its position as a radical enterprise, located in the intellectual hinterlands of sociology.
More recently, however, some ethnomethodologists — e.g., Pollner (1987: xvi) and
Hilbert (1992) — have emphasized that the enterprise is not as distanced from ‘main-
stream’ sociology as was earlier contended, and that time is ripe for ethnomethodolo-
gists to initiate a dialogue with and within sociology.
Ethnomethodology is a continually developing field of inquiry; it is built upon radi-
cal foundations and offers a unique and challenging perspective on human sociation;
its research methodologies are eclectic, some might say haphazard and arbitrary, relying
on a seemingly capricious assortment of ethnographic field work, audio- and video-
recordings, experimentations, field notes, transcriptions, first-hand observations, and
more besides. Given all this, the ‘dialogue’ that Pollner and Hilbert call for should also be
extended to neighboring disciplines, one of the most relevant of which is pragmatics.
References
Bar-Hillel, Y. (1954). Indexical expressions. Mind 63: 359–379.

Bilmes, J. (1986). Discourse and behavior. Plenum Press.
——— (1993). Ethnomethodology, culture, and implicature. Pragmatics 3: 387–409.
Boden, D. (1994).The business of talk. Polity Press.
Button, G. (ed.) (1991). Ethnomethodology and the human sciences. Cambridge University Press.
Button, G. & W.W. Sharrock (1994). Practices in the work of software development. In A. Firth (ed.),
The discourse of negotiation. Pergamon Press.
Cicourel, A.V. (1968). Police practices and official records. In R. Turner (ed.) (1974): 85–95.
——— (1973). Cognitive sociology. Penguin.
Clayman, S.E. & D.W. Maynard (1994). Ethnomethodology and conversation analysis. In P. ten Have &
G. Psathas (eds.), Situated order. University of America Press.
Coulter, J. (1979). The social construction of mind. Macmillan.
——— (1983). Rethinking cognitive theory. Macmillan.
——— (ed.) (1990). Ethnomethodological sociology. Edward Elgar.
Douglas, J. (ed.) (1971). Understanding everyday life. Routledge & Kegan Paul.
Garfinkel, H. (1967). Studies in ethnomethodology. Prentice Hall.
——— (1974). The origins of the term ‘ethnomethodology’. In R. Turner (ed.): 15–18.
Garfinkel, H. & H. Sacks (1970). On formal structures of practical actions. In J.C. Mckinney &
E.A. Tiryakian (eds.), Theoretical sociology: 338–366. Appleton Crofts.
78 Alan Firth
Garfinkel, H. et al. (1981). The work of a discovering science construed with materials from the optically
discovered pulsar. Philosophy of the Social Sciences 11: 131–158.
Garfinkel, H. & D.L. Wieder (1992). Two incommensurable, asymmetrically alternate technologies
of social analysis. In G. Watson & R.M. Seiler (eds.), Text in context: 175–206. Sage.
Gilbert, G.N. & M. Mulkay (1984). Opening Pandora’s box. Cambridge University Press.
Heritage, J. (1984). Garfinkel and ethnomethodology. Polity Press.
——— (1987). Ethnomethodology. In A. Giddens & J. Turner (eds.), Social theory today: 224–272.
Polity Press.
Hilbert, R.A. (1992). The classical roots of ethnomethodology. Chapel Hill Press.
Hill, R.J. & K.S. Crittenden (eds.) (1968). Proceedings of the Purdue symposium on ethnomethodology.
Institute for the Study of Social Change, Purdue University.
Leiter, K. (1980). A primer on ethnomethodology. Oxford University Press.
Lynch, M. (1985). Art and artifact in laboratory science. Routledge & Kegan Paul.
McDermott & M. Wertz (1976). Doing the social order. Reviews in Anthropology 3: 160–174.
McHoul, A.W. (1988). Language and the sociology of mind. Journal of Pragmatics 12: 339–386.
Mehan, R.P.H. & H. Wood (1975). The reality of ethnomethodology. John Wiley & Sons.
Moerman, M. (1968). Being Lue. In R. Turner (ed.) (1974): 54–68.
Parsons, T. (1937). The structure of social action. McGraw Hill.
Peirce, C.S. (1932). Collected papers, vol. 2. Harvard University Press.
Pollner, M. (1987). Mundane reason. Cambridge University Press.
Sacks, H. (1972). On the analysability of stories by children. In J.J. Gumperz & D. Hymes (eds.),
Directions in sociolinguistics: 325–345. Holt, Rinehart & Winston.
Schütz, A. (1962). Collected papers, vol. 1. Martinus Nijhoff.
——— (1964). Collected papers, vol. 2. Martinus Nijhoff.
Smith, D.E. (1984). Textually mediated social organization. International Social Science Journal
36: 59–75.
Suchman, L. (1987). Plans and situated actions. Cambridge University Press.
Sudnow, D. (1967). Passing on. Prentice Hall.
——— (1978). Ways of the hand. Bantam Books.
Turner, R. (1974). Practical reasoning in organizational settings. In R. Turner (ed.): 83.
——— (ed.) (1974). Ethnomethodology. Penguin.
Wieder, D.L. (1974a). Language and social reality. Mouton.
——— (1974b). Telling the code. In R. Turner (ed.): 144–172.
Wilson, T.P. & D.H. Zimmerman (1980). Ethnomethodology, sociology, and theory. Humboldt Journal
of Social Relations 7: 52–88.
Wittgenstein, L. (1958). Philosophical investigations (2nd ed.). Blackwell.
Zimmerman, D.H. (1971). The practicalities of rule use. In J. Douglas (ed.), Understanding everyday
life: 221–238. Routledge & Kegan Paul.
Zimmerman, D.H. & M. Pollner (1971). The everyday world as a phenomenon. In J. Douglas (ed.),
Understanding everyday life: 80–103. Routledge & Kegan Paul.
Erving Goffman
Jim O’Driscoll
University of Huddersfield
1. Introduction
Erving Goffman (1922–1983) was not a linguist. Although he admired linguists as the
only group of scholars with “the capacity to study the small behaviours of their own
society and to treat the conduct of their own familiars objectively” (1971: xviii), only one
aspect of language form, tangentially, attracted his attention (see end of Section 4). He
was not interested in semantics at all. He did, it is true, quite frequently have recourse
to examples of language in his writings, but his examination of these was always per-
functory in the extreme because, for him, they were never the object of examination
themselves; they were there only as illustrations of something else.
What can such a scholar possibly have to say to pragmaticians? The answer, as
this paper hopes to show, is quite a lot. In fact, Goffman’s work had many of the same
concerns as pragmatics. The most complete overlap is the concern for language-use in
situated contexts. However, Goffman’s approach to this subject-matter was from the
opposite direction to that of pragmatics. Historically, pragmatics started with linguistic
meanings and then added on the situations of their use. Goffman started with situational
meanings and then added on the language. A central question of pragmatics, as a
branch of linguistics, might be expressed as: what can we learn about language and
communication from the study of its cultural, social and cognitive aspects? One question
for Goffman, as a sociologist, was: what can we learn about culture and society from
the study of its communicative aspects?
However, this was not his main question. For one thing, he was interested in com-
munication only in as far as it is part of something bigger. His purview was all aspects of
people’s behaviour when they are together. The word he used to denote all these aspects
was ‘interaction’, and it was ‘interaction’ in a very wide sense (see Section 4 below).
Secondly, the focus of Goffman’s work gave ‘culture and society’ a relatively
peripheral place. The recurring argument and invariant practice of his whole oeuvre
was the study of interaction in its own right. For him, the central question was simply:
what can we learn about interaction itself? The oft-professed thrust of his work was to
explore what he called the ‘traffic-rules of interaction’ and move towards ‘a sociology of
occasions’. In the introduction to one collection of his essays (1967: 1–3), he describes
the objectives of this enterprise as twofold. One is the identification and description
of interactional units. The other is the exploration of the ‘behavioural’ and “expressive
80 Jim O’Driscoll
order prevailing within and between these units”. The first of these is also a major
objective of conversation analysis. But Goffman, unlike CA, limits himself neither to
empirical data nor to conversation nor, when asking what properties interactants must
possess in order for them to interact in the first place, to sociolinguistic properties. The
second objective which bears a close relation to the ethnography of communication,
except that Goffman’s exploration does not refer to a cultural order in the first instance
but to an (autonomous) interactional one.
Of course, interaction is the place where people and society meet. Therefore, in
learning about interaction, we inevitably learn something about people and something
about that aspect of culture and society which is constituted of interaction. He allows
for this second outcome in the introduction to one of his books (1963b: 3–4) and
argues its benefits in his final work (1983). However, in the preface to another (1971: xi),
he pokes gentle fun at the notion that interactive data is merely a source of evidence for
social structure. His focus is fixed on interaction because, for him, the social aspect
of people and the interactional aspects of society are not merely instantiated in
interaction — they are essentially a product of it. While Descartes told us the essence
of a human being is in thought, Sartre that is in mere existence, and the modern reli-
gion of the body tells us ‘you are what you eat’, for Goffman, we are what we project.
To be a given kind of person, then, is not merely to possess the required attributes,
but also to sustain the standards of conduct and appearance that one’s social grouping
attaches thereto.” (1959: 81).
In this paper, I first identify the two central tenets, the twin foundations upon which
Goffman constructed his explorations of the nature of interaction (Sections 2 and 3).
Then, in Section 4, I attempt an outline of his investigative framework through a typology
of the concepts he employed, his building materials, as it were. (For the sake of brevity,
I make no attempt to trace the development of his thought. This is Goffman ‘in the
round’.) Finally, in Section 5, I discuss his significance for pragmatics. This paper is
chiefly intended for pragmaticians largely unfamiliar with Goffmans’s work. At the
same time, it is inevitably a personal interpretation and in that respect I hope it will
also be of interest to those who already are.
2. The primacy of the situation
Goffman’s arguments in favour of interaction as a locus of research in the realm of human

activity take the form of claiming its autonomy with regard to other possible loci. On the
one hand, he claims the independence of interaction from the ‘macro’ concerns of social
roles and social structure because “whatever social role the individual plays during
Erving Goffman 81
a conversational encounter, he will in addition have to fulfil the role of interactant”

(1967: 116). Thus interactants’ roles are analytically separate from social roles.
But it is not just that the roles which people play and the ground rules by which
they play them are a part of social organisation. These rules, he notes, “can regulate
dealings when those who participate share hardly any additional organization at all”
(1971: xiii). Moreover, the properties and structure of instances of spoken interaction
involve more than merely the intersection of properties of speech and social attributes
of the interactants. The whole is greater than the sum of its parts. You can’t study
either (properties of speech or social correlation) without getting drawn into aspects
of situation. He points out that when people study the impact of the social attributes
that interactants bring along with them, they are not studying the significance of the
attributes themselves but rather the value placed on them in the situation (1964: 134).
“Face-to-face interaction has its own regulations” (1964: 136).
On the other hand, he espouses the value, and to some degree the pre-eminence,
of the study of interaction for the ‘micro’ concerns of interpersonal relationships and
psychological reality. With regard to the former, he points out that interactants express
evaluation of participants in interaction because they express a view of the situation,
and not the other way around (1967: 5). Similarly, people’s rational objectives regarding
the actions or feelings of other people are pursued by influencing their definition of
the situation (1959: 15).
With regard to the latter, he repeatedly recommends that psychological phenomena
be studied from the vantage point of interaction, working inwards rather than out-
wards from the presumed deep recesses of the individual. He argues, for instance, that
psychological traits such as self-deception can be seen as occurring “when two differ-
ent roles, performer and audience, come to be compressed in the same individual”
(1959: 86–87). He suggests (1967: 136) that through the study and understanding of
how a person becomes alienated from interaction “we can learn something about the
way in which he can become alienated from things that take much more of his time”
such as “his occupational career, his political involvements, his family membership”
because “there will be a sense in which these wider matters consist in recurrent occa-
sions of…. interaction”. Thus Goffman sees interaction as providing a window not only
outwards onto social realities but also inwards onto psychological ones, and in one
paper, Mental Symptoms and Public Order (1967: 137–148), he makes a direct plea that
psychotic phenomena be studied from this vantage point.
In fact, Goffman regarded much psychiatric practice with not a little scorn and it
was sometimes the target of his sarcasm. In this same paper, he remarks of psychia-
trists that “few professions … have so well been able to institutionalize, to sell on the
social market, their own fantasies of what they were engaged in doing” (1967: 139).
For Goffman, mental illness is a social creation (see e.g. 1971: 335–390) and therefore
82 Jim O’Driscoll
his preference is not to study the putative psychological causes of offensive behaviour
but rather to study the “improprieties themselves” (1967: 138 — see also 1963b: 3–12).
This involves examining “the general rule of conduct of which the offensive behavior is
an infraction”, then proceeding to identify the set of rules of which it is part “and at the
same time try to get a glimpse of the social circle or group that sustains the rules and
is offended by the infraction of any one of them” (1967: 140). Thus he places offensive
behaviour in the context of codes of conduct and the values held by social groups.
So far in this account of Goffman’s reifying of interaction as a worthy field of
study, there has been much that pragmaticians can easily approve of, or even cheer
along with. After all, interaction is meat and drink to pragmatics; even those scholars
who eschew its more sociological aspects (see, for example, many of the entries in the
handbook by Horn & Ward 2004) appreciate interaction, or at least a consideration of
it, as a source of data. What may be less easy to swallow (and perhaps just plain inedible
to the latter group) is the distinctly small portion of this field he allots to language.
He repeatedly insists that spoken communication is only one aspect of face-to-face
interaction, so that any ‘rules of conduct’ we might identify refer not to communication
in the first instance but rather to behaviour in the presence of others.
His argument goes beyond the obvious one of subordination, that on the one
hand plenty of interaction takes place without any talk at all and on the other hand
speech nearly always takes place within situations of interaction, so that these situa-
tions are “the natural home of speech” (1964: 135). It is also that even when speech
does occur, it is more often than not in situations where “a conversation is not really
the context of the utterance; a physically elaborated, nonlinguistic undertaking is,
one in which nonlinguistic events may have the floor” (1981: 141). This is true both
of brief service contacts and those between strangers and also of extended pieces of
interaction such as mother-child paediatric consultations, mechanics working on a
car or a game of bridge, where utterances are only meaningful as part of “a presumed
common interest in effectively pursuing the [non-linguistic] activity at hand”
(1981: 143). Thus the characteristics of the encounters which take place in interac-
tion, although “often expressed through a linguistic medium”, are “not intrinsically
linguistic in character” (1964: 136).
Among other observations he makes in support of this viewpoint are: that many
encounters are begun with non-verbal moves (as in approaching the check-out at a
supermarket); that some queries can be answered verbally or gesturally (you can tell
someone the time or show them your watch); that responses are often directed not
to a verbal prompt but rather to the situation which prompted the previous words
(as when someone accepts another’s apologies). These points are made in a paper
(1981: 5–77) where he argues that even a minimal mutual greeting (hullo — hullo)
is better seen not as a response from one person to another but rather as mutual
responses to the situation.
Erving Goffman 83
Another argument is that examples of speech can be found which cannot possibly
be part of a talk or anything like that because they occur in isolation. When alone but
in the presence of strangers, we sometimes blurt out imprecations such as “oops” or
“eek” so as to be audible to all those present, so that they will realise that what they have
seen of our plight “is not something that should be taken to define us” (1981: 136). In
so doing, we try to elicit a response but not a reply, to acquire overhearers but not con-
versational partners.
For all these reasons, Goffman believes that the thing utterances are part of cannot
be called a speech event and that an analysis of any bit of spoken language needs to
refer to the situation as a whole, not just to the totality of speech within that situation,
and to take account of the role of everybody else who might be present, not just those
engaged in the spoken encounter (1981: 137). This can only be done if the primary
unit of analysis is the situation itself, and for Goffman it should remain so even if it is
the spoken bits of the situation on which an analysis wishes to focus.
3. Ritual and the sacred self
The emphasis on situation as the starting point for all analysis of interaction is essen-
tially a matter of analytical procedure. The other foundation of Goffman’s work is an
explanatory one. It addresses how interaction is possible at all and why it takes forms
that it does. His answer is contained in his conception of the interacting individual as
a vaunted being of sacred status — not, it should be hastily added, the individual of
socio-political theory but rather the ‘performed self ’.
Goffman argues the importance and nature of this self partly from logical reason-
ing and partly from observation. The reasoning aspect involves a consideration of the
nature of interaction for those taking part in it. The ‘reality’ of a situation can never be
fully known by an interactant, so appearances must be used instead and people have
to act and treat others on the basis of impressions. Given this, an interactant cannot
help conveying impressions, regardless of intention, and “others act as if he conveyed
a particular impression” (1959: 18). As a result, interaction has a moral character, the
impressions being received as claims and promises. A particular claim morally obliges
the claimant to be truthful about this claim and morally obliges others to accept it
(1959: 24). Thus, those involved in interaction often pay heed to the impression that
others receive of them and focus their efforts on creating the right impression, one
that can be sustained. Knowing that others do the same thing, and also because of the
fragility of these performed selves (see 1971: 28–61), they typically pay heed to the
impressions of others that they receive and, in the expectation that their own projec-
tions will be respected, respect them. This courtesy in turn becomes part of their own
selves to be sustained and so mutual courtesy becomes a mutual expectation.
84 Jim O’Driscoll
Following Durkheim, the name that Goffman gives to the mechanism by which
this moral salience is enacted is ritual. He defines it as “a perfunctory, conventional-
ized act through which an individual portrays his respect and regard for some object
of ultimate value to that object of ultimate value” (1971: 62). He argues and then
exemplifies at length (1971: 62–186) that the interpersonal rituals by which people
attest to their own civility and goodwill and others’ worthiness — to the “small patri-
mony of sacredness” (1971: 63) they are each thereby accorded — are ubiquitous in
everyday life, that “whenever one individual rubs up against another, he is likely to
say hullo or excuse me” (1971: 64). and that this ubiquity testifies to the omnipresent
concern which people have for each others’ selves.
Elsewhere he observes (1967: 30–31) that so essential is mutual courtesy to
interactants that reciprocal self-denial (e.g. “No. After you.”) is commonplace and
often includes “negative bargaining” which “as a form of exchange perhaps is more
common than the economist’s kind”. He also expounds (1967: 102–103) on the tact
with which people attempt to hide their knowledge that a fellow interactant is embar-
rassed. In fact, so inbuilt are the mutual expectations that interactants will uphold each
others’ projections of themselves that they become mutually dependent. Shamelessness
damages not only one’s own ‘face’ but that of others, because everybody is implicated
in everybody else’s face (1967: 14) and
the mere witnessing of an involvement offense … can cause a crime against the
interaction, the victim of the first crime himself being made a criminal. Thus … when
one individual is stricken with uneasiness, others often come down with the disease.
(1967: 125–126)
For Goffman, the moral element, and thus the moral character of the mutual expecta-
tions that result, which themselves become moral rules, are not merely an integral part
of interaction — they are also an essential prerequisite for it. It is through such moral
rules that
socially responsible people [are transformed] into people who are interactively
responsible as well … [and] …society is made safe for the little worlds sustained in
face-to-face encounters. (1967: 118).
He also argues (1981: 16–19) that what is necessary for orderly communication to take
place consists of more than the ‘system requirements’ which make effective transmission
of messages possible. There are ritual requirements as well and these “safeguard[s] not
only feelings [cf. above] but communication too” (1981: 18). He notes that features often
taken to be system requirements, such as back channel cues and relevance norms,
simultaneously serve ritual concerns as well as transmission and that therefore ritual
constraints subsume system ones. In any case, they cannot be separated out in terms
of their enactment. For example, in the request “Do you have the time?” the ‘remedy’
Erving Goffman 85
(that is, the bit that begs sufferance for a potential offense) is encoded in the request
itself (it is not separate from the bit that indicates a desire for the information). Ritual
concerns thus have explanatory power.
Thus Goffman casts the self as an omni-salient element of all interaction, and
thus something that the analyst of interaction cannot ignore. It should be added here
that the performed nature of the self, the fact that it is “something of collabora-
tive manufacture” (1959: 245), a dramatic effect, is for Goffman not something on
that account to be dismissed as a charade. The extended dramatic metaphor which
he employed in The Presentation of Self in Everyday Life (1959) has led some to regard
his view of interpersonal relations as cynical. But Goffman frequently writes warmly
of the ritual order in which the self participates, painting a picture in the final pages of
his essay On Face-work, for example (1967: 5–45), of a co-operative system whereby
individuals can shelter themselves and each other from whatever they regard as the
more unpalatable aspects of reality. In Stigma (1963a), he writes with ironic compassion of
the plight of those whose condition means they are unable to enjoy some of the com-
forts of this order (the irony being directed at ‘we normals’) and, in Asylums (1961),
with something akin to moral outrage at the Catch 22 situation of inmates of mental
institutions, who find that all the everyday props with which the self can be performed
have been stripped away from them and any of the strategies it can normally deploy
to resist or offset this loss are interpreted as confirming their abnormality (see also
1967: 67). Hence his scorn noted above. Elsewhere, he describes mental hospitals as
“hopeless storage dumps trimmed in psychiatric paper” whose effects on the incarcer-
ated are “not merely a bad deal [but] a grotesque one” (1971: 336).
Ultimately, the performed nature of the self is sacred because it is what makes life
bearable. Presumably, the conception of drama that Goffman had in mind was not the
modern one but the Ancient Greek.
4. Goffman’s working framework
For the purposes of studying interaction and its ‘traffic rules’ Goffman offers “the units
of association which make these rules possible” (1967: 143). This section attempts an
exegesis of these and the kinds of action accompanying them, in other words the main
concepts which Goffman employed. (This is not, of course, a complete list. I have chosen
what I regard as the most crucial ones, in most cases because they can be found repeatedly
in his work. For this latter reason, I do not provide citations — there would be just too
many — unless I use direct quotes.)
As a starting point, Goffman identifies three units. First, there is the situation
(sometimes called ‘social situation’), ‘an environment of mutual monitoring possibilities’
for the people present; that is, these people find themselves accessible to each other
86 Jim O’Driscoll
through their ‘naked senses’. A situation could be anything from a small room to a long
stretch of beach. The aggregate of all the people present in a situation is a gathering.
There are of course different kinds of gatherings with differing reasons for existence.
To denote the auspices under which situations occur (e.g. a funeral, a workday at the
office), Goffman uses the term occasion (sometimes, ‘social occasion’). While the limits
of a particular situation and its gathering from any one participants’ viewpoint are the
result of the ‘system requirements’ that make mutual monitoring possible, and so may
be objectively defined, occasions are social creations. It is therefore possible that different
(groups of) members of a gathering may not concur about what sort of behaviour is
appropriate to it (see 1963b: 20 for examples). How people manage to arrive at and
negotiate definitions of the situation is something that Goffman explores at length
through the concept of frame (see below).
There are two basic types of gathering, focused and unfocused. As the less elaborated,
I will take the latter first. An unfocused gathering is one in which unfocused interaction
takes place in that the members of the gathering “have not extended to each other
the status of co-participants” (1967: 145). If every member is a ‘single’ (they have not
entered the situation already in a ‘with’), this is likely to comprise the sum total of
interaction that goes on in many train carriages and urban and suburban streets. One
might wonder what such a situation has to do with interaction at all. However, it is axi-
omatic to Goffman (spelled out most extensively in 1963b: 24–88) that all members of
a gathering, however unfocused, are at least minimally and visibly alive to each other’s
presence, mutually monitoring and holding themselves in some kind of readiness for
some kind of encounter.
In an encounter (also called a ‘face engagement’), participants “jointly ratify one
another as authorized co-sustainers of a single… focus of visual and cognitive atten-
tion” (1964: 135) who are “open to each other for talk or its substitutes” (1967: 144).
The people thus involved are designated ratified participants and focused interaction
results. Some encounters comprise the whole gathering — for example, private sex.
But if it takes place within a wider gathering, whether this consists entirely of unfo-
cused interaction (for example, somebody buying a book over the counter while
everybody else in the shop quietly browses), or several other encounters (a couple on
the dance floor full of similarly engaged couples), it is an accessible engagement and
thus, within the orbit of the encounter, there will be people who, while not partaking
in the abovementioned mutual ratification, are party to what goes on. They are unrati-
fied participants, who Goffman calls bystanders. The fact that the actions of those
engaged in encounters, as well as those of singles, are partly designed to take account of
bystanders (see 1981: 85–98 for examples) shows the importance of addressing analyses
to all members of a gathering.
So far, I have avoided using examples of encounters where talk is essential. But
Goffman appears to fall into this trap when identifying two types of bystander: those
Erving Goffman 87
who are party to the doings of the encounter by accident and those who are party
by surreptitious intent. The former he calls overhearers, the latter eavesdroppers, both
implying an aural channel for their intelligences. And when he comes to study the
doings of ratified participants of an encounter, those who have ratified each other as
such, it is talk that he focuses on. Nevertheless, he shows (1981: 23–29) that any one
doing is not the same thing as a turn at talk or an utterance and therefore proposes
a move as the elementary unit of analysis in an encounter. A move is anything with a
unitary bearing on the circumstances of the gathering.
When an encounter takes place in the presence of bystanders, various kinds of
subordinate communication, organised so as not to interfere with the main focus of
attention, are possible. These are: byplay, involving a subset of ratified participants;
crossplay, which takes place between (a) ratified participant(s) and (a) bystander(s);
and sideplay, involving bystanders only. All of these may take place with no effort at
concealment, or may involve collusion. Collusion can be accomplished through concealing
the very fact of subordinate communication, or through “affecting that the words the
excolluded can’t hear are innocuous, or through allusion which, although ostensibly
meant for all, only some participants will catch” (1981: 134).
The possibilities of subordinate communication are obviously greater when more
than two people are ratified participants to an encounter because in these cases one or
more of the recipients of any one move, while still fully ratified, may be not be directly
addressed. Thus Goffman distinguishes between addressed and unaddressed recipients.
He recognises, however, that matters may not be that simple. In ‘podium events’, for
example, there are special kinds of recipient who, whether addressed (e.g. lectures,
charismatic religious services, people watching the news on TV) or unaddressed (the
audience at a stage play) have particular rights and obligations which are different to
those of fellow conversationalists (1981: 137–140).
The distinction between ratified participants and bystanders is not always straight-
forward either. When talk is subordinated to some other non-talk task at hand, in
which stretches of silence are neither exactly interludes between encounters nor pauses
within them, an open state of talk prevails in which those involved are neither exactly
ratified nor bystanders “but a peculiar condition between” (1981: 135).
Goffman sees the above categories as a way of escaping from the unanalysed
notion of ‘hearer’. He also deconstructs the notion of ‘speaker’, identifying three roles.
These are: the animator, the ‘talking machine’, the individual active in making language
noises; the author, the selector of the sentiments expressed and the words in which
they are encoded; and the principal, who is the person whose position is established
by the words spoken and committed to what the words say and thus “a person active
in some particular social identity or role” (1981: 145). This, of course, may change
rapidly, even though animator and author remain constant (as when we change hats
in committee meetings or code-switch). Canonically, these three roles are combined
88 Jim O’Driscoll
in the same person, but there are exceptions. For example, a person reciting or reading
aloud a is animator but neither author nor principal; a politician giving a speech is ani-
mator and principal but often not author; a person reporting what someone else said is
animator and author but not principal. (In Goffman’s terms, I am animator and author
of this section, but to a large extent and most of the time Goffman is its principle, the
person committed to its meanings. But only ‘to a large extent’ because my organization
of this presentation stresses some concepts and the links between them at the relative
expense of others, and only ‘most of the time’ because at some points, most obviously
here, I stand back and offer comment. I hope I make it clear at any one time whether I
am merely reporting or doing something more than that.)
The configuration of these roles at any one time, constitutes the production format
of an utterance. Goffman also uses the term participation status to denote the role or
function of any one person in a gathering, relative to an utterance or all the activity in
a situation, and participation framework to denote the pattern of participation statuses
in a gathering. He is ambiguous, however, about whether he intends these terms to
encompass productions formats or just all recipients.
So far, this exposition has presented static categories of participant. In reality,
of course, the participation status of those involved in talk is changing all the time,
not just because who is talking and who they address can change but, more crucially,
because of what is said. Part of Goffman’s ritual model assumes that anything done in
interaction always and inevitably carries implications about the character of the person
who says it, that person’s evaluation of other participants and the relationship between
them (see, for example, 1981: 19–22). To capture these potentially infinitely mutable
personal circumstances, Goffman uses the term footing, which he defines as “the align-
ment we take up to ourselves and others present as expressed in the way we manage
the production and reception of an utterance” (1981: 128). In addition, anything said
also carries an implication about what sort of thing is going on, about how the speaker
construes the experience. This is where Goffman’s use of the concept of frame comes
in. For him, a frame is “principles of organization which govern events … and our
subjective involvement in them” (1974: 10). Now, it should be obvious that framing
does not take place independently of the abovementioned concerns. When something
is said which effectively claims a definition of the situation, it simultaneously makes
claims about the participants and their relationships. Therefore, “a change in footing is
another way of talking about a change in our frame for events” (1981: 128).
Footing and framing are complicated by the phenomenon which Goffman calls
embedding. At its simplest level, this happens when utterances involve a figure, a person
“who belongs to the world that is spoken about, not the world in which the speaking
occurs” (1981: 147). A figure can be different from the principal even when we say
“I”, because the figure “I” may not embody the same social role as that we embody
as speakers committed to our words (e.g. when telling a story about what happened
Erving Goffman 89
to us). Embedding means that an utterance can involve more than one animator, as
when we begin with “I said” (or “he” or “you”). Embedded authors and principals
are also possible. Putting on a mock voice and employing an adage or proverb, and
thus an appeal to a wider authority, are other cases of embedding which, by being a
kind of reporting, involve changes in footing. Moreover, it is not just utterances which
can be embedded but participation frameworks themselves, and therefore frames, and
“within one alignment, another can be fully enclosed” (1981: 155). Among examples
he offers are: collusion where nobody is excolluded (as in the semaphored elbow nudge
or wink); the mock whisper as if bystanders were present when they aren’t; and affecting
a podium speech register within an informal conversation.
What nature divides, talk frivolously embeds, insets, and intermingles. As dramatists
can put any world on their stage, so we can enact any participation framework and
production format in our conversation. (1981: 155)
5. Goffman’s influence and significance
Goffman’s influence on pragmatics has been simultaneously great and small; far-reaching
in its extent but remarkably lacking in localised impact. This paradox has much to do
with his singular discursive and methodological practices.
At the most general level, he is a source of inspiration, an expander of conscious-
ness. All those who have read him (would) testify to the sheer excitement of the
experience. First, there are the frequent gasps of recognition when readers relate his
descriptions to their own lives. But it’s not just our own individual, inevitably limited,
lives. Another aspect of the excitement is the vast range of human behaviour that comes
within Goffman’s sights. Leaving tooth marks in a cake (1971: 47), the ritual of gas-
chamber executions (1971: 115), the functions of seconds in duels (1967: 27), bursting
bubble-gum bubbles (1967: 87), the physical movements of puppets (1971: 149), the
significance of different reading material when eating alone in public (1963b: 52) are
all grist to his mill. Goffman’s capacity to illuminate human experience, together with
his matchless prose style, has meant that, as Grimshaw (1990: 4) notes, even literary
scholars sometimes feel obliged to read him.
Another excitement is his frequently subversive approach to a phenomenon, his
angle being diametrically opposed to the default expectation. Thus, when considering
those whose behaviour is in some way offensive (1963a, 1963b), he turns it round and
looks at the offended and asks: what are the rules we live by which causes us to be
offended? Similarly, starting from the philosophical question of ‘what is reality’ (1974),
he turns it round and looks at the perceiver, asking: why and how do we perceive some
things as real (or not)? That is, “a problem having to do with the camera and not what
it is that the camera takes pictures of ” (1974: 2).
90 Jim O’Driscoll
The above features have made Goffman, for those who favour his general approach
and share his substantive interests a source of wisdom and authority and sometimes a
major inspiration for new fields of studies (see below). His name crops up in references
with impressive frequency in an impressive range of studies. The adjective ‘Goffmanian’
has taken its place in the study of language-related human activity alongside ‘Bakhtinian’,
‘Chomskyan’, ‘Foucauldian’, ‘Hallidayan’ and a handful of others.
However, when it comes to the detail, we see the other side of the coin. The
numerous frameworks and concepts he offered and explored have not been systematically
developed to any appreciable extent and have led only indirectly to theory-building.
There is no Goffmanian school of anything. In explanation, we can again appeal to his
working practices.
Many of the books through which his writings have reached a wide academic public
are collections of essays, many of which were previously published. In his prefaces or
introductions to these books, he makes no apologies for this lack of integration. Indeed,
he extols the virtues of approaching the same general issue from different vantage points
even in those books written more-or-less in one go. Even in those volumes when he
introduces, sustains and develops a single perspective and set of concepts throughout,
he does not normally pursue them further, exhibiting a cavalier disregard for them in
his subsequent writings, a promiscuous, love-‘em-and-leave-‘em attitude.
His concept of ‘face’ is a case in point. At the start of his 1955 paper On Face-Work,
he defines it as “the positive social value a person effectively claims for himself by
the line others assume he has taken during a particular contact” (1967: 5) and then
proceeds to explore it and its ramifications at some length. But The Presentation of the
Self in Everyday Life, originally published in 1956, and similarly concerned with how
interactants express definitions of the situation by managing each others’ situationally
contingent identities, makes no use whatever of the term he had so carefully nurtured
only a year before. Not so many years later, he published Stigma (1963a), which, it has
been remarked (Friedson 1983), is a kind of mirror image of this previous work, the
presentation of the discredited self. In it, he distinguishes between a person’s ‘virtual’
and ‘actual’ social identity (1963a: 12), the former circumscribed in a way which is
remarkably similar to ‘the line others assume he has taken’ (cf. above). In the same
work, he contrasts ‘ego identity’ on the one hand and personal and social identity
on the other, the former defined as a person’s “subjective sense of his own situation”
(1963a: 129) and an image that “the individual constructs … himself ” (1963a: 130),
thus bearing a close relation to ‘the positive social value a person effectively claims for
himself ’ (cf. above). It is not stretching a point, therefore, to suggest that in these two
constructs we may detect an exploration of the elements involved in his 1955 face. And
yet, once again, there is no reference to it.
As with the concepts and perspectives, so with the works themselves. As is evident
from the quality of his writing, Goffman lovingly gestated them and gave birth to them
Erving Goffman 91
but then left them to fend largely for themselves. In only one of them (1974), is there
anything more than very occasional reference to any of his previous ones. Other than
intermittently herding some of the smaller ones together into a book, he did nothing
in his professional practice to nurture their reception. In the words of one admirer:
He seems to have entirely eschewed the ritual apparatus of institutional continuity:

never editing collections of papers by his students or imitators; never contributing a
prefatory endorsement of another’s book; discouraging the growth of commentaries
and secondary studies of his work; and, until 1981 [two years before his death], never
publicly responding to published criticisms. (Williams 1988: 64)
Finally, to this untended riot of theoretical perspectives and conceptual resources

he adds an ill-mannered riot of empirical ones: fieldwork notes, more informal
ad hoc observations, introspection, etiquette manuals, biographies, fiction, press
cuttings, anything. His method is described by Schegloff (1988: 101) as ‘analytical
pointillism’. The cumulative effect of these characteristics has been that he stands
accused of lack of cumulation, of theoretical discontinuity, of failing to produce a
‘body of work’.
Goffman was not unaware of his unorthodox modus operandi and sometimes
commented on it. He regarded his theoretical constructs as working ones, necessary
because they are “implied in what has gone before and required for what is to follow”
(1959: 26), but without claim to wider validity. With reference to his conceptual
frameworks he more than once uses the metaphor of scaffolding (1959: 246; 1967: 47)
and “scaffolds, after all, are to build other things with, and should be erected with an
eye to taking them down” (1959: 246). Those with a penchant for following theories
might object that the ‘other things’ which the scaffolding was purportedly supporting
never emerge into the light of day. Goffman’s interest remains directed to his subject-
matter, not to theory-building, of whose efficacy he is lukewarm: “Better, perhaps,
different coats to clothe the children well than a single, splendid tent in which they
all shiver” (1961: xiv).
His justification for lack of empirical rigor is similar: “a loose speculative approach
to a fundamental area of conduct is better than a rigorous blindness to it” (1963b: 4).
Scholars, depending on their methodological stance, regard this aphoristic advocacy
coupled with irony as either wise caution or just plain lazy. But in another preface
(1971: xv–xviii), he makes a more explicit case for his method of “unsystematic,
naturalistic observation” and embarks on a scathing attack on the attempt to apply
the traditional, hypothesis-testing research design to this subject-matter. He claims
that this kind of research has yielded no new fields of study, no new significant con-
cepts of social activity and no new frameworks capable of accommodating more facts.
“Understanding of ordinary behavior has not accumulated; distance has”. It is this
understanding, not academic conceptualization, on which Goffman remains focused.
92 Jim O’Driscoll
(For a convincing, more elaborated justification of his working practices than Goffman
could ever be bothered to supply, see Williams 1988.)
These contradictions emerge most notably in regard to Goffman’s relation to the
practice of conversation analysis (CA). Goffman was a major influence at the inception of
CA, whose central aim has been described as to “uncover the sociolinguistic com-
petencies underlying the production and interpretation of talk in organized sequences
of social interaction” (Hutchby & Drew 1995). As such, by making great strides in
providing more detailed analytical tools for one part (the spoken part) of one kind (the
focused kind) of interaction, it may be seen as having fleshed out an area of inquiry for
which Goffman provided only the bones. However, to put it like this is to locate CA as
a Goffmanian sub-field, which may not sit well with those conversational analysts who
emphasise its empirical and methodological rigor. The practice of CA involves minute,
exhaustive study of instances of real-time spoken data and on careful operational defi-
nitions of units of talk which are supposed to be transferable to any situation of talk.
This is why Schegloff (1988) takes exception to Goffman’s exploration (1981: 124–159)
of the concept of the adjacency pair (which is one such unit) and its limits in terms
of a prototype. He assumes that Goffman is discussing the validity of the concept. But
for Goffman, ‘adjacency pair’ is just another piece of scaffolding. As for interpretation,
Schegloff argues that, despite Goffman’s professed aim to view interaction as autono-
mous and to discover its traffic rules, he never quite got over his ‘addiction’ for face and
ritual, so that he remained focused too much on the drivers. For Goffman, however,
such a focus is essential, not only because the drivers go some way to explaining the
nature of the traffic rules but also because, ultimately, they are the reason for the
existence of the rules in the first place.
Nevertheless, a great deal of work has found the concepts of participation frame-
works, framing and footing fundamental to the analysis of talk-in-interaction in a
variety of settings. See, for example, most of the contributions in the volumes edited
by Grimshaw (1990), Drew & Heritage (1992), Tannen (1993) and Sarangi & Roberts
(1999). Of particular note is the notion of ‘frame attunement’, proposed by Kendon
(1990). Although Kendon himself applies this notion to spatial orientation, the empha-
sis on process implied means that it can be applied to the minute details of ongoing
talk (e.g. Hutchby 1999).
Frame is, of course, not a concept which Goffman invented. But his exploration
of how frames are established and modified within interaction in Frame Analysis has
proved hugely influential. It has, in fact, mushroomed in all kinds of directions, many
of them by now a long way from what Goffman had in mind (or, indeed, from
interaction — see e.g. Benford & Snow 2000). This kind of tendency, whereby one of
Goffman’s concepts is picked up and then carried far away, is also evident in the field
which has become known as ‘politeness studies’. This was kick-started when Brown &
Levinson ([1978] 1987) picked up on his concept of face-work (1967: 5–45), bifurcated
Erving Goffman 93
it with reference to Durkheim’s positive and negative rites and proceeded to typologize
its instantion in great detail (much of this suggested by Goffman’s exemplifications of
supportive and remedial interchanges in his 1971 work, Relations in Public). A huge
amount of work has subsequently emerged, but, despite recent attempts to prick this
face-balloon and return to a more Goffmanian spirit (e.g. Bargiela-Chiappini 2003;
Watts 2003), in most of it Goffman is no more than a revered forerunner, a John-the-
Baptist to Brown & Levinson’s Christ. The sheer inspirational qualities of Goffman’s
writings seem to carry their own dangers, with many of his terms being bandied about
imprecisely, reduced to the status of buzzwords.
But the above comments are not to deny the potential value of such cherry-picking.
To give just two examples: Cromdal & Aronsson (2000) show how the concept of footing
can be used to illuminate the phenomenon of code-switching; O’Driscoll (2001) picks
up on the relation between face and ‘line’ (Goffman 1967: 5–45) to develop a model of
language-choice in face-to-face interaction.
These two examples above raise the question of Goffman’s cross-cultural appli-
cability. Goffman himself was not primarily interested in intercultural comparisons.
But he is always careful to qualify any generalisations regarding the instantiation of
the phenomena he observes and occasionally recommends enquiry in a cross-cultural
direction, for example, the origins of and methods for dealing with embarrassment
(1967: 101). In particular, while insisting that ritual requirements are always there, he
observes that the form they take are likely to be radically cross-culturally dependent
(1981: 17). And in advising that it is these kinds of differences we should first look to
in trying to account for improper behaviour rather than to individual personality
(1967: 124–125), he expresses what has since become a commonly cited raison d’etre for
studies in cross-cultural communication.
A great deal of work in the above-mentioned politeness tradition has had a
cross-cultural focus and, notwithstanding its conceptual flaws, much of it has proved
illuminating. Certainly, as Kendon (1988) observes, it is a promising area for future
Goffman-inspired work. The other main area which Kendon recommends is system-
atizing and elaboration of Goffman’s units of analysis, a notable attempt at which is
Levinson (1988) with respect to categories of participation framework.
It remains to be seen, however, whether such ventures will prove fruitful. What
has already proved so is Goffman’s unwavering attention to interaction itself and his
exposition of so many of the factors involved. That this has meant locating spoken
interaction within something bigger has, in this writer’s opinion, strengthened the
analyses of talk. It has also liberated it from its former limitation to ‘conversation’.
All sorts of possibilities open for the study of talk in the wider context of unfocused
interaction (the modern ubiquity of the mobile phone use being just one), although
this part of Goffman’s terrain remains notably fallow. His greatest value has been to
point up that moves, verbal or otherwise, achieve something. This emphasis on process
94 Jim O’Driscoll
is reflected in (what might be called) the ‘negotiational’ turn which pragmatic studies
of language-in-use have taken in the last few decades, which emphasis that frames in
interaction are jointly constructed and where illocution and perlocution are consid-
ered together (see Aronsson 1996). And Goffman’s simultaneous emphasis on the self,
reminding us that there are people out there who achieve these things, has had the
same effect. Identities, roles, group membership, footings, meanings and reference are
all up for grabs all the time.
Finally, it is worth re-emphasising that in all this constant shifting for position, in
his presentation of face concerns as sometimes fraught with anxiety, in his use of the
imagery of the stage, in his recognition of “a potentially infinite cycle of concealment,
discovery, false revelation and rediscovery” (1959: 20), Goffman does not present a
miserable view of interactive life. Nor does his insistence on the primacy of situation
belittle human agency. Goffman revels in the sheer complexity of human interaction.
The effect of his writings is not to point up the puny insignificance of the individual,
but rather to celebrate the amazing intricacy of our togetherness. The mutual obliga-
tions of interactants and fellow-interactants, recognised by each and modifying the
behaviour of each,
form the bridge that people build to one another, allowing them to meet for a moment
of talk in a communion of reciprocally sustained involvement. It is this spark, not the
more obvious kinds of love, that lights up the world. (1967: 116–117)
Note
The Goffman references in this list are incomplete in two respects. First, they refer only to the volumes
to which I had recourse; his papers, collected in many of these, are not recorded separately. Second,
it is not a bibliography; only works cited in the text are recorded here. For attempts at comprehensive
bibliographies, see Ditton (1980) and www.tau.ac.il/~algazi/mat/goffman.htm.
References
Aronsson, K. (1996). Collaboration in dialogues. Handbook of Pragmatics. John Benjamins.

Bargiela-Chiappini, F. Face and politeness: new (insights) for old (concepts). Journal of Pragmatics
35(10/11): 1453–1469.
Benford, R. & D. Snow (2000). Framing Processes and Social Movements: An Overview and Assessment.
Annual Review of Sociology 26: 11–39.
Brown, P. & S. Levinson (1987). Politeness: some universals in language usage. Cambridge University
Press.
Cromdal, J. & K. Aronsson (2000). Footing in bilingual play. Journal of Sociolinguistics 4(3):435–457.
Ditton, J. (1980). A bibliographical exegesis of Goffman’s sociology. In J. Ditton (ed.), The view from
Goffman. St. Martin’s Press.
Erving Goffman 95
Drew, P. & J. Heritage (eds.) (1992). Talk at work: interaction in institutional settings. Cambridge
University Press.
Drew, P. & A. Wooton (eds.) (1988). Erving Goffman: exploring the interaction order. Polity Press.
Friedson, E. (1983). Celebrating Erving Goffman. Contemporary Sociology 12(4): 359–262.
Goffman, E. (1959). The Presentation of Self in Everyday Life. Penguin.
——— (1961). Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Penguin.
——— (1963a). Stigma: Notes on the Management of Spoiled Identity. Penguin.
——— (1963b). Behavior in Public Places: Notes on the Social Organization of Gatherings. Free Press
of Glencoe.
——— (1964). The Neglected Situation. American Anthropologist 66(6), part II (Special issue): 133–136.
——— (1967). Interaction Ritual: Essays on Face-to-Face Behavior. Penguin.
——— (1971). Relations in Public: Microstudies of the Public Order. Allen Lane (Penguin).
——— (1974). Frame Analysis: An Essay on the Organization of Experience. Peregrine Books (Penguin).
——— (1981). Forms of Talk. Blackwell.
——— (1983). The Interaction Order. American Sociological Review 48: 1–17.
Grimshaw, D. (ed.) (1990). Conflict Talk: Sociolinguistic investigations of arguments in conversations.
Horn, R, & G.Ward (2004). The Handbook of Pragmatics. Blackwell.
Hutchby, I. (1999). Frame attunement and footing in the organisation of radio talk openings. Journal
of Sociolinguistics 3(1):41–63.
Hutchby, I. & P. Drew (1995). Conversation Analysis. Handbook of Pragmatics (Manual).
Kendon, A. (1988). Goffman’s approach to face-to-face interaction. In P. Drew & A. Wooton (eds.):
14–40.
——— (1990). Conducting Interaction: Patterns of Behaviour in Focused Encounters. Cambridge
University Press.
Levinson, S. (1988). Putting Linguistics on a proper footing. In P. Drew & A. Wooton (eds.): 161–227.
O’driscoll, J. (2001). A face model of language choice. Multilingua 20(3): 245–268.
Sarangi, S. & C. Roberts (eds.) (1999). Talk, work and institutional order: Discourse in medical, mediation
and management settings. Mouton de Gruyter.
Schegloff, E. (1988). Goffman and the analysis of conversation. In P. Drew & A. Wooton (eds.): 89–135.
Tannen, D. (ed.) (1993). Framing in discourse. Oxford University Press.
Watts, R. (2003). Politeness. Cambridge University Press.
Williams, R. (1988). Understanding Goffman’s methods. In P. Drew & A. Wooton (eds.): 64–88.
Interactional linguistics
Jan Lindström
1. Background
In the dawn of the twenty-first century, interactional linguistics was recognized

as a new, internationally emerging direction in the field of linguistics. While most
adherents of this direction are heavily influenced by the methodology of conversa-
tion analysis (CA), interactional linguists have backgrounds in and orientations to a
diversity of traditions, including discourse analysis, (interactional) sociolinguistics,
anthropological linguistics, (discourse) functional linguistics, construction grammar,
and grammaticalization theory. It is thus fair to say that interactional linguistics is
interdisciplinary within linguistics but also connected to other sciences, in the first
place sociology (Selting & Couper-Kuhlen 2001). The unifying perspective is to
describe linguistic structures and meanings as they serve social goals in naturally
occurring spoken, in a broad sense, conversational language, viz. ‘talk-in-interaction’.
In this perspective, linguistic structures are seen as resulting from the practical needs
of (repeated) interaction(s) as well as giving form to (particular) interaction(s), thus
providing a trajectory of an on-going interaction for the speakers. From the point of
view of pragmatics, research in interactional linguistics contributes to an empirically
based understanding of language use and the dependency of linguistic form on social
action, and vice versa.
Overarching topics within interactional linguistics comprise the syntactic and
sequential organization of turns and utterances, their prosodic features, and the
lexico-semantics of interactionally central functional elements of language. Many of
the topics were projected in the first published conversation analytic, micro-sociological
works in the 1970s and early 1980s (e.g. Sacks, Schegloff & Jefferson 1974; Schegloff,
Jefferson & Sacks 1977; Goodwin 1981; Pomerantz 1984). In these studies several
points of connection between patterns of language use and structures of social action
were observed, including the design of a turn, the adjacency pair framework and the
formal reflections of preference organization; moreover, the need of developing a ‘syntax-
for-conversation’ with the help of linguists was expressed (very explicitly in Schegloff
1979). Meanwhile, linguistic research in discourse and pragmatics began to focus on
interactive facets in the structure and style of speech as well as on the role of intonation
as a possible unit-delimiter (Tannen 1984; Chafe & Danielewicz 1987). Cross-linguistic
and cultural dimensions were connected to this line of research especially in the direc-
tions called interactional sociolinguistics and anthropological linguistics (Gumperz
1982; Schieffelin & Ochs 1986). As these currents continued to attract the interest
of researchers with their specialization in linguistics, the need of a framework for
a systematic and empirical study of spoken, interactive language and its structure
became evident. Such an orientation can also be seen as a reaction against the dominant
formal, non-empirical and written language biased directions in linguistics (see Linell
2005). Accumulating evidence for motivated relations between the linguistic form and
interactional function spurred a view of grammar as a dynamic, locally adjustable
resource of communication rather than as a self-contained, static and abstract system
(Hopper & Thompson 1984; Hopper 1998). From these orientations developed a
tradition called discourse functional linguistics which has offered a theoretically relevant
framework for many CA-inspired linguists especially in the USA.
Research at the interface between linguistics and conversation analysis has gained
the label interactional linguistics in Europe but also in a more global perspective.
Important for the international establishment of this linguistic direction was the pub-
lication of the volumes Interaction and Grammar (Ochs, Schegloff & Thompson 1996),
Studies in Interactional Linguistics (Selting & Couper-Kuhlen 2001), The Language of
Turn and Sequence (Ford, Fox & Thompson 2002a), and Syntax and Lexis in Conversa-
tion (Hakulinen & Selting 2005). There is a generation of researchers developing this
direction in their own languages, thus also enabling further cross-linguistic studies of
universals and differences involving the regularities of talk-in-interaction; pioneer-
ing studies concern English, German, Dutch, Japanese, Finnish, Danish, Swedish and
Norwegian, but the number of languages studied is constantly increasing.
2. Points of departure
The basic discovery procedure in interactional linguistics relies on the radically induc-
tive methodological tools developed within conversation analysis. The starting point
is an audio or (preferably) a video recording of naturally occurring interaction, for
example an informal get-together among friends, which for a better analytic access
is carefully transcribed. The research objects and categories are reconstructed from
collections of sequences that represent instances of a recurrent interactional or lin-
guistic phenomenon. It is analytically important that phenomena are not studied in
isolated utterances but in the original prior and subsequent sequential context around
the utterance or action of interest. Of course, there are instances which can be properly
understood only when the whole preceding interaction or the comprehensive activity
is taken into account (Linell 2005), which calls for a qualitative rather than a quantitative
98 Jan Lindström
approach to the data. As in CA, the analysis departs from a participant perspective and
establishes the relevance of structures and categories as they can be seen to be made
relevant by the interactants themselves in the data.
Interactional linguistics builds on the same assumption as CA, namely, that
ordinary conversation is an ordered, structurally organized phenomenon, and that the
structures of language on different levels are subordinated, moulded or influenced by
the general normative aspects of social interaction. Basic interactive constraints on the
speakers’ linguistic choices are turn-taking organization, sequence organization, prefer-
ence organization and repair organization. A special sequential constraint is temporal;
that is, talk-in-interaction takes place in real time. Thus, utterances are not planned to
their end before the speakers set out producing them. The production is incremental
in nature and in this respect also influenced by the conduct of the recipient to whom
the current speaker orients, possibly changing the course of a current utterance/turn
as a response to recipient reactions (cf. Goodwin 1981). The production and interpre-
tation of talk-in-interaction is thus situated in time and in a surrounding physical and
participant space.
Turns in talk-in-interaction are made of turn constructional units (TCUs), which
are defined in terms of action and linguistic structure, including syntax and prosody;
at the completion of such a unit, turn transition becomes relevant (Sacks, Schegloff &
Jefferson 1974). Studying turn organization and turn allocation is thus a matter of
linguistic form and the speakers’ linguistic competence to a high degree. The syntactic
design of an utterance is also dependent on its position in an interactional sequence
and the type of action carried out (Schegloff 1996). Responses are often produced
in a characteristic manner, usually not in full form but in an ‘elliptical’ fashion;
however, how ellipsis should be treated in grammatical terms is a largely unsolved
problem. Furthermore, responses that are socially dispreferred, viz. generally negative
or rejecting, differ formally from unproblematic responses; the dispreferred ones tend
to be delayed, contain characteristic discourse markers and are followed by accounts
(Pomerantz 1984). Finally, the question of grammaticality can be seen in a new light
when repair is considered. So called false starts, self- and other corrections do not nec-
essarily follow from the speaker’s grammatical mistakes but are prompted by problems
of hearing or understanding (Schegloff, Jefferson & Sacks 1977). What is produced in
syntactic terms may in many cases be fully grammatical up to the point where repair is
initiated; moreover, repair itself is carried out in an orderly fashion within or between
utterances/turns (Schegloff 1979, 1996).
Given the above, the research questions in interactional linguistics focus on the
relation between language structure and the social organization of talk-in-interaction.
Basically, there are two sorts of orientations: (i) the research can start from a particu-
lar linguistic form and explore its association with interactional function(s) or (ii) it
can start from a particular interactional function and then specify which linguistic
form(s) typically realize that function (Selting & Couper-Kuhlen 2001). These orienta-
tions give different results, but both types of results are relevant. The first approach
enables a thorough account of diverse potential uses of an element of language
(e.g. a certain discourse marker) but not of other resources that could alternatively
realize the interactional task in question. The second approach gives a rich descrip-
tion of the linguistic resources that are used in order to carry out a certain action
or practice (e.g. declining an offer) but it does not account for potential uses of the
observed linguistic elements outside of the focused contexts (Hakulinen & Selting
2005). In practice, the two approaches complement each other for a fuller linguistic
and sequential description of the regularities of talk-in-interaction.
3. Topics
Syntactic units were identified as central resources for the construction of turns in
ground breaking CA studies. Hence, it is no wonder that syntax in conversation has
attracted the interest of interactional linguists. Many studies have been concerned
with the syntactic shaping of turn constructional units: what makes them identifi-
able as completed units and orientable in turn transition (Selting 1996, 2000, Tanaka
1999; Steensig 2001); how can they be expanded with increments or parentheses (Auer
1996; Ford, Fox & Thompson 2002b, Couper-Kuhlen & Ono 2007); how does word
order correlate with interactional function (Auer 2005; J. Lindström 2006); what are
the structural regularities in repair segments (Fox & Jasperson 1995); what is formally
characteristic of different request and response formats (A. Lindström 1999)? This
vein of research has made important contributions towards a new understanding of
syntax and grammar: It is not sufficient to account for the internal structure of utter-
ances (‘sentences’), since the internal structure is relative to the external structure of
the utterance, viz. the sequential and actional context. Syntactic structure provides a
resource of organizing an interactional task, e.g. in a story-telling sequence, a repair
sequence, or an adjancency pair sequence, and its resulting shape is a reflection of the
contingencies involved in the task (see Helasvuo 2001). In other words, syntax cannot
be fully understood without an account of its interactional inhabitat in a turn.
In close association with conversational syntax, more specific aspects of interac-
tional meaning and lexis in conversation are studied. Most attention has been directed
to interactionally loaded ‘form words’, that is, various discourse markers. This line of
research is directly connected to earlier investigations in a more general pragmatic
perspective (cf. Östman 1981; Schiffrin 1987). Studies of response tokens of the type
yes and no have given new insights into their discourse functions that go beyond sim-
ply confirming or negating a prior utterance (A. Lindström 1999; Sorjonen 2001); one
characteristic function is their use as ‘continuers’ and as tokens of feedback. There are
100 Jan Lindström
also interesting cross-linguistic similarities and differences in the use and repertoire of
response tokens. Another class of interactionally central form words is sentence con-
nectors. For example, conjunctions and adverbs have uses in talk-in-interaction that
do not naturally surface in written language. Conjunctions can signal affiliation or dis-
affiliation with the prior speaker, continuity or discontinuity of a topic, relatedness or
unrelatedness to the prior turn (Günthner 1996; Mazeland & Huiskes 2005). In addition
to responsive and connective elements, reference practices, and particularly the use of
pronouns, have been studied in an interactional perspective (Laury 1997). It is fair to
assume that in talk-in-interaction, which occurs ‘here and now’, reference practices
may show other kinds of complexity and dynamicity than in written exposition.
A third major subfield within interactional linguistics deals with prosody in con-
versation. This line of research has contributed to the defining characteristics of TCUs
which are not only syntactic but also prosodic (and pragmatic) units. Are there regular
features of prosody that can project or mark a turn transition place? Results in this area
are yet somewhat conflicting but it seems clear that prosodic cues are important for
the delimitation of units of action, not only marking closure but also new beginnings
(Auer 1996; Ford & Thompson 1996; Selting 1996). To take another example of the
communicative import of sound patterns, studies of lexical prosody have shown that
the interactional function of an expression relies not only on the lexical choice itself
but also on the phonetic design. Response tokens can thus have affiliating or non-
affiliating interpretations depending on the phonetic delivery (A. Lindström 1999);
moreover, their phonetic design may indicate if a response token was a complete
response or if there is more to come. Yet another field of interest is prosodic stylization,
a feature that may have been overlooked as paralinguistic and thus not really linguistic
(Ogden, Hakulinen & Tainio 2004). By stylization the speaker typically communicates
modal aspects such as whether an action is to be interpreted as problematic or not, as
newsworthy or not. All in all, prosody in conversation is a challenging field in many
respects, not least because there is a great lack of professional phonetic basic research
on conversational speech in practically all languages (Kelly & Local 1989 is one of the
first openings in this field). Another challenge is that large scale generalizations are
difficult to make because prosodic patterns may vary considerably within dialects of
one language, let alone between different languages.
4. Possibilities and challenges
Interactional linguistics is a rapidly growing field with intriguing prospects in many

directions of linguistic inquiry. Some approaches imply a progress towards more for-
malistic, although empirically based, accounts of grammar(s) for spoken, dialogic
language (Du Bois 2003; Linell 2004). However, this endeavor may involve some tension
between the terms and methods of CA on the one hand and traditional linguistics on
the other hand. Practitioners of a CA-inspired linguistics must come to terms with
how to draw generalisations that are necessary for the description of the language, not
only for the description of particular instances of interactions. While plain counting
can never satisfy a methodologically informed interactional linguist, the analytic work
may still have to be complemented with quantification and computer corpus studies at
least to some degree. The same tension is related to acoustic measurements of sound
patterns; these are probably needed in order to be able to objectively show that a
phenomenon really is there and not only in the analyst’s mind.
The biggest challenge for interactional linguistics, however, is to enable and
establish a better understanding between socio-pragmatically oriented linguistic
directions and traditional, even formal linguistics. One of the tasks of interactional
linguistics is to explore the relevance of grammatical categories, e.g. the subject or the
object, for the orientations that the speakers are making in talk-in-interaction. If rel-
evant, exact linguistic terms may provide an analytic access to interactional linguistic
phenomena that could be too bluntly expressed by the (deliberately) vernacular terms
used in CA (e.g. beginning, continuer, cut-off). A dialogue with grammarians, for
example practioners of construction grammar, may provide interactional linguistics
with tools for making (careful) formalistic accounts of certain regular phenomena of
talk-in-interaction. Furthermore, contributions to and from typological linguistics
are of relevance since interactional linguistics is gradually evolving in a truly interna-
tional and cross-linguistic direction.
References
Auer, P. (1996). On the prosody and syntax of turn-continuations. In E. Couper-Kuhlen & M. Selting
(eds.), Prosody in conversation: 57–100. Cambridge University Press.
——— (2005). Projection in interaction and projection in grammar. Text 25: 7–36.
Chafe, W. & J. Danielewicz (1987). Properties of spoken and written language. In R. Horowitz &
S.J. Samuels (eds.), Comprehending oral and written language: 83–113. Academic Press.
Couper-Kuhlen, E. & C.E. Ford (2004). Sound patterns in interaction. Benjamins.
Couper-Kuhlen, E. & T. Ono 2007: ‘Incrementing’ in conversation. A comparison of practices in
English, German and Japanese. — Pragmatics 17: 513–52.
Du Bois J. (2003). Discourse and grammar. In M. Tomasello (ed.), The new psychology of language,
vol. 2: Cognitive and functional approaches to language structure: 47–87. Lawrence Erlbaum
Associates.
Ford, C.E. & S.A. Thompson (1996). Interactional units in conversation: Syntactic, intonational,
and pragmatic resources for the management of turns. In E. Ochs et al. (eds.), Interaction and
grammar: 134–184. Cambridge University Press.
Ford, C.E., B.A. Fox & S.A. Thompson (eds.) (2002a). The language of turn and sequence. Oxford
University Press.
102 Jan Lindström
——— (2002b). Constituency and the grammar of turn increments. In C.E. Ford et al. (eds.), The
language of turn and sequence: 14–38. Oxford University Press.
Fox, B.A. & R. Jasperson (1995). A syntactic exploration of repair in English conversation. In P.W. Davis
(ed.), Alternative linguistics. Descriptive and theoretical modes: 77–134. Benjamins.
Goodwin, C. (1981). Conversational organization: Interaction between speakers and hearers. Academic
Press.
Gumperz, J.J. (1982). Discourse strategies. Cambridge University Press.
Günthner, S. (1996). From subordination to coordination? Verb-second position in German causal
and concessive constructions. Pragmatics 6: 323–370.
Hakulinen, A. & M. Selting (eds.) (2005). Syntax and lexis in conversation. Benjamins.
Helasvuo, M.-L. (2001). Syntax in the making: the emergence of syntactic units in Finnish conversation.
Benjamins.
Hopper, P.J. (1998). Emergent grammar. In M. Tomasello (ed.), The new psychology of language: Cognitive
and functional approaches to language structure: 155–175. Lawrence Erlbaum Associates.
Hopper, P.J. & S.A. Thompson (1984). The discourse basis for lexical categories in universal grammar.
Language 60: 703–751.
Kelly, J. & J. Local (1989). Doing phonology. Manchester University Press.
Laury, R. (1997). Demonstratives in interaction: The emergence of a definite article in Finnish. Benjamins.
Lindström, A. (1999). Language as social action. Grammar, prosody, and interaction in Swedish
conversation. Diss. Uppsala.
Lindström, J. (2006). Grammar in the service of interaction: Exploring turn organization in Swedish.
Research in Language and Social Interaction 39: 81–117.
Linell, P. (2004). On some principles of a dialogical grammar. In K. Aijmer (ed.), Dialogue Analysis
VIII. Understanding and misunderstanding in dialogue: 7–23. Niemeyer.
——— (2005). The written language bias in linguistics. Its nature, origins and transformations. Routledge.
Mazeland, H. & M. Huiskes (2005). Dutch ‘but’ as a sequential conjunction: Its use as a resumption
marker. In M. Selting & E. Couper-Kuhlen (eds.), Studies in interactional linguistics: 141–169.
Benjamins.
Ochs E., E.A. Schegloff & S.A. Thompson (eds.) (1996). Interaction and grammar. Cambridge
University Press.
Ogden, R., A. Hakulinen & L. Tainio (2004). Indexing ‘no news’ with stylization in Finnish. In
E. Couper-Kuhlen & C.E. Ford (eds.), Sound patterns in interaction: 299–334. Benjamins.
Östman, J.-O. (1981). You know. A discourse functional approach. Benjamins.
Pomerantz, A. (1984). Agreeing and disagreeing with assesments. In J.M. Atkinson & J. Heritage (eds.),
Structures of social action: 57–101. Cambridge University Press.
Sacks, H., E. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn-taking
for conversation. Language 50: 696–735.
Schegloff, E. (1979). The relevance of repair for syntax-for-conversation. In T. Givon (ed.), Syntax
and semantics 12. Discourse and syntax: 261–288. Academic Press.
——— (1996). Turn organization: one intersection of grammar and interaction. In E. Ochs et al. (eds.),
Interaction and grammar: 52–133. Cambridge University Press.
Schegloff, E., G. Jefferson & H. Sacks (1977). The preference for self-correction in the organization of
repair in conversation. Language 53: 361–382.
Schieffelin, B. & E. Ochs (eds.) (1986). Language socialization across cultures. Cambridge University
Press.
Schiffrin, D. (1987). Discourse markers. Cambridge University Press.
Selting, M. (1996). On the interplay of syntax and prosody in the construction of turn constructional
units and turns in conversation. Pragmatics 6: 357–388.
——— (2000). The construction of units in conversational talk. Language in Society 29: 477–517.
Selting, M. & E. Couper-Kuhlen (eds.) (2001). Studies in interactional linguistics. Benjamins.
Sorjonen, M.-L. (2001). Responding in conversation: A study of response particles in Finnish. Benjamins.
Steensig, J. (2001). Notes on turn-construction methods in Danish and Turkish conversation.
In M. Selting & E. Couper-Kuhlen (eds.), Studies in interactional linguistics: 259–286. Benjamins.
Tanaka, H. (1999). Turn-taking in Japanese conversation: A study in grammar and interaction.
Benjamins.
Tannen, D. (1984). Conversational style: Analyzing talk among friends. Ablex.
Listener response
Deng Xudong
National University of Singapore
1. Introduction
The conversational phenomenon of listener response has attracted a great deal of

attention during the past five decades from such diverse scholarly disciplines as linguis-
tics, conversation analysis, (cross-cultural) communication studies, sociolinguistics,
and experimental and social psychology. A number of terms have been used to describe
this kind of listener behaviour, including ‘signals of continued attention’ (Fries 1952),
‘recognition’ (Rosenfeld 1966, 1967), ‘concurrent feedback’ (Krauss & Weinheimer
1966), ‘accompaniment signals’ (Kendon 1967), ‘listener responses’ (Dittmann &
Llewellyn 1967, 1968; Bavelas, Coates, & Johnson 2002), ‘assent terms’ (Schegloff 1968;
Leet-Pellegrini 1980), ‘back channels’ or ‘backchannel responses’ (Yngve 1970; Duncan
1972, 1973; Duncan & Niederehe 1974; Duncan & Fiske 1977, 1985; Ward & Tsukahara,
2000; Cutrone 2005; Li 2006), ‘encourager’ (Edelsky 1981), ‘continuers’ (Schegloff 1982),
‘limited feedback’ (Kraut, Lewis, & Swezey 1982), ‘responsive listener cues’ (Miller,
Lechner, & Rugs 1985), ‘minimal responses’ (Fishman 1978; DeFrancisco 1991; Bennett &
Jarvis 1991), ‘reactive tokens’ (Clancy et al. 1996; Young & Lee 2004), ‘acknowledg-
ment tokens’ (Jefferson 1984, 1983/1993; Drummond & Hopper 1993a, 1993c), ‘receipt
tokens’ (Heritage 1984), ‘response tokens’ (Gardner 2001, 2007) and ‘project makers’
(Bangerter, Clark, & Katz 2004). For the sake of generality and easy comprehensibility,
the term ‘listener response’ is used in this paper to refer to the verbal and nonverbal
behaviours of a listener in response to his or her co-conversationalist’s talk.
The wide range of interests in the study of listener responses is not surprising as
it is such a ubiquitous phenomenon in our daily conversations, but nonetheless one
very seldom attended to by conversational participants in a conscious way. Such a
conversational phenomenon, trivial as it may seem, carries a large burden of interac-
tional meaning, plays a significant role in the emergent construction of a conversation,
and contributes to the interpersonal and social context on an ongoing basis (McCarthy
2003). Its study becomes even more important in cross-cultural communication set-
tings. It is widely documented in cross-cultural communication literature that people
from different cultures may follow a different conversation norm and can have different
conversational styles, of which the use of listener responses in conversations is an
important part (Tannen 1984). Interactional sociolinguists have shown repeatedly
that when speakers with diverse conversational practices interact with each other,
communicative difficulties or even miscommunication are most likely to occur,

which can further result in negative cultural evaluations and stereotyping (Gumperz
1982; Tannen 1990).
2. Approaches to the study of listener responses
In the history of research on listener response, two major strands of study can be
identified which are representative of two different approaches to its study. One is the
lumping approach, which treats as a single category or class a group of different forms
of listener responses. The other is the splitting approach which is taken mainly by
ethnomethodological conversation analysts. This approach analyses one or more discrete
listener responses in their sequential context and tries to demonstrate that each token
of listener response can perform distinctive interactional functions.
2.1 The lumping approach

The lumping approach to the study of listener response is widely used in the fields of
linguistics, sociolinguistics (most prominently, language and gender studies), cross-
cultural communication, and experimental and social psychology. This approach
probably starts with Fries’ (1952) study of American English sentence patterns, though
earlier more general allusion to listener response could arguably have been made by
Bales (1950) in one of his major groups of interaction process categories called ‘Positive
Reactions’. Fries (1952), in his discussion on the classification of different kinds of
sentences, distinguished three large classes of sentences based on the “characteristic
responses that regularly followed or attended their utterance” (p. 51). One of the three
large classes is ‘those single free utterances in a series that have as responses continued
attention’ (p. 49). He found in his telephone-conversation data the following brief oral
signals of attention, listed in order of their relative frequency of occurrence: ‘yes’, ‘unh
hunh’, ‘yeah’, ‘I see’, ‘good’, ‘oh’, ‘that’s right’, ‘yes I know’, ‘oh oh’, ‘fine’, ‘so’, ‘oh my goodness’,
and ‘oh dear’ (pp. 49–50). These signals of attention, according to Fries, “do not interfere
with the continuous flow of the utterances of the speaker” and “simply serve to give
something of the hearer’s reaction and to signal the fact that he is listening attentively
to the speaker” (p. 50).
Earlier studies of listener response in the lumping approach come largely from
experimental and social psychology, commencing in the 1960s (e.g., Dittmann &
Llewellyn 1967, 1968; Kendon 1967; Rosenfeld 1966, 1967) and continuing through
the twenty-first century (e.g., Bennett & Jarvis 1991; Bavelas, Coates, & Johnson 2002).
The earlier studies focused mainly on two general themes: the structural characterisation
of listener response and its roles in conversation in general or more specifically in the
conversational encoding and decoding process (cf Duncan & Fiske 1985).
106 Deng Xudong
2.1.1 Structural properties of listener response

A number of studies, which were concerned with the first theme, i.e., the structural
description of listener response in the conversation, typically dealt with it in connection
with its non-verbal versions such as headnods, gaze, and smiles (e.g., Birdwhistell
1962; Kendon 1967; Dittmann & Llewellyn 1967, 1968; Brunner 1979). Others have
mainly been concerned with its non-turn status in conversation (e.g., Yngve 1970;
Duncan 1972, 1973; Duncan & Niederehe 1974; Duncan & Fiske 1977, 1985).
Kendon (1967), looking at functions of gaze direction in dyadic conversation, also
examines gaze direction in relation to the occurrence of short utterances, one of which
is ‘accompaniment signals’. Based on where the listener places his/her signals in relation
to the speaker’s behaviour (particularly the gaze-direction in the course of producing
these signals), Kendon identifies two main classes of accompaniment signals: attention
signal and point granting or assenting signal. With respect to the attention signals such
as “yes quite”, “surely”, and “I see”, the listener appears to do no more than signal to the
speaker that s/he is attending, and following what is being said (Kendon 1967/1990: 73).
As regards point granting or assenting signals, the speaker structures his/her argu-
ment in such a way that “his continuing is dependent upon his interlocutor consenting
to, or specifically granting him, the points that he is making” (Kendon 1967/1990: 73).
Kendon found that the two types of accompaniment signals are clearly distinguished
from each other in terms of gaze-direction. In almost every case, when the listener
produces an attention signal, s/he continues to look steadily at the speaker. But when
s/he produces a point-granting or assenting signal, s/he looks away.
Dittmann & Llewellyn’s study (1967) of listener response is aimed at providing
evidence that spoken language is decoded by the listener in word groups called phonemic
clauses, which are defined as clauses having a certain rhythmic structure involving a
primary stress and terminal juncture (p. 345). Their data are simulated telephone con-
versations between college students. They found that vocal listener responses occur
almost exclusively in the junctures surrounding speakers’ phonemic clauses. In their
subsequent study, Dittmann & Llewellyn (1968) examined the use of vocal listener
responses in relation to a visual one, the head nod. They found that the head nod
tended to occur in the same location (i.e., at the end of phonemic clauses) with vocal
listener responses (or to be more exact, head nods slightly preceding vocal responses),
and that the two types of signals, visual and vocal, co-occur more often than would
be expected by chance. They also conducted a content analysis of these co-occurring
visual and vocal listener responses and found that they usually serve an interpersonal
function: the wish of the listener to speak or the wish of the speaker for feedback.
Dittmann & Llewellyn’s discussion of the positions of listener responses in con-
versation is illuminating as it begins to take into consideration the interactional
environment in which listener responses occur. Their examination of the location of
listener responses with reference to phonemic clauses, being based upon a theory of
sentence structures (Trager & Smith 1957), was later taken up and improved by scholars
in other fields such as cross-cultural communication studies in the investigation of the
positions of listener response in the conversational context. For example, Clancy et al.
(1996) examined the positions of listener responses with reference to what they call
the ‘Complex Transition Relevance Places’ which comprise grammatical, pragmatic and
intonational completion points (see also Ford & Thompson 1996; cf Sacks, Schegloff &
Jefferson 1974).
Yngve (1970) introduced the most common term currently in use for conversa-
tional listener responses — ‘back-channel communication’. As the term ‘back channel’
implies, Yngve assumed the existence of two channels in a conversation: the main
channel and the back channel. He argued that short utterances such as ‘uh huh’ and
‘okay’ took place in the back channel, whereas the activities of the primary teller took
the main channel. Yngve (1970: 568) defines back channels as a device which allows
the person who does not hold the turn to send “short messages such as ‘yes’ and ‘uh
huh’ ” without forcing his/her partner to relinquish the turn.
Duncan and his associates carried out a series of studies in which they attempted
to ‘discover some of the structural properties of dyadic conversations among speakers of
American English’ (Duncan 1973: 29). In these studies, back channel responses were
discussed with respect to the turn taking mechanism they aimed to develop. The key
concern of this line of work is the identification of the signals with their constituent
cues and of the rules governing turn taking in conversations, specifically in two-person
face-to-face conversations.
Duncan’s early work (Duncan 1972, 1973, 1974; Duncan & Niederehe 1974) was
based on two dyadic interactions. One was an interview between a 40-year old male
therapist and a female client in her early 20s, while the other was a conversation
between the male therapist who participated in the first interview, and a second male
therapist, also 40 years old. On the basis of the exploratory data, Duncan (1972) for-
mulated a system of signals and rules, called the turn-taking mechanism (later referred
to as ‘the turn system’ [Duncan & Fiske 1977, 1985]), that he hypothesised to govern
turn taking in dyadic face-to-face interaction.
The turn-taking mechanism proposed by Duncan (1972, 1973; Duncan & Niederehe
1974) consists of three basic signals, one of which is the back channel communication,
or ‘auditor back-channel responses’ (Duncan & Fiske 1985), defined as the display of
at least one of its six constituent forms as below (ibid: 58–59):
1. M-hm: a group of readily identified verbalisations, such as “m-hm,” “yeah,” “right,”

“yes quite,” “surely,” “I see,” and “that’s true.” These back channels may be used
singly, in combination, or repeated in groups, such as “yeah, yeah.”
2. Sentence completion: the auditor completing the speaker’s sentence.
3. Request for clarification: the auditor making brief requests for clarification.
108 Deng Xudong
4. Brief restatement: the auditor giving a brief restatement of an immediately preceding

thought expressed by the speaker.
5. Head nods and shakes.
6. Smiles.
In addition to the three basic signals described above, Duncan & Fiske (1977, 1985)
also proposed a set of rules specifying the appropriate action by one participant in
response to a signal by the other (Duncan 1973: 29; Duncan & Fiske 1985: 51). The
rules proposed for auditor back-channel responses are as follows:
Auditor back channels … do not constitute speaking turns or claims of the speaking
turn. They appear to indicate continuing attentiveness and responsiveness of one
sort or another to the speaker’s message. The back channel appears to provide the
auditor with a way of actively participating in the conversation, thus facilitating
the general coordination of action by both participants within the structure of the
interaction. (Duncan & Fiske 1985: 58)
After the formulation of the turn-taking mechanism, a number of studies were done
to test its validity, but with varying results (Duncan 1972; Dittmann 1973; Duncan &
Niederehe 1974; Duncan & Fiske 1977; Beattie 1978; Wiemann & Knapp 1975;
Trimboli & Walker 1984; Walker & Trimboli 1984; for review, see Wilson, Wiemann &
Zimmerman 1984).
With specific reference to their discussion of back channels, Duncan & his associates’
studies are important as they are among the first to provide a systematic classification
of back channels in respect of their non-turn status in the turn system. But the major
problem in this classification is the difficulty in determining the status of an utterance
in conversation as a turn or a non-turn. In the work of Duncan & his associates, no
explicit definition of the concept of a turn is provided, though Wilson, Wiemann, &
Zimmerman (1984) manage to deduce it as “a continuous period during which a par-
ticipant has the undisputed right to speak” (p. 164). In fact, Duncan & his associates
themselves find problematic the distinction between back-channels and turns
In summary, early structural descriptions of listener response in the lumping
approach have focused on the relationship between verbal and non-verbal listener
responses, the location where they occur in conversation with reference to phonemic
clauses, and their non-turn status with respect to the turn system in general. These
studies, especially those by Duncan & his associates, provide a systematic base for the
classification of listener responses, which has served as analytic framework in many
subsequent studies (e.g., Marche & Peterson 1993).
2.1.2 Roles and functions of listener responses in conversation

In addition to the structural description of listener responses, the study of their roles
and functions in conversation, or in interpersonal communication more generally, is
another frequent theme in the study of listener responses in the fields of experimen-
tal and social psychology and communication studies. This theme may have its origin
in the study of the effects of feedback on human communication, a more general
term which covers virtually all kinds of responses (visual or vocal) to a speaker ranging
from headnods and smiling to interrupting and question-asking (Leavitt & Mueller
1951; Argyle, Lalljee, & Cook 1968; Rosenfeld 1966, 1967; Krauss et al. 1977; Davis &
Perkowitz 1979; Kraut, Lewis, & Swezey 1982; Vogal, Keane, & Conger 1988; Bennett &
Jarvis 1991; Li 2006). For example, Leavitt & Mueller (1951) study the effects of feed-
back giving and withholding on the transmission of information from one person
to another. In their study, feedback is a much wider notion than, for instance, the
concept of Duncan & his associates’ back channels (Duncan & Fiske 1977, 1985), and
includes any form of verbal or expressive language such as visibility of conversation
partners, question asking and interrupting (pp. 402–403). They found that when an
instructor is giving a description of some patterns, s/he can present the information
more accurately and the students can understand it better with the availability of
feedback on the part of the students.
One of the earliest studies of the interactional roles and functions of listener
responses with a more restricted sense (i.e., the Duncanian sense) may be the one
by Rosenfeld (1966, 1967). Rosenfeld (1967) studies the reciprocation of approval-
related responses, which he divides into either approving responses or disapproving
responses. Approving responses include nonverbal reactions (such as smile, positive
head nod, and gesticulation) and verbal acknowledgment or ‘recognition’. ‘Recognitions’,
more or less equivalent to the notion of listener responses or back channels, refer to
‘a broad class of usually brief responses to an utterance which indicate attentiveness
to the other person, but apparently add no other information to the conversation’, and
include ‘verbal reinforcers’ such as ‘mm-hmm,’ and attentional responses such as ‘I see,’
‘no kidding?’ and ‘really?’ (p. 105). Disapproving responses comprise frown, nega-
tive head nod, and verbal disparagement. Rosenfeld asked the interviewer to present
questions to 48 ninth-grade students from 14 to 16 years. He found that the student
interviewee emits a significantly higher percentage of smiles and positive head nods
in response to the approving interviewers than to the disapproving or nonresponsive
interviewers. But the recognitions by the students occurred too infrequently to permit
analysis. Rosenfeld suggests that one can use appropriate tactics to gain or maintain
another individual’s approval.
Krauss et al. (1977) study the role of audible and visible back-channel responses in
interpersonal communication. Specifically, they want to find out whether visible back-
channel responses have the same functions as audible ones and how the speaker would
respond with only the presence of visible back-channel responses. Krauss et al. used
simulated telephone conversations as their data but manipulated the responses of the lis-
tener into four different types: (1) visible back-channels only; (2) audible back-channels
110 Deng Xudong
delayed for one second; (3) both visible and audible back-channels; (4) visible back-
channels plus delayed audible ones. They found that when audible back-channels are
delayed, the speaker has greater difficulty in encoding the information (i.e., uses more
words in the encoding process). But if visible back-channels are available, this diffi-
culty decreases even though the audible ones are delayed. Krauss et al. thus conclude
that visible back-channel responses are functionally equivalent to vocal back-channel
responses, at least in situations in which the vocal responses are unavailable (p. 527).
The 1980s also saw a number of studies concerned with the role and functions
of listener responses in conversation. Kraut, Lewis, & Swezey (1982) examine how
feedback influences the production and reception of information in interaction.
They asked 76 university students to watch a movie and then summarise it to one
or two listeners. The listeners provide the speaker with varying amounts of feed-
back: (1) unrestricted feedback, where the listener can provide any form of feedback
including asking questions and interrupting; (2) limited feedback, where the listener
can only provide brief listener responses such as “h-hmm,” “I see,” “huh?” “who?” or
“really!”; (3) no feedback, where the speaker receives no feedback from the listener
at all. Kraut, Lewis, & Swezey found that the more feedback speakers received from a
partner, the more comprehensible their summaries were to the listeners. In addition,
feedback individuated communication; that is, the listener who provided the feedback
understood the movie better than the listener who listened to the same conversation
but provided no feedback. They conclude that feedback plays an important role in the
coordination of conversation.
The study on the function of listener responses in conversation continues to the
1990s and the 2000s. Bennett & Jarvis (1991) studied whether withholding minimal
responses would affect how listeners perceive an interaction. They asked eighty under-
graduate students to listen to a recording of conversation between two people in two
different versions, one being the original recording and the other being an edited
version in which minimal responses such as “mmm,” “hmm,” “yeah,” and “aha” had
been deleted. The students were then asked to make judgments about various aspects
of the interaction (as provided in a questionnaire). Their results showed that minimal
responses have two main functions, that is, to denote agreement and to suggest a
context of informality. Li (2006) examined the relationship between the frequency of
backchannel responses (any verbal or nonverbal act occurring during the conversation
in a non-intrusive manner) and listener recall scores in inter- and intra-cultural con-
versations. With forty simulated physician-patient interactions between 40 Chinese
and 40 Canadians, Li found that while backchannel responses facilitated content com-
munication in the two intra-cultural groups (i.e., Chinese physician/Chinese patient
and Canadian physician/Canadian patient), these responses seem to serve as misleading
feedback and cause miscommunication in the two inter-cultural groups (Chinese
physician/Canadian patient and Canadian physician/Chinese patient).
In summary, the above studies have in various ways examined the roles and functions
of listener responses in conversation and most of these studies have shown, in one way
or other, that listener responses play an important role in successful communication
(except probably in intercultural communication contexts).
2.2 The splitting approach

The other approach to the study of listener responses is what can be termed the
splitting approach. This approach, mainly taken by conversation analysts, has examined
some discrete listener response tokens in their sequential contexts. Unlike the lump-
ing approach, it is not concerned with the relationship between the occurrences of
listener response tokens and the external variables. Rather, it is mainly concerned with
their occurrences with respect to the operation of the turn-taking organization (Sacks,
Schegloff, & Jefferson 1974). In the conversation analytic literature, several listener
response tokens have received an intensive systematic study. Each of them is found
to be distinctive in terms of its placement and roles in the sequential environment
and its consequences for subsequent turns. These tokens include: ‘yeah,’ ‘uh huh,’ and
‘mm hm’ (Schegloff 1982; Jefferson 1983/1993, 1984; Drummond & Hopper 1993a,
1993b, 1993c); ‘oh’ (Heritage 1984); assessments such as ‘wow’ and ‘good’ (Goodwin
1986); high-grade assessments such as ‘lovely’ and ‘brilliant’ (Antaki 2002; Antaki,
Houtkoop-Steenstra, & Rapley 2000); ‘okay’ (Beach 1993, 1995; Pillet-Shore, 2003);
‘mm’ (Gardner 1997a, 1997b, 1998, 2001) and ‘right’ (Gardner 2007).
Schegloff (1982), for example, studied listener response tokens like ‘uh huh’. He
advises that discourse be studied as an interactional achievement, which is partially
shaped by its turn-taking organisation. He observed that vocalisations like ‘uh huh’ in
their sequential context can have two main and related usages: ‘a usage as continuer
and a usage to pass an opportunity to initiate repair’ (p. 88). According to Schegloff,
the most common usage of vocalisations such as ‘uh huh’ is as continuer, the function
of which is to encourage the previous speaker to continue talking, and by producing
‘uh huh’, etc, the producer passes the opportunity to take a fuller turn at talk (p. 81).
Listener response tokens like ‘uh huh’ occur frequently in an environment when an
extended unit of talk by another is underway.
Jefferson (1983/1993, 1984) examined listener response forms like ‘mm hm’ and
‘yeah,’ which she labelled ‘acknowledgment tokens’. She found that the two tokens are
functionally and sequentially different from each other in that the former (i.e., ‘mm
hm’) indicates more of a passive recipiency and the latter (i.e., ‘yeah’) is more related
to full speakership incipiency. Passive recipiency, according to Jefferson (1984), means
that “its user is proposing that his co-participant is still in the midst of some course of
talk, and shall go on talking” (p. 200). This is consistent with the above observation by
Schegloff (1982) that listener response tokens like ‘uh huh’ and ‘mm hm’ serve mainly
112 Deng Xudong
as continuers. Drummond & Hopper (1993a, 1993b) later took up the theme, attempt-
ing to reassess in a quantitative mode Jefferson’s claim about speakership incipiency
of ‘yeah’ and passive recipiency of ‘mm hm’. Their studies received a critical response
from Zimmerman (1993), although their findings reaffirmed Jefferson’s claim.
Also on the basis of the analysis of sequential organisation, Goodwin (1986)
distinguished between continuers (e.g., ‘uh huh’) and assessments (like ‘wow’ and
‘good’). He claimed that while continuers serve as bridges between turn-construction
units and tend to overlap with the primary speaker’s next unit, assessments normally
end in the current unit and do not overlap with the speaker’s next unit. He also noted
that the recipient speaker of the continuer orients to it by continuing to speak whereas
that of the assessment may see the telling as an ending.
Heritage (1984) did an extensive study of one listener response token ‘oh’, which
he called a ‘change-of-state token’. By this he means that an ‘oh’ is used to “propose that
its producer has undergone some kind of change in his or her locally current state of
knowledge, information, orientation or awareness” (p. 299). ‘Oh’ was found to occur in
a variety of conversational sequences such as in informings, question-elicited inform-
ings, counterinformings, other-initiated repair, understanding checks and in displays
of understanding. In all these environments, its generic change-of-state usage holds
though with slight variations of meaning in each sequence. Additionally, Heritage
observed that the sequential role of ‘oh’ is “essentially backward looking and scarcely
ever continuative” (p. 336). That is, ‘oh’ by itself does not invite or promote any contin-
uation of an informing from the primary speaker. This is accounted for by the fact that
‘oh’ most regularly occurs either in conjunction with additional turn components such
as assessments or requests for further information, or in company with some further
talk from the ‘oh’ producer. The change-of-state usage of ‘oh’ and its non-continuative
sequential role makes it distinctive from such receipt tokens as ‘yes’ and ‘mm hm’. The
latter, unlike ‘oh’, avoid or defer treating prior talk as informative and are regularly used
as continuers in extended tellings (pp. 305–306).
Building upon previous studies on ‘okay’ usages in phone call openings and clos-
ings (e.g., Schegloff 1968, 1979, 1986; Schegloff & Sacks 1973), in service-encounters
(e.g., Merritt 1984), and in simulated family interactions (e.g., Condon 1986), Beach
(1993) goes on to examine “the interactional work giving rise to ‘Okay’ usages, par-
ticipants’ orientations to them, and their consequences for subsequent talk” (p. 328).
He found that ‘okay’ has a dual character, by which he means that it is used at or near
transition/opportunity spaces as responsive to the current speaker’s prior talk on the
one hand and displaying ‘state of readiness’ for movements to next-positioned matters
on the other.
Gardner (1997a, 1997b, 1998) studies the token ‘mm’, which he calls as a ‘weak
acknowledging’ token. He observes that this token is very common in Australian and
British English, but not in American English. Unlike previous researchers of listener
response tokens, who focus mainly on the examination of the sequential placement
and the speakership incipiency of the tokens, Gardner also looks at their prosodic
shape and pause environment; these latter two he regards as being crucial in distin-
guishing between different uses of ‘mm’ and other related tokens such as ‘mm hm’ and
‘yeah’. According to Gardner, when ‘mm’ takes on a falling intonation contour, it is
used as a weak acknowledging token, which is its most common or its canonical use.
When ‘mm’ has a fall-rising intonation contour, it is used as a continuer-like object. It can
also be used as a weak assessment token, but then it takes on the rise-falling contour.
Gardner (1997a) also found that ‘mm’ displays a speakership incipiency about midway
between ‘uh huh’ and ‘mm hm’ on the one hand, which are rarely followed by same-
speaker talk, and ‘yeah’ on the other, which is frequently followed by the same-speaker
talk. Moreover, ‘mm’, with whatever prosodic shape, was found to be topically disalli-
gning in that “its speaker has nothing further to say on the topic of the talk to which it
is oriented, so either the prior speaker continues, on or off topic, or the Mm producer
continues, but off topic” (Gardner 1997a: 133).
Most recently, Gardner (2007) turns his attention to the token ‘right’. In addi-
tion to the two common uses of ‘right’ (one denotes ‘correct’, ‘true’ and is similar to
the epistemic confirmation token That’s right; and the other is somewhat a truncated
version of Alright and serves, like the token Okay, as a boundary marker or change-of-
activity marker such as a warrant to signal the end of a conversation), he focuses more
particularly on the uses of ‘right’ as what he calls an ‘epistemic dependency marker’. By
this, he means that the ‘right’ is used by its producer not only to acknowledge a turn
but also to claim understanding of the relevance of this turn in relation to something
prior (p. 336). According to Gardner (2007), ‘right’ as an epistemic dependency token
is found frequently in information- and advice-giving sequences in Australian and
British English.
In summary, the conversation analytic approach to the study of listener responses
examines discrete listener response tokens in its sequential context. Studies in this
approach found that each of these tokens is distinctive from the other and each is
a separate token on its own. Different listener response tokens can occur in different
sequential environments, have different roles and functions in these environments
and/or project different trajectories for subsequent talk.
3. Classification of listener response
As studies of listener responses in the conversation analytic perspective seek mainly

to establish the uniqueness of individual tokens, the various classification systems of
listener responses derive mostly from studies in the lumping approach. A number of
these systems have been based upon Duncan & his associates’ classification of ‘auditor
114 Deng Xudong
back channel responses’ (e.g., Marche & Peterson 1993; for Duncan & his associates’
classification, see Duncan & Fiske 1985). Their classification distinguishes auditor
back channel responses from other listening and speaking behaviours on the basis
of the former’s non-turn status. That is, auditor back channel responses, according to
Duncan & Fiske (1985), do not constitute a turn.
In addition to the use of turn, a few other researchers have sought to look into
other criteria in the identification and classification of listener responses. These crite-
ria include, most notably, the concept of ‘floor’ (e.g., Hayashi & Hayashi 1991) and the
form and/or sequential organisation of listener responses (Tottie 1991; Clancy et al.
1996). For example, Clancy et al. (1996), which classified listener responses (or Reactive
Tokens in their terminology) based partly on their form and partly on their sequential
function, distinguished five types of Reactive Tokens:
1. Backchannel: a non-lexical vocalic form, which serves as a ‘continuer’, display of

interest, or claim of understanding. Examples of backchannels include ‘hm,’ ‘huh,’
‘oh,’ ‘mhm,’ and ‘uh huh’.
2. Reactive Expression: a short non-floor-taking lexical phrase or word” produced
by the non-primary speaker. Typical Reactive expressions include, for example,
‘yeah,’ ‘sure,’ ‘exactly,’ ‘shit,’ and assessments such as ‘good’ and ‘wow’.
3. Collaborative Finish: an utterance produced by the non-primary speaker to finish
a previous speaker’s utterance.
4. Repetition: the non-primary speaker repeats a portion of the speech of the pri-
mary speaker.
5. Resumptive Opener. Resumptive Openers (RO) share the same form as Backchan-
nels (BC), i.e., they are both non-lexical vocalisations. But RO is distinguished
from BC in that RO is followed by a full turn whereas BC stands alone. There is
normally a short pause between a RO and the full turn following it. In the sequen-
tial context, RO serves to acknowledge the prior turn and commence a new turn.
Thus, ROs are parallel to Jefferson’s acknowledgment tokens which signal speaker-
ship incipiency (Jefferson 1984, 1983/1993; Drummond & Hopper 1993a).
Clancy et al.’s classification system of listener responses provides a promising link

between the aggregate and the individualised studies of the conversational phenomena.
4. Cross-cultural study of listener response
Though scholarly interest in the conversational phenomenon of listener response dates

back to the fifties, cross-cultural studies of this phenomenon are more recent, originating
probably from the comparative studies of the conversational conventions of Japanese
and Anglo-American native speakers of English (e.g., Lebra 1976; Hinds 1978; Clancy
1982; Mizutani 1982; LoCastro 1987; Maynard 1986, 1987, 1989, 1990, 1997; White
1989; Yamada 1989; White 1997; Hayashi 1988, 1990, 1991; Hayashi & Hayashi 1991;
Clancy et al. 1996; Ward & Tsukahara 2000; Cutrone 2005). This is probably because
Japanese pay exceptional attention to the interactional and affective aspects in conver-
sations. In fact, unlike most other languages such as English and Chinese, the Japa-
nese language has a special term to describe the use of short listener responses, called
aizuchi. The literal meaning of the term aizuchi refers to “the joint hammering of and
swordsmiths’ pounding on a sword’s blade or the hammer two workers use to drive a
large wooden stake into the ground” (Hirokawa 1995: 40). This was later extended to
mean an act of indicating agreement with another party or that of going along with
the other party so as not to cause a conflict (ibid). According to Clancy et al. (1996),
the use of aizuchi is a matter of everyday discussion among Japanese people and it is
common to comment on other people’s over- or under-use of aizuchi.
The comparative studies of the use of listener responses by Japanese and Americans
have shown that the two groups of people differ greatly in their use of this conversa-
tional strategy in terms of the frequency of its use, its placement in the conversation,
and its functions in the conversational context. With respect to the frequency of the
use of listener responses, previous studies have most consistently shown that Japanese
speakers produce listener responses more frequently than do Anglo-Americans (Hinds
1978; LoCastro 1987; Maynard 1986, 1987, 1989, 1990, 1997; White 1986, 1989; Yamada
1989; Hirokawa 1995; Cutrone 2005; but for exception, see Clancy et al. 1996). The
more frequent use of listener responses by Japanese was evidenced in a variety of con-
versational contexts: (1) in casual conversations (Maynard 1986, 1987, 1989, 1990,
1997; White 1989; Hirokawa 1995), formal conversations (Hinds 1978) and in business
negotiations (White 1997); (2) in dyadic (Maynard 1986, 1987, 1989, 1990, 1997; White
1989; Hirokawa 1995; Cutrone 2005) and multiparty conversations (Yamada 1989;
LoCastro 1987); (3) in intracultural and intercultural conversations (Maynard 1986,
1989, 1997; White 1989; Hirokawa 1995; Yamada 1989).
With regard to the placement of listener responses in the conversational context,
a number of studies have also observed differences between Japanese and American
speakers (Maynard 1986, 1989, 1990, 1997; Hirokawa 1995; Clancy et al. 1996).
For example, Maynard (1997) found that Japanese listeners frequently send back
channels during a brief speaker pause which she terms ‘Pause-bounded Phrasal
Unit’ (p. 45) whereas American listeners give back channels at the grammatically
significant breaks, i.e., at the end of the clause and at the sentence-final position
(somewhat equivalent to the notion of Sacks, Schegloff, & Jefferson’s Transition
Relevance Place or TRP). In other words, Japanese listeners may tend to produce
back channels at non-TRP while their American counterparts may be more likely
to utter back channels at TRP.
116 Deng Xudong
A few studies have also attempted to identify some specific functions of listener
responses in the conversational context and to determine whether differences exist
between Japanese and American speakers. Maynard (1986, 1989, 1997), for example,
specified six categories of the functions of back channels, including: (1) “continuer”
(Schegloff 1982); (2) display of understanding of content; (3) support toward the
speaker’s judgment; (4) agreement; (5) strong emotional response; (6) minor addition,
correction, or request for information. She observes that while back channels sent by
both Japanese and Americans function in all six categories, in Japanese the display of
understanding of content is more often used as a kind of moral support for the primary
speaker, whereas in American English the function as “continuer” is the more primary
function (Maynard 1997: 46).
Several studies have also examined the types of listener responses different cultural
groups of speakers tend to use. In this respect, Japanese speakers have consistently
been found to be more likely to produce semantically empty listener responses (like
‘mmhm’ and ‘uh huh’) than Americans, who, in turn, prefer to use contentful ones
more (like ‘yeah’) (White 1989; Clancy et al. 1996; Hayashi & Hayashi 1991; White
1997; but see Maynard 1997). For example, Sheida White (1989) compared the use by
Japanese and American speakers of the five most frequently-occurring backchannels
in her conversational data: ‘mmhm,’ ‘yeah,’ ‘uh-huh,’ ‘oh,’ and ‘hmm’. She found that
Japanese speakers use all of the five backchannels more than Americans do in intrac-
ultural interactions, with the exception of ‘yeah’.
The cross-cultural differences in the use of listener responses between Americans
and Japanese have largely been accounted for by different cultural values of the two
groups. Japanese people are said to be more concerned for harmony and cooperation
(Lebra 1976; LoCastro 1987; Clancy et al. 1996; White 1989), more sensitive toward
“self-contextualization” (i.e., “the ongoing process of continually defining oneself in
relation to one’s interactional environment”) (Maynard 1997: 54), and more other-
oriented and listening-oriented (Hirokawa 1995). Americans, on the other hand, are
more concerned for self expression and frankness (LoCastro 1987; White 1989) and
more self-involving and speaking-oriented (Hirokawa 1995).
The comparative study of the use of listener responses by cultural groups other
than Japanese is a relatively recent development and does not seem to receive much
attention. A few exceptions include the study of Finnish conversation behaviours
(e.g., Lehtonen & Sajavaara 1985), the use of feedback between French native speakers
and American advanced learners of French (Wieland 1991), British and American
use of backchannels (Tottie 1991), Swedish and (American and British) English use of
acknowledgment tokens (Beach & Lindstrom 1992), Maori and Pakeha’s use of verbal
feedback in New Zealand English (Stubbe 1998), Chinese use of listener responses
(e.g., Tao & Thompson 1991; Günthner 1993; Clancy et al. 1996), and Korean and
American use of reactive tokens (Young & Lee 2004).
Lehtonen & Sajavaara (1985), in their discussion of the Finnish listening behaviours,
reported that vocalisations and verbal backchannel signals are less frequently used in
Finnish than in Central European languages or in British and American Englishes
(pp. 195–196). According to them, verbal backchannel signals are used mostly in
informal and enthusiastic discourse, but their too frequent use is considered intrusive
and can even be taken as behaviour “typical of drunken people” (p. 196). They conclude
that the typical Finn is a ‘silent’ listener (p. 196). Lehtonen & Sajavaara (1985) contend
that a Finnish listener’s silence or the absence of verbal signals may be misinterpreted
as being inattentive, indifferent, sullen, or even hostile on the part of the Finnish inter-
locutor. It is also sometimes interpreted by their foreign counterparts as showing that
the Finn is feeling anxiety and would like to end the conversation. Misinterpretations
like these necessarily result in a communication breakdown and negative cultural
evaluations and stereotypes.
Wieland (1991) analysed conversations between French native speakers and
American advanced learners of French. She found that Americans use a lot of hearer
signals (e.g., ‘um hum,’ ‘uh huh,’ ‘huh,’ ‘oui’). French speakers, on the other hand, only
infrequently employ such signals. Instead, they use quite a lot of ‘minor contributions’
(viz., short phrases that are uttered during the primary speaker’s turn in reaction to
what is being said). But their use of these minor contributions is sometimes inter-
preted by American speakers as an interruption rather than as feedback.
Tottie (1991) compared the use of backchannels in British and American English
conversations. She found that American English conversations, with 16 backchannels
per minute, contain more backchannels than do British English conversations which
have only 5 backchannels per minute.
Beach & Linstrom (1992) is one of the very few studies which comparatively
examines the interactional work done by acknowledgment tokens in Swedish and
English conversations. Through detailed analysis of these tokens in their sequential
context, they found that “Swedes and Americans rely upon the same or similar inter-
actional resources, acknowledgment tokens being a prime example, while organizing
such activities as stories or topics” (pp. 36–37). Beach & Linstrom (1992) specifically
compared how speakership and recipiency are achieved through the use of acknowl-
edgment tokens in Swedish and English conversations. They observed that although
in Swedish some different forms of acknowledgment tokens are sometimes used
such as ‘eh’, similar organising principles are complied with by Swedish and English
conversations to achieve passive recipiency and exhibit incipient movements toward
speakership and topic shift/change. They conclude that the routine achievement
of Swedish talk “may not be a radically different enterprise from, for example, the
achievements comprising interaction with English-speaking cultures,” thus refuting
the commonly-held beliefs that Swedes are incapable of providing adequate feedback
and are conversationally inept (p. 37).
118 Deng Xudong
Stubbe (1998) compared the use of verbal feedback by Maori and Pakeha (i.e., people
of European descent) speakers of New Zealand English. In her study, feedback is clas-
sified into minimal responses and cooperative overlaps, with the former (i.e., minimal
responses) being further classified into neutral (such as ‘mm’) and overtly supportive
minimal responses (such as ‘oh gosh’). She found that Pakeha speakers produced more
verbal feedback in general and more minimal responses in particular than their Maori
counterparts. But she did not provide results for neutral and overtly supportive minimal
responses, which may prove to be useful in showing whether the two groups differ
in any way in the different types of minimal responses. Further, more data need to
be included than her eight dyadic conversations (with two independent variables of
ethnicity and gender) to have a more valid claim for ethnic and gender differences in
the use of verbal feedback.
As to Chinese speakers, existing studies seem to show that they are very infre-
quent users of listener responses compared with some other cultural groups such as
Japanese (Clancy et al. 1996), Americans (Tao & Thompson 1991; Clancy et al. 1996),
Germans (Günthner 1993) and Australians (Deng 2008). Clancy et al. (1996), for
example, compared the use of reactive tokens in three languages – English, Japanese,
and Chinese. They demonstrated that the three languages differ in the use of reactive
tokens in several ways. Specifically, Chinese speakers use reactive tokens less than half
as frequently as English and Japanese speakers, and they tend to use reactive tokens
which are lexically contentful more than Japanese speakers, but a little less than English
speakers. Chinese also place about the same percentage of reactive tokens at points of
grammatical completion as English speakers but a higher percentage than do Japanese
speakers. Clancy et al. (1996) suggest that Chinese listener behaviour “is part of a ‘non-
coercive cultural orientation’ that places high value on personal autonomy and avoids
putting oneself above others” (p. 382). They also propose that avoidance of backchan-
nelling by the Chinese reflects an appropriate stance of non-interference toward the
speaker and represents an interactional style which values respectful deference on the
part of Chinese interactants (pp. 382–383).
In summary, previous studies of cross-cultural differences in the use of listener
responses have clustered in the examination of their use in Japanese and American
conversations. More recently the patterns of their use by other cultural groups have
also been investigated. These studies have to a great extent provided evidence that
people from different cultural groups may use listener responses differently in terms
of their frequency of use, their placement in the conversational context and in terms of
the different types of listener responses. These studies, by comparing the pattern of use
of listener responses in different languages, have largely shown that speakers of different
cultural groups use listener responses differently, which can result in communicative
difficulties and negative cultural evaluations and stereotyping.
5. Gender-differentiated use of listener response
Listener responses are one of the most widely studied conversational phenomena which
have been claimed to show gender-related differences. Although a few studies seem to
provide evidence to the contrary (i.e., no or little gender-differentiated use of listener
responses) (e.g., Kollock, Blumstein, & Schwartz 1985; Marche & Peterson 1993), a
majority of studies have supported the claim that women use more listener responses
than men do. These studies include research in varieties of the English language such
as in British and American English (e.g., Hirschman 1973/1994; Leet-Pellegrini 1980;
Fishman 1978; Roger & Schumacher 1983; Roger & Nesshoever 1987; Tottie 1991),
New Zealand English (e.g., Hyndman 1985; Gilbert 1990; both cited in Holmes 1995),
and Indian English (e.g., Valentine 1986). A similar pattern of gender differentiated
use of listener responses has also been documented in some other languages including
Greek (Makri-Tsilipakou 1994) and Swedish (Nordenstam 1992).
Hirschman (1973/1994) is one of the first to note gender-differentiated use of listener
responses, although earlier allusions have been made in studies of sex-role behaviour
in small groups, to the effect that men were more task-oriented whereas women were
more socio-emotion-oriented in interaction and that men tended to “pro-act,” i.e., give
opinions, suggestions, and information while women tended to positively “react” to the
contributions of others such as showing solidarity and agreeing (e.g., Strodtbeck 1951;
Parsons & Bales 1955; Strodtbeck & Mann 1956; Strodtbeck, James, & Hawkins 1957;
Bennett & Cohen 1959; Heiss 1962; Borgotta & Stimson 1963; Gouran 1968). Hirschman
(1973/1994) analysed six conversations on love-related themes between four university
students, two male and two female. She divided “responses made to the speaker” into two
categories: “affirmative” (‘yeah,’ ‘ok,’ ‘mm hmm,’ ‘right,’ ‘all right’) and “other” (‘oh,’ ‘well’
in utterance-initial position) (p. 434–435). In terms of the use of affirmative responses,
she found that females had a higher frequency than males. More specifically with the use
of the token ‘mm hmm’, Hirschman found that it was a predominantly female speech
form, as it was used much more frequently by the two female speakers than by the two
males. But most of these ‘mm-hmm’s occurred in female-female interaction.
Fishman (1978) studied 52 hours of conversations between three heterosexual
couples in their homes. She found that women used what she called “attention begin-
nings” (like “this is interesting” and its variations) much more frequently than their
partners did. With respect to minimal responses such as ‘yeah,’ ‘umm,’ and ‘huh’, Fish-
man observed that whereas men and women both used minimal responses, they used
them in quite different ways. Men might give only a minimal response at the end of a
woman’s lengthy remark, while women would insert these responses throughout the
stream of men’s talk, signalling their constant attention. Fishman thus maintained that
women use minimal responses for support work, whereas the male usages of them
120 Deng Xudong
displayed lack of interest. Her conclusion was that “women are the ‘shitworkers’ of
routine conversation” (Fishman 1978: 405).
Leet-Pellegrini (1980) studied conversation between 70 pairs of unacquainted
college students. She reported that women used more ‘assent terms’ such as ‘yeah,’
‘right,’ ‘uh-huh,’ and ‘that’s true’ than men did. Some other experimental studies have
also produced the results to the effect that women use more listener responses than
men do both in same-gender conversation (e.g., Roger & Schumacher 1983) and in
mixed-gender interaction (e.g., Roger & Nesshoever 1987).
Studies on gender-related patterns in the use of listener responses have largely
been based on English data. Very few studies have been done in this respect in other
languages. Two exceptions are Nordenstam’s (1992) study of Swedish conversation and
Makri-Tsilipakou’s (1994) study of Greek conversation. Both studies have revealed that
women used more listener responses than men did, though in Nordenstam’s study, this
pattern occurred only in same-sex groups and not in married couples.
6. Future research
This paper has depicted a wide range of research traditions for the study of the
conversational phenomenon of listener responses. The scholarly interest in such a
phenomenon does not seem at all to have subdued. Studies of individual listener
responses or clusters of listener responses to examine their nuanced meanings and
functions in local sequential contexts as well as in entire conversations are continuing
(e.g., Bangerter, Clark, & Katz 2004; Gardner 2007). In addition to the expanding
number of studies of listener responses in cross-cultural contexts, more studies of
their acquisition by first and second language speakers and of their production in
various sociopragmatic contexts are also expected, in both the quantitative and the
qualitative modes (Blum-Kulka, House, & Kasper 1989). The increasing availability
of various properly transcribed conversational corpora has made possible the large
corpus study of listener responses and other conversational behaviours, which will
undoubtedly contribute to the understanding of their meanings and functions in
multiple contexts (e.g., McCarthy 2003). As new communication media such as
Instant Messaging and other online chatting programs have become increasingly
popular in this era, studies of the use of conversational behaviours like listener
responses in these relatively untapped media may also begin to emerge.
References
Argyle, M., M. Lalljee & M. Cook (1968). The effects of visibility on interaction in a dyad. Human
Relations 21(1): 3–17.
Bales, R.F. (1950). Interaction process analysis: A method for the study of small groups. Addison-Wesley.
Bangerter, A., H.H. Clark & A.R. Katz (2004). Navigating joint projects in telephone conversations.
Discourse Processes 37(1): 1–23.
Bavelas, J.B., L. Coates & T. Johnson (2002). Listener responses as a collaborative process: The role of
gaze. Journal of Communication 52(4): 566–580.
Beach, W.A. (1993). Transitional regularities for ‘casual’ “Okay” usages. Journal of Pragmatics
19: 325–352.
——— (1995). Preserving and constraining options: “Okays” and ‘official’ priorities in medical
interviews. In Morris, G. & R. Cheneil (eds.), Talk of the clinic: 259–289. Lawrence Erlbaum.
Beach, W.A., & A.K. Lindstrom (1992). Conversational universals and comparative theory: Turning to
Swedish and American acknowledgement tokens-in-interaction. Communication Theory 2: 24–49.
Beattie, G.W. (1978). Floor apportionment and gaze in conversational dyads. British Journal of Social
and Clinical Psychology 17: 7–16.
Bennett, M., & J. Jarvis (1991). The communicative function of minimal responses in everyday
conversation. Journal of Social Psychology 131(4): 519–23.
Birdwhistell, R.L. (1962). Critical moments in the psychiatric interview. In T.T. Tourlentes (ed.),
Research approaches to a psychiatric problem: 179–188. Grune & Stratton.
Blum-Kulka, S., J. House & G. Kasper (eds.) (1989). Cross-cultural pragmatics. Ablex.
Brunner, L.J. (1979). Smiles can be back channels. Journal of Personality and Social Psychology
37: 728–734.
Clancy, P.M. (1982). Written and spoken style in Japanese narratives. In D. Tannen (ed.), Spoken and
written language: 55–76. Ablex.
Clancy, P.M., S.A. Thompson, R. Suzuki & H. Tao (1996). The conversational use of reactive tokens
in English, Japanese, and Mandarin. Journal of Pragmatics 26: 355–87.
Condon, S.L. (1986). The discourse functions of OK. Semiotica 60: 73–101.
Cutrone, P. (2005). A case study examining backchannels in conversations between Japanese-British
dyads. Multilingua 2: 237–274.
Deng, X. (2008). The use of listener responses in Mandarin Chinese and Australian English
conversations. Pragmatics: The Quarterly Journal of the International Pragmatics Association
18(2): 303–328.
Dittmann, A.T. (1973). Style in conversation. Semiotica 3: 241–271.
Dittmann, A.T. & L.G. Llewellyn (1967). The phonemic clause as a unit of speech decoding. Journal
of Personality and Social Psychology 6: 341–8.
——— (1968). Relationship between vocalizations and head nods as listener responses. Journal of
Personality and Social Psychology 9: 79–84.
Drummond, K. & R. Hopper (1993a). Backchannels revisited: Acknowledgement tokens and
speakership incipiency. Research on Language and Social Interaction 26: 157–177.
——— (1993b). Some uses of yeah. Research on Language and Social Interaction 26: 203–212.
——— (1993c). Acknowledgment tokens in series. Communication Reports 6(1): 47–53.
Duncan, S. (1972). Some signals and rules for taking speaking turns in conversations. Journal of
Personality and social Psychology 23: 283–292.
——— (1973). Toward a grammar for dyadic conversation. Semiotica 9: 29–46.
——— (1974). On the structure of speaker-auditor interaction during speaking turns. Language in
Society 2: 161–180.
Duncan, S. & D. Fiske (1977). Face-to-face interaction: Research, methods, and theory. Lawrence Erlbaum.
——— (1985). Interaction structure and strategy. Cambridge: Cambridge University Press.
Duncan, S. & G. Niederehe (1974). On signalling that it’s your turn to speak. Journal of Experimental
Social Psychology 10: 234–47.
122 Deng Xudong
Edelsky, C. (1981). Who’s got the floor? Language in Society 10: 383–421.
Fishman, P. (1978). Interaction: The work women do. Social Problems 24: 397–406.
Fries, C.G. (1952). The Structure of English. Harcourt Brace.
Gardner, R. (1997a). The conversation object mm: A weak and variable acknowledging token.
——— (1997b). The listener and minimal responses in conversational interaction. Prospect 12(2): 12–32.
——— (1998). Between speaking and listening: The vocalisation of understanding. Applied Linguistics
19(2): 204–224.
——— (2001). When listeners talk: Response tokens and listener stance. Benjamins.
——— (2007). The Right connections: Acknowledging epistemic progression in talk. Language in
Society 36: 319–341.
Goodwin, C. (1986). Between and within: Alternative sequential treatments of continuers and
assessments. Human Studies 9: 205–217.
Gumperz, J.J. (1982). Discourse strategies. Cambridge University Press.
Günthner, S. (1993). German-Chinese interactions: Differences in contextualization conventions and
resulting miscommunication. Pragmatics: The Quarterly Journal of the International Pragmatics
Association 3(3): 283–304.
Hayashi, R. (1988). Simultaneous talk: From the perspective of floor management of English and
Japanese speakers. World Englishes 7(3): 269–288.
Hayashi, R. (1990). Rhythmicity, sequence and syncrony of English and Japanese face-to-face
conversation. Language Sciences 12(2/3): 155–195.
——— (1991). Floor structure of English and Japanese conversation. Journal of Pragmatics 16(1): 1–30.
Hayashi, T. & R. Hayashi (1991). Back channel or main channel: A cognitive approach based on floor
and speech acts. Pragmatics and Language Learning Monograph Series 2: 119–138
Heritage, J. (1984). A change-of-state token and aspects of its sequential placement. In M. Atkinson &
J. Heritage (eds.), Structures of social action: Studies in conversation analysis: 299–345. Cambridge
University Press.
Hinds, J. (1978). Conversational structure: An investigation based on Japanese interview discourse.
In J. Hinds & I. Howard (eds.), Problems in Japanese syntax and semantics: 79–121. Kaitakusha.
Hirokawa, K. (1995). The expressions of culture in the conversational styles of Japanese and Americans.
Unpublished Ph.D. Thesis, The University of Michigan.
Hirschman, L. (1973). Female-male differences in conversational interaction. Paper given at annual
meeting of the Linguistic Society of America, San Diego, California. (Later published in Language
in Society 23: 427–42. [1994]).
Holmes, J. (1995). Women, men and politeness. Longman.
Jefferson, G. (1983). Two explorations of the organization of overlapping talk in conversation: Notes
on some orderlinesses of overlap onset. Tilburg Papers in Language and Literature No. 28.
——— (1984). Notes on a systematic deployment of the ackowledgement tokens ‘yeah’ and ‘Mm hm’.
Papers in Linguistics 17: 197–216.
——— (1993). Caveat speaker: Preliminary notes on recipient topic-shift implicature. Research on
Language and Social Interaction 26: 1–30. (Original work published 1983).
Kendon, A. (1967). Some functions of gaze direction in social interaction. Acta Psychologica 26: 22–63.
(Reprinted in A. Kendon (1990). Conducting interaction: Patterns of behavior in focused encounters:
51–8. Cambridge University Press.).
Kollock, P., P. Blumstein & P. Schwartz (1985). Sex and power in interaction: Conversational privileges
and duties. American Sociological Review 50: 34–46.
Krauss, R.M. & S. Weinheimer (1966). Concurrent feedback, confirmation, and the encoding of
referents in verbal interaction. Journal of Personality and Social Psychology 4: 342–346.
Kraut, R.E., S.H. Lewis & L.W. Swezey (1982). Listener responsiveness and the coordination of
conversation. Journal of Personality and Social Psychology 43(4): 718–731.
Leavitt, H.J. & R.A.H. Mueller (1951). Some effects of feedback on communication. Human Relations
4: 401–410.
Lebra, T.S. (1976). Japanese patterns of behavior. University of Hawaii Press.
Leet-Pellegrini, H.M. (1980). Conversational dominance as a function of gender and expertise. In
H. Giles, W.P. Robinson & P.M. Smith (eds.), Language: Social psychological perspectives: 97–104.
Pergamon Press.
Lehtonen, J. & K. Sajavaara (1985). The silent Finn. In D. Tannen & M. Saville-Troike (eds.), Perspective
on silence: 193–201. Ablex.
Li, H.Z. (2006). Backchannel responses as misleading feedback in intercultural discourse. Journal of
Intercultural Communication Research 35(2): 99–116.
Locastro, V. (1987). Aizuchi: A Japanese conversational routine. In L.E. Smith (ed.), Discourse across
cultures (pp. 101–113). New York: Prentice Hall.
Makri-Tsilipakou, M. (1994). Interruption revisited: Affiliative vs. disaffiliative intervention. Journal
of Pragmatics 21: 401–426.
Marche, T.A. & C. Peterson (1993). On the gender fifferential use of listener responsiveness. Sex Roles
29(11/12): 795–816.
Maynard, S.K. (1986). On back-channel behavior in Japanese and English casual conversation.
Linguistics 24(6/286): 1079–1108.
——— (1987). Interactional functions of a nonverbal sign: Head movement in Japanese dyadic
conversation. Journal of Pragmatics 11: 589–606.
——— (1989). Japanese conversation: Self-contextualization through structure and interactional
management. Ablex.
——— (1990). Conversation management in contrast: Listener response in Japanese and American
English. Journal of Pragmatics 14: 397–412.
——— (1997). Analyzing interactional management in native/non-native English conversation: A case
of listener response. IRAL XXXV(1): 37–60.
McCarthy, M. (2003). Talking back: “Small” interactional response tokens in everyday conversation.
Merritt, M. (1984). On the use of ‘O.K.’ in service encounters. In J. Baugh & J. Scherzer (eds.), Language
in use: 139–147. Prentice-Hall.
Miller, L.C., R.E. Lechner & D. Rugs (1985). Development of conversational responsiveness: Preschoolers’
use of responsive listener cues and relevant comments. Developmental Psychology 21: 473–480.
Mizutani, N. (1982). The listener’s response in Japanese conversation. Sociolinguistics Newsletter
13(1): 33–38.
Nordenstam, K. (1992). Male and female conversational style. International Journal of the Sociology
of Language 94: 75–98.
Pillet-Shore, D. (2003). Doing “Okay”: On the multiple metrics of an assessment. Research on Language
and Social Interaction 36(3): 285–319.
Roger, D. & W. Nesshoever (1987). Individual differences in dyadic conversational strategies: A further
study. British Journal of Social Psychology 26: 247–255.
Roger, D.B. & A. Schumacher (1983). Effects of individual differences on dyadic conversational
strategies. Journal of Personality and Social Psychology 45(3): 700–705.
Rosenfeld, H.M. (1966). Approval-seeking and approval-inducing functions of verbal and nonverbal
responses in the dyad. Journal of Personality and Social Psychology 6: 597–605.
Rosenfeld, H.M. (1967). Nonverbal reciprocation of approval: An experimental analyses. Journal of
Experimental Social Psychology 3: 102–111.
124 Deng Xudong
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn-taking
for conversation. Language 50: 696–735.
Schegloff, E.A. (1968). Sequencing in conversational openings. American Anthropologist 70(6):
1075–95.
Schegloff, E.A. (1979). Identification and recognition in telephone conversation openings. In G. Psathas
(ed.), Everyday language: Studies in ethnomethodology: 23–78. Irvington.
Schegloff, E.A. (1982). Discourse as an interactional achievement: Some uses of ‘uh huh’ and other things
that come between sentences. In D. Tannen (ed.), Analyzing discourse: Text and talk (Georgetown
University Round Table on Language and Linguistics 1981): 71–93. Georgetown University Press.
Stubbe, M. (1998). Are you listening? Cultural influences on the use of supportive verbal feedback in
conversation. Journal of Pragmatics 29: 257–289.
Tao, H. & S.A. Thompson (1991). English backchannels in Mandarin conversations: a case study of
superstratum pragmatic ‘interference.’ Journal of Pragmatics 16(3): 209–233.
Tannen, D. (1984). Conversational style: Analyzing talk among friends. Ablex.
——— (1990). You just don’t understand: Women and men in conversation. William Morrow.
Tottie, G. (1991). Conversational style in British and American English: The case of backchannels.
In K. Aijmer & B. Altenberg (eds.), English corpus linguistics: Studies in honour of Jan Svartvik:
254–335. Longman.
Trimboli, C. & M.B. Walker (1984). Switching pauses in cooperative and competitive interactions.
Journal of Experimental Social Psychology 20: 297–311.
Valentine, T. (1986). Language and power: Cross-sex communicative strategies in Hindi and Indian
English. Economic and Political Weekly 21: 75–87.
Vogel, J., S.P. Keane & J.C. Conger (1988). A content analysis of the conversational behavior of accepted
and rejected children. Journal of Psychopathology and Behavioral Assessment 10: 49–64.
Walker, M.B. & C. Trimboli (1984). The role of nonverbal signals in co-ordinating speaking turns.
Journal of Language and Social Psychology 3(4): 257–272.
Ward, N., & Tsukahara, W. (2000). Prosodic features which cue back-channel responses in English
and Japanese. Journal of Pragmatics 32: 1177–1207.
Wiemann, J.M. & M.L. Knapp (1975). Turn-taking in conversations. Journal of Communication 25: 75–92.
White, R. (1997). Back channelling, repair, pausing, and private speech. Applied Linguistics
18(3): 314–343.
White, S. (1989). Backchannels across cultures: A study of Americans and Japanese. Language in
Society 18: 59–76.
Wieland, M. (1991). Turn-taking structure as a source of misunderstanding in French-American
cross-cultural conversation. Pragmatics and Language Learning Monograph Series 2: 101–118.
Wilson, T.P., J.M. Wiemann & D.H. Zimmerman (1984). Models of turn taking in conversational
interaction. Journal of Language and Social Psychology 3(4): 159–183.
Yamada, H. (1989). American and Japanese topic management strategies in business meetings.
Unpublished Ph.D. Thesis, Georgetown University.
Yngve, V.H. (1970). On getting a word in edgewise. Papers from the Sixth Regional Meeting of the
Chicago Linguistic Society: 567–577. Chicago Linguistic Society.
Young, R.F. & J. Lee (2004). Identifying units in interaction: Reactive tokens in Korean and English
conversations. Journal of Sociolinguistics 8(3): 380–407.
Zimmerman, D.H. (1993). Acknowledgement tokens and speakership incipiency revisited. Research
on Language and Social Interaction 26: 179–194.
Participation
Jack Sidnell
University of Toronto
Whenever people talk together they establish and sustain a little world of shared
attention and involvement, a “communion of mutual engagement,” as Goffman (1957)
described it. As such participation, or better co-participation, in an activity cannot be
reduced to mere physical co-presence. Rather, participation (or non-participation) in
talk is organized moment-by-moment, incarnately in and through that talk. While
this may seem obvious, it raises a series of important questions about the organization
of human conduct. Specifically, what are the practices of speaking, listening, gesturing,
posturing and so on by which persons constitute themselves and co-present others as
participants in social interaction and participants of a particular kind? What forms
of participation are possible and how are these realized in the particular activities
with which they are associated? In this chapter I trace ideas about this fundamental
aspect of human interaction from an initial conception of the “speech circuit” through
mid-century notions of “channel” and “phatic communion” to Goffman’s innovative
rethinking of the issue in terms of “mutual monitoring possibilities” and his trail-
blazing analysis in the late essay “Footing”. I conclude with a discussion of current
research in conversation analysis, linguistic anthropology and pragmatics that inves-
tigates the complex structures and practices of participation that characterize human
social interaction.
1. “Phatic communion” and the practices of participation
In his, “The problem of meaning in primitive languages”, Malinowski argued for a

behavioralistic theory of language in which the meaning of an utterance is connected
with the speaker or hearer’s behavior, with the “purpose of what they are doing.”
Malinowski develops his theory in a discussion of talk in task activity (e.g. fishing) but
notes that there are types of speech that do not fit this mold. Specifically, there is talk-for-
talk’s sake or, as he puts it, “the exchange of words almost as an end in itself ” (1923: 312):
“a flow of language, purposeless expressions of preference or aversion, accounts of
irrelevant happenings, comments on what is perfectly obvious” (1923: 314). This type
of speech which he termed “phatic communion,” posed a problem for Malinowski’s
theory in which words are understood to get their meanings from the activities in
126 Jack Sidnell
which they are used and the actions with which they are associated. The problem arose
as a result of Mailnowski’s highly simplistic understanding of “action” and “activity”
which he thought of more or less completely in terms of task. So his theory works best,
perhaps, for cases such as “a party of fisherman on a coral lagoon, spying for a shoal of
fish, trying to imprison them in an enclosure of large nets, and to drive them into
small net-bags” (1923: 310) and using single word commands to coordinate their
efforts. Although Malinowski was aware of the ways language played a role in the coor-
dination of such activities, he seems not to have grasped the fact that language is itself
an instrument of action.1 As such conversation – in which Malinowski’s “expressions of
preference or aversion” become the vehicles of actions such as praising and blaming,
complimenting and excusing, complaining and sympathizing and so on – posed a prob-
lem for his theory as did narrative. He solved the problem for conversation by arguing
that (1923: 315):
But what can be considered as situation when a number of people aimlessly gossip
together? It consists in just this atmosphere of socialbility and in the fact of the
personal communion of these people. But this is in fact achieved by speech, and the
situation in all such cases is created by the exchange of words, by the specific feelings
which form convivial gregariousness, by the give and take of utterances which make
up ordinary gossip. The whole situation consists in what happens linguistically. Each
utterance is an act serving the direct aim of binding hearer to speaker by a tie of some
social sentiment or other. Once more language appears to us in this function not as an
instrument of reflection but as a mode of action.
Malinowski suggests here and elsewhere that much of what we do when we talk is
reaffirm social relationships and contrasts such uses of language with more genuinely
“informational” communication. Malinowski used the term phatic to talk about role
of talk in reaffirming social bonds.
1. Jakobson, of course did. He disagrees with efforts to distinguish poetics from linguistics on the
grounds that the former but not the latter is concerned with “evaluation”:
This separation of the two fields from each other is based on a current but erroneous
interpretation of the contrast between the structure of poetry and other types of verbal
structure: the latter are said to be opposed by their “casual,” designless nature to the
“noncasual” purposeful character of poetic language. In point of fact, any verbal behavior
is goal-directed, but the aims are different and the conformity of the means used to the
effect aimed at is a problem that evermore preoccupies inquirers into the diverse kinds
of verbal communication. (Jakobson 1960)
Participation 127
So, in attempting to solve fundamental problems in his theory of meaning,

Malinowski actually put his finger on something vitally important that, though it may
have been noticed before him, had not been accorded any real significance. The notion
of “phatic communion” focused attention on the practices of speaking that underlie all
human interaction insofar as establishing contact between the parties to talk is clearly
prerequisite to any form of interaction.
Jakobson incorporated these ideas into his model of “speech functions”. Here
again, the notions of “contact” and of the “phatic” crop up almost as a by-product
of other considerations. The “phatic” dimension is discussed on two occasions in
Jakobson’s writings: in the first (Jakobson 1960) he is primarily concerned with the
“poetic function” (“the set toward the message”), in the second (Jakobson 1980 [1957])
with the “metalingual” function. Thus, in his discussion of the poetic function he
writes, “Language must be investigated in all of the variety of its functions. Before
discussing the poetic function, we must define its place among the other functions
of language.”
CONTEXT
(referential)
MESSAGE
(poetic)
ADDRESSER --------------------------- ADDRESSEE

(emotive) (conative)
CONTACT
(phatic)
CODE
(metalingual)
Figure 1. Schematic representation of the functions of language, adapted from Jakobson 1960
Jakobson’s (1985: 115) discussion of the phatic dimension is brief enough that it may
be reproduced here in its entirety:
There are messages primarily serving to establish, to prolong, or to discontinue

communication, to check whether the channel works (“Hello, do you hear me?”),
to attract the attention of the interlocutor or to confirm his continued attention
(“Are you listening?” or in Shakespearean diction, “Lend me your ears!” – and on
the other end of the wire “Um-hum!”). This set for contact, or in B. Malinowski’s
terms phatic function (1923) may be displayed by a profuse exchange of ritualized
128 Jack Sidnell
formulas, by entire dialogues with the mere purport of prolonging communication.

Dorothy Parker caught eloquent examples: “Well!” she said. “Well, here we are,” he
said. “Here we are,” she said, “Aren’t we?” “I should say we were,” he said, “Eeyop! Here
we are.” “Well!” she said. “Well!” he said, “well.” The endeavor to start and sustain
communication is typical of talking birds; thus the phatic function of language is the
only one they share with human beings when conversing with them. It is also the first
verbal function acquired by infants; they are prone to communication before being
able to send or receive informative communication.
Jakobson used the term “phatic” to focus in on particular aspects of the commu-
nicative situation – what he thought of, somewhat inadequately, as the “channel”
of communication. This contrasts with Malinowski for whom “phatic” is more
about the social relationships involved. Jakobson introduced another crucial mod-
ification in suggesting that uses of speech were not categorically either “phatic”,
“referential”, “meta-lingual”, “connotative” or whatever else. Rather these different
functions were combined in a single instance of speaking though their relative
importance varies.
Although we distinguish six basic aspects of language, we could, however, hardly

find verbal messages that would fulfill only one function. The diversity lies not in a
monopoly of some one of these several functions but in their different hierarchical
order. The verbal structure of a message depends primarily on the predominant
function. But even though a set (Einstellung) toward the referent, […], briefly the
so-called REFERENTIAL, “denotative”, “cognitive” function – is the leading task
of numerous messages, the accessory participation of the other functions in such
messages must be taken into account by the observant linguist.
Jakobson’s important move was to think in terms of concurrent functions rather

than categorical utterance or activity types. However, because he continued to think
in terms of “predominant functions” that shape “verbal structure” the consequences
of this remain somewhat muted. Consider in this respect a few of Jakobson’s exam-
ples from the passage on the phatic function. Jakobson suggests that there are utter-
ances designed “to attract the attention of the interlocutor or to confirm his
continued attention” and proposes that “Are you listening?” might be an example of
the latter. But in fact studies by Goodwin (1980, 1981) for example clearly show that
this kind of thing is typically not accomplished by an “utterance” devoted to the
task but rather by various forms of same-turn self repair and especially by turn-
restarts. And, inversely, a speaker who says, in so many words, “are you listening?”
will likely not be heard simply as “confirming” the continued attention of the inter-
locutor but reprimanding her for a suspected lapse thereof. The problem here lies in
Jakobson’s failure to distinguish between what is announced (by an explicit question
Participation 129
such as “are you listening” to which other actions are bound to attend) and what is
embodied (by various forms of self-repair for instance) (See Schegloff & Sacks
1973). The distinction is also relevant to his discussion of “entire dialogues with the
mere purport of prolonging communication.” The example he provides looks very
much like what Schegloff and Sacks would describe as a floor-passing exchange
which acts as a “possible pre-closing” and thus serves not to prolong conversation
but rather to bring it to a close. According to this argument, an exchange such as
this “embodies” (rather than announces) a warrant for closing the conversation –
each participant is presented with an opportunity to raise a topic of conversation and
each passes that opportunity thereby conveying “together” that they have nothing left
to talk about. Because, Jakobson had no account of turn-taking he was not able to
see what an exchange such as this might be doing.
But perhaps more importantly there is a sense in which, by equating the “phatic”
with, for instance, the “referential” or “the meta-lingual” Jakobson committed a
category mistake (Ryle 1949) since, on one analysis at least, these are not, in fact,
comparable phenomena.
The terms “referential” and “metalingual” describe unique features of human
language but, as Jakobson himself noted, the “phatic” is a much more general feature
of social interaction.2 We can see this, for instance, in the fact that it is in no way spe-
cific to interaction between humans. Non-human primates, wolves and other mam-
mals use various means to get a conspecific’s attention and by this establish contact
in order to interact. Moreover, it is clear that children establish contact, and thereby
show sensitivity to the phatic dimension, long before they begin talking (Trevarthan
1998). And as this entails, “phatic” communication or the phatic dimension of com-
munication is not limited to speech as both Jakobson and Malinowksi writings on
the topic imply. Indeed, it would seem that gaze as well as head and body orientation
more generally are much more important resources in this area than is speech per se.
To anticipate the discussion that follows somewhat, by limiting a consideration of
the “phatic dimension” to speech, Jakobson and Malinowski cast the associated phe-
nomena as speaker’s practices and thus failed to take account of other participants
including the recipient. This meant that they were never able to conceptualize the
phenomena as interaction between speaker and recipient, that is, to move beyond the
activity of single speaker.
2. For one proposal on the unique features of human language see Hockett (1960) for more
recent discussion see Tomasello (2003, 2008).
130 Jack Sidnell
The importance of gaze and body orientation in establishing contact or what we

will eventually come to call, after Goffman and the Goodwins, participant frameworks
is obvious. The visual arts provide particularly clear examples here.
Figure 2. Officer and Laughing Girl, c. 1657. Johannes Vermeer (1632 – 1675)
In Vermeer’s “Officer and laughing girl,” the artist has captured a moment of
mutual gaze. Notice also the way, by the arrangement of their bodies, the partici-
pants are oriented to one another (on this see Schegloff 1998; Goodwin 2002) in
this way sustaining co-participation in the activity of the moment. Consider, by
way of contrast, the image of “Venus with an organist” by Titian. As Schegloff notes
Participation 131
in his discussion of a similar painting by Titian, the body of the organist is torqued
with the bottom half (from the waist down) set towards the organ and the top half
gazing towards the body of Venus. Schegloff shows that such positionings as this –
body torque – can convey instability and project change. Moreover, the positioning
appears to suggest multiple involvements – the one associated with the orientation
of the head/gaze having being managed within the framework of the one estab-
lished by the lower part of the body. In this version of the painting, there is a second
body torque depicted with Venus herself attending to cupid at her side. So here the
painting suggests that the organist has not established “contact” with Venus, rather,
he is apparently “sneaking” a glimpse of the beauty while he continues the “official”
activity of playing the organ and while she is momentarily distracted by an engage-
ment with Cupid.
Figure 3. Venus and Cupid with an Organist. c.1548. Titian
Goffman (1963: 43ff) noted the possibility of multiple involvements: “main” and
“side” or “dominant” and “subordinate”. Consider in this respect the painting of a peas-
ant dinner in which the main involvement between the man on the right and the man
on the left is sustained via mutual gaze while a subordinate, side involvement between
the man on our left and the woman is underway. Notice that by positioning his hand to
receive the glass, the man displays attention to this task and a willingness to participate
in that subordinate activity of object transfer.
132 Jack Sidnell
Figure 4. Peasants’ Dinner. c. 1618. Diego Velázquez
Consider now the well-known print by Norman Rockwell entitled “Homecoming”.

Here we see a group of co-present persons who together form what Goffman described
as a “focused encounter”.
It is by virtue of the participants’ gaze that we can see they are engaged in a single,
joint activity. With the help of the title, we can see that the marine is telling the story of
his service. Notice in this respect that though no one in the scene is apparently talking
at this moment we can nevertheless recognize that the marine is the one telling the
story. That is, the others are, each one of them and together, collectively, arranged or
oriented in such a way as to show that they are acting as recipients.3
Once we recognize the crucial role that gaze and body orientation play in the orga-
nization of participation, it is possible to see that contact is a pervasive concern and a
contingent accomplishment which is achieved moment-by-moment in the details of
interaction. This of course leads to an account of “involvement” and ultimately to a
3. On the problem of depicting and seeing “non-verbal” communication in the still images of
painting see Gombrich (1972).
Participation 133
Figure 5. Homecoming. Norman Rockwell
theory of “participation”. The pioneer here is the sociologist Erving Goffman and it is
to his early studies that we turn next.
2. Goffman: Attention, involvement and focused encounters
For all their insight, Malinowski and Jakobson thought in terms of “contact” or “chan-
nel”. Although this was a significant improvement on the Saussurean notion of a
“speech circuit”, famously diagrammed in the Cours, in so far as it made the means by
which contact is established a possible topic of investigation, it nevertheless encourages
a massively over-simplified view of the relevant range of phenomena.
134 Jack Sidnell
A B
Figure 6. Saussurean speech circuit
Most problematic is the fact that notions of “contact” and “channel” conceptualize
the phenomena in terms of a binary, on-off function: contact is either established or
not. Once it is established it is assumed to remain in the “on” position until there is
reason to suspect otherwise.
In some of his earliest writings Goffman developed a radically alternative account.
Goffman begins not with language or speech but with social interaction and concep-
tualizes participation therein not in terms of contact but rather “involvement”. For
Goffman each episode of interaction, each encounter, constitutes a self-sustaining
system in which there is, normatively, “a single, albeit moving, focus of visual and
cognitive attention” (Goffman 1964: 135).4 Inevitably, we speak in “social situations”
and Goffman complains that this basic fact of social life has been handled in a “happy-
go-lucky” way, typically in terms defined by research on some other topic.
If one is dealing with the language of respect, then social situations become occasions
when persons of relevant status relationships are present before each other, and
a typology of social situations is drawn directly and simply from chi-squaredom:
high-low, low-high and equals. (Goffman 1964: 34)
Such an approach, according to Goffman, wrongly assumes that “social situations do

not have properties and a structure of their own, but merely mark, as it were, the
geometric intersection of actors making talk and actors bearing particular social attri-
butes” (1964: 134). Against this, Goffman suggests:
Your social situation is not your country cousin. It can be argued that social situations,
at least in our society, constitute a reality sui generis as He used to say, and therefore
need and warrant analysis in their own right, much like that accorded other basic
forms of social organization.
4. Later in his career Goffman noted various permutations of by-play, cross-play and side-play
(see Goffman 1981; Goodwin 1997). See also Egbert 1994.
Participation 135
Goffman (1964: 135) goes on to define the social situation as an environment of

“mutual monitoring possibilities, anywhere within which an individual will find him-
self accessible to the naked senses of all the others who are ‘present’, and similarly find
them accessible to him.”
Within the general category of social situations, Goffman distinguishes unfocused
interactions (such as a group of people waiting at a bustop) from focused interac-
tions which he terms “encounters” (see Goffman 1964). The latter include “card games,
ball-room couplings, surgical teams in operation and fist fights.” Goffman (1964: 135)
continues, saying:
These ventures in joint orientation might be called encounters or face engagements. A

preferential mutual openness to all manner of communication is involved. A physical
coming together is typically also involved, an ecological huddle wherein participants
orient to one another and away from those who are present in the situation but not
officially in the encounter.
It is within these kinds of social arrangements, these focused interactions, these

“encounters” that speaking takes place according to Goffman (talk which does not,
such as talk to oneself, represents a departure from the norm according to Goffman
1964: 136).
And, thus, for Goffman underlying all speech are the practices of participation
by which the environments for talk – social situations – are constituted. He describes
this in terms of “functional gestures” which, as the passage quoted below makes clear,
is meant to be a cover term for a wide range of practices ranging from actual hand
gesturing and body-positioning to intonation and prosody.
Utterances must be presented with an overlay of functional gestures – gestures

which prop up states of talk, police them, and keep these little systems of activity
going. Sounds are used in this gestural work because sounds, in spoken encounters,
happen to be handy; but everything else at hand is systematically used too. Thus
many of the properties of talk will have to be seen as alternatives to, or functional
equivalents of, extra-linguistic acts, as when, for example, a participant signals his
imminent departure from a conversational encounter by changing his posture, or
redirecting his perceivable attention, or altering the intonation contour of his last
statement. (Goffman 1964: 136)
So here, rather that the abstract “speech circuit”, Goffman is proposing a dynamic
system of interaction the moment-by-moment maintenance of which is sustained by
“all-manner of communication.” Various studies by Goodwin (1980, 1981, 1986), Kendon
(1990), Streeck (1992, 1993) and others provide ample evidence that gesture, gaze, and
body orientation are crucially involved here. All speech takes place within some such
interactive system and thus we can see that the Saussurean model of a “speech circuit”
136 Jack Sidnell
and Jakobson’s notion of “contact” and “channel” are abstractions. What they gloss
and thereby obscure are the actual practices of interaction that are prerequisite to any
speech in context: talk-in-interaction.
According to Goffman, face-to-face interaction is an incredibly delicate thing: to
maintain the fiction and the feel of ease each participant must dutifully do their part,
attending to the right things at the right moments, conveying just the right degree
of involvement and so on. In “Alienation from interaction,” Goffman described this
delicate balancing act by which we engage “in a reciprocally sustained communion
of involvement.” He does this by describing the various ways in which participants in
interaction can become alienated from it. He writes (1957: 47):
I want to consider the ways in which the individual can become alienated from a
conversational encounter, the uneasiness that arises with this, and the consequence
of this alienation and uneasiness upon the interaction. Since alienation can occur in
regard to any imaginable talk, we may be able to learn from it something about the
generic properties of spoken interaction.
Goffman argues that in order for interaction to come off smoothly, participants must
satisfy a variety of “involvement obligations” by paying attention in just the right measure
at just the right moments. The problem is that one cannot do this self-consciously
for, as soon as one does, attention is redirected from the course-of-action underway
towards the problem of being spontaneously involved in it. As such, there are any
number of ways in which a participant may become alienated from interaction. Goffman
(1957) describes four such ways.
1. External preoccupation: “The individual may neglect the prescribed focus of

attention and give his main concern to something that is unconnected with what
is being talked about at the time and even unconnected with the other persons
present, at least in their capacity as fellow-participants.” (1957: 49)
2. Self-consciousness: “At the cost of his involvement in the prescribed focus of
attention, the individual may focus his attention more than he ought upon him-
self – himself as someone who is faring well or badly, as someone calling forth a
desirable or undesirable response from others.” (1957: 50)
3. Interactional consciousness: “A participant in talk may become consciously con-
cerned to an improper degree with the way in which the interaction, qua interaction,
is proceeding, instead of becoming spontaneously involved in the official topic of
conversation.” (1957: 50)
4. Other-consciousness: “During interaction, the individual may become distracted
by another participant as an object of attention-exactly as in the case of self-
consciousness he can become distracted by concern over himself.” (1957: 51)
Participation 137
In this early work, Goffman was describing interaction from the point of view of
the individual and he focuses on the causes and consequences of alienation for the
individual participant specifically. The insights of Goffman’s early researches may,
however, be applied to a study of interaction as such. For instance, with the idea of
focused encounter we can consider the practices by which participation is constituted
within any given instance of face-to-face interaction. And we can find cases that are
illuminating precisely because the question of whether some individual is or is not a
participant is called into question.
Consider in this respect the following fragment of interaction from a backyard
barbeque in which three men (Curt, Gary, Mike), one child and one woman (Carney)
are seated at a table. The men have been telling “dirty jokes.” As seen in Figure 7 in one
configuration, Curt, Gary and Mike are arranged so as to form a focused encounter:
an environment of “mutual monitoring possibilities”. Carney, in contrast sits with her
back to the table, facing in the other direction. This organization provides for a kind of
equivocal participation in so far as Carney is present and can hear what the others are
saying without necessarily counting as a recipient of the talk at the table.
Carney
Mike
Gary
Curt
Figure 7. Equivocal participation
The participants can themselves be seen to orient to this equivocality. Thus at one
point Curt is telling a joke to Gary, claiming that he and Mike were in Vietnam and
were “captured by the enemy.”
138 Jack Sidnell
(1) Auto discussion pp. 45

01 CUR: =D’d I ever tell you about th’time Mike
02 en I were in Vietnam en’ got captured by
03 the enemy¿
04 (0.4)
05 MIK: Oh[:: come o::n.]
06 CUR: [Mike en I en,] Y[eah.
07 GAR: [Mike en I?
08 CUR: Mike en [I yeh
09 GAR: [Mackina:w?
10 GAR: [ (Mike’n I.) ]
11 CUR: [ N o m e ] ’n Mike wz in Vietnam we got
12 captured by the
13 Me[‘n Mike ‘n,
14 GAR: [Oh:: horseshit,=
15 CUR: =Ye::s we di:d!
16 (0.7)
17 CUR: ‘n we wz inen there wz me’n Mike en,
18 en another guy en,
19 GAR: Alright, kee[p goin Curt,
20 CUR: [Ennuh:::
21 GAR: [Keep bullshittin] a w a y , ]
22 CUR: [ W e d i : d ? ] Didn’ we.] r’member
23 Mike?
24 CAR: ( [ )
25 GAR: [R’member Mike?
26 MIK: Dah, [yeh, sure enou.gh,
27 GAR: [You remember Mike, he wz
28 [pretty nice ( ).]
29 CUR: [Remember we wz in there]
30 he [s a y[s-
31 GAR: [˙uhh! [
32 CAR: -> [You ferget I’m here.
33 CUR: Ulright,
34 CAR: [Teh hhah hah hah
35 CUR: [Mhh hehh heh[heh
36 MIK: [heh-heh-heh hah-hah
37 [hah heh heh
38 CUR: [ehh heh heh
39 CAR: Go o:n,
40 CUR: We:ll? We wz just in there et the
41 prison camp ennuh,
Here then Curt, who is Carney’s cousin, is telling a joke which he presents as a true
story about the time that he and Mike were captured by the enemy in Vietnam. Gary,
Participation 139
the primary recipient for the joke, expresses doubt saying, in line 07 ‘Mike en I?’ and
then in, line 14, ‘Oh:: horseshit,=’.5 In response to these repeated displays of doubt,
Curt attempts to enlist Mike with ‘Didn’ we. r’member Mike?’ at lines 22–23. At this
point, Carney turns to her cousin Curt and says ‘You ferget I’m here.’ With this Carney
suggests that she can disconfirm what Curt is saying. Indeed, Curt, who has persevered
in the face of multiple challenges by Gary, here appears to acquiesce to Carney. Not
only does he suspend the telling in progress saying ‘ulright’, he joins with Carney in a
bout of laughter. In this way, Curt displays that he has been ‘caught out’ by Carney and
does not in fact proceed with the telling until Carney instructs him to with ‘Go o:n,’.
Notice then that in formulating this in the way she does, Carney characterizes her
own participation as something that has been treated by others as equivocal, ambiguous
or peripheral – as someone, who though present, can be, and has been to this point,
disattended, in fact, forgotten. In a study of this episode (Sidnell frth) I argue that
this equivocality is carefully managed by the participants, and especially by Carney,
in such as way as avert potential recipient-design problems implicated in the telling
of dirty jokes.
3. Goffman: Footing
Goffman’s “Footing” marked a watershed in the analysis of participation. Although

anticipated in many respects by his earlier discussion of “involvement” and face-to-
face encounters and even more so by what he called in 1974 “the frame analysis of
talk,” “Footing” broke new ground and established a touchstone for future studies.
Indeed, the ideas presented have generated a sizeable secondary literature which
branches into conversation analysis, linguistic anthropology, interactional socio-
linguistics and pragmatics.
Goffman begins this article with a news item reporting on an interaction in the
Oval office between (then) president Nixon and a female news reporter. In this first
section of the article, Goffman seems to be talking about “footing” as something akin
to his earlier notion of “role-distance” (Goffman 1961) but he quite quickly shifts
gears again and links footing to what we would today describe as speaker’s stance:
shifts in the alignment the speaker takes up with respect to what she is saying. Indeed,
5. Gary is initiating repair of the phrase ‘Mike and I’ – he appears to finds the use of the (gram-
matically correct) nominative pronoun in this position ‘presumptuous’ – ‘putting on airs’ as it were.
This is a rather interesting case in which the ‘incorrect’ form is oriented-to as the appropriate
one. It may also be that Gary hears this as ‘formulaic’ and thus by initiating repair of it means to
be highlighting its status as an index of the story’s inauthenticity. Note in any case that when he
continues Curt “corrects” to “Me and Mike” and “Mike and me”. On the use of questioning repeats
to challenge what another speaker has said see Schegloff (1997), Sidnell (frth.).
140 Jack Sidnell
in this first section of the paper Goffman links “footing” to a rather bewildering array
of practices: shifts in alignment, code-switching, the framing of utterances, bracketing
different episodes, “keyings” of talk as serious or joking, marking role-distance.
This virtual barrage of connections does not bode well for the argument and it ini-
tially looks as though “footing” is being used to talk about just about everything
and nothing in particular. Thankfully, Goffman reins things in at the conclusion of
this section remarking:
I want to make a pass at analyzing the structural underpinnings of changes in footing.
The task will be approached by re-examining the notions of speaker and hearer, and
some of our unstated presuppositions about spoken interaction. (Goffman 1981: 128)
The rest of the paper is in fact a reasonably focused consideration of different kinds of
participation in interaction. It is this discussion which, for the most part has provided
a foundation for subsequent work in this area. Goffman begins with a critique of what
he describes as the “traditional analysis of saying” which, he claims, is based upon
two basic yet faulty assumptions. First, the traditional analysis of speaking assumes
just two persons: a speaker and a hearer. Second, the traditional analysis fails to take
account of whatever else the participants are doing while, in and through speaking.
This model, we are told “informs the underlying imagery we have about face-to-face
interaction” (Goffman 1981: 129) despite the fact that it can only adequately handle
a small slice of talk-in-interaction. Goffman’s critique of the traditional analysis has
several barbs. These include the following:
1. The unanalyzed notions of speaker and hearer seem to imply that voice and sound
are all that is involved in talking. Goffman notes that this is problematic insofar as
sight and even touch are of obvious importance to a range of practices (including
those involved in the distribution of turns-at-talk for instance).
2. The traditional analysis does not recognize that any bit of talk is always part of “a
talk”: “a substantive, naturally bounded stretch of interaction comprising all that
relevantly goes on from the moment two (or more) individuals open such deal-
ings between themselves and continuing until they finally close this activity out.”
(1981: 130)
3. The traditional notions of “speaker” and “hearer” presuppose a dyadic event and
do not take account of the range of other participants potentially involved in any
occasion of talk.
4. The traditional notions “speaker” and “hearer” are not precise enough. Finer
discriminations are required to describe the range of participant roles that these
terms gloss.
The third and fourth points are directly concerned with the analysis of participation
and it is these that I will focus on in what follows.
Participation 141
Goffman’s argument is that the notions of “speaker” and “hearer” should be decom-
posed. On the speaking side, Goffman suggests that we should think in terms of a
“production format”. On the side of the hearer, he proposes a “participant framework”.
Following Levinson (1988) I substitute the terms “production roles” for “production
format” and “reception roles” for “participant framework”. “Participant roles” serves
as a cover term for the entire range of roles. Goffman uses this typology of different
kinds of participation to construct a corresponding typology of different kinds of talk.
Thus besides the dominating communication among ratified participants, there can be
various forms of subordinate communication. “By-play” is subordinate talk between
ratified participants (see Goodwin 1997), “cross-play” is talk between a ratified partici-
pant and one or more bystanders, and “side-play” is talk among by-standers.
Table 1. Goffman’s participation roles6
Production roles
1. Animator ‘the sounding box’ (p. 144–145)
2. Author ‘the agent who scripts the lines’ (p. 144–145)
3. Principal ‘the party to whose position the words attest’ (p. 144–145)
Reception roles
A: ratified
1. A
ddressed recipient ‘the one to whom the speaker addresses his visual attention and to whom,
incidentally, he expects to turn over his speaking role’ (p. 133)
2. Unaddressed recipient (p. 133) the rest of the ‘official hearers’, who may or may not be listening
B: unratified
1. Over-hearers ‘inadvertent’, ‘non-official’ listeners (p. 132) or bystanders
2. Eavesdropper ‘engineered’, ‘non-official’ followers of talk (p. 132)
Consider the “production roles” first. The animator role is introduced in recogni-
tion of the fact that a speaker may produce talk that is not his or her own. Obvious
examples include a herald (Wagner 1939), a town-crier, a bellman or a beadle who,
in the course of fulfilling his occupational duties, produces talk of which he is not the
author. In reporting speech, at least of the “direct type” (see Holt 1996), the speaker
disclaims authorship of what is said. Actors who speak the lines of a playwright (under
instructions from a director) and hosts who speak while possessed by spirits similarly
fall into this category.7 The Wolof Griot as described by Irvine (1992) presents a more
6. Adapted from Levinson (1988). Page references are to Goffman (1981).
7. On the griot, see Irvine (1992, 1996). The literature on reported speech is voluminous, Holt
and Clift ed. (2007) provides an excellent introduction with a focus on conversation analytic
approaches (See also Volosinov 1973). The literature on spirit possession is similarly expansive, see
142 Jack Sidnell
complicated case. While the Griot is clearly an animator, she is typically not solely an
animator. Rather, as a Griot, the speaker provides both the sounding box and some
aspects of the linguistic and poetic form while at the same time “speaking on behalf
of another”. She may thus be both animator and author but not the principal (see my
discussion of Irvine 1996 below). In the case of talk delivered by a press secretary, the
secretary is often only the animator of the talk, while the author may be an individual
or a committee charged with the task of “spinning” what is to be said (a speech-writer).
The principal, the person ultimately responsible for what is said, is the head of office
whom the press secretary represents.
Writing in 1949, Jakobson (1987) provides a lovely example illustrating the
complexity of “production”. He writes:
Recently, aboard a train, I overheard a scrap of conversation. A man said to a young
Lady, “They were playing ‘The Raven’ on the radio. An old record of a London actor
dead for years. I wish you had heard his Nevermore”. Although I was not the addressee
of the stranger’s oral message, I received it nevertheless and later transposed this
utterance first into handwritten and then into printed symbols; now it has become a
part of a new framework – my message to the prospective reader of these pages.
Jakobson explains the layers of authorship that went into the production of “Nevermore”.
The stranger had resorted to a literary quotation, which apparently alluded to an
emotional experience shared with his female interlocutor. He referred to a performance
allegedly transmitted by broadcast. A dead British actor was the original sender of a
message addressed “to whom it may concern”. He, in turn, had merely reproduced
Edgar Allan Poe’s literary message of 1845. Furthermore, the American poet himself
was ostensibly only transmitting the confession of a “lover lamenting his deceased
mistress” – perhaps the poet himself, perhaps some other man, real or imaginary.
Within this monologue, the word nevermore is attributed to a talking bird, with the
further implication that that one word uttered by the Raven had been caught from
some unhappy master, as the melancholy burden of his customary laments.
Thus the same single word was successively set in motion by the hypothetical “master”,
the Raven, the lover, the poet, the actor, the radio station, the stranger on the train,
and finally, by the present author. The “master” repeatedly exteriorized the elliptic
one-word sentence of his inner speech, nevermore; the bird mimicked its sound
sequence; the lover retained it in his memory and reported the Raven’s part with
reference to its probable provenience; the poet wrote and published the lover’s story,
actually inventing the lover’s, Raven’s, and master’s roles; the actor read and recited for
Boddy (1994) for a useful overview and Lambek (1981) for a classic case study. Urban (1989) and
DuBois (1986) provide useful theoretical framings of these issues.
Participation 143
a recording the piece assigned by the poet to the lover with its nevermore attributed
by the lover to the Raven; the radio station selected the record and put it on the air; the
stranger listened, remembered, and quoted this message with reference to its sources,
and the linguist noted his quotation, reconstituting the whole sequence of transmitters
and perhaps even making up the roles of the stranger, the broadcaster, and the actor.
This is a chain of actual and fictitious senders and receivers, most of whom merely
relay and to a large extent intentionally quote one and the same message, which, at
least to a few of them, was familiar beforehand.8
In these examples there is a clear temporal dimension to the sequence of production

roles: the words originate with an author or principal and are only later articulated by
the animator. But different temporal permutations are possible. Consider for instance
the following example from the famous “Frost-Nixon” interview. This begins with
Frost pressing Nixon on his characterization of what he did as consisting of “mistakes”.
Nixon responds by asking Frost what word he would use to express this. In answering,
Frost mentions “three things I would like to hear you say”. For at least the last two of
these, Frost actually animates an imagined Nixon, saying “I did (.) abuse the power
I had as President.” (line 28–29) and “I: put the American peup-people through two
years of needless agony and I apologize for that.” (lines 31–33).
(2) Frost-Nixon – Frost’s “Three things I would like to hear you say”
01 Frost: Would you: (.) go further than “mistakes”?
02 (0.2)
03 Thet y- you’ve explained how you got caught up
04 (.) in this thing. Youv- You’ve explained your
06 motives. .hhh I don’t wanta quibble about any of
07 that.
08 (0.4)
09 But just coming to the shee:r substance,
10 (0.6)
11 wouldju go further than “mista:kes”.The wo:rd,
12 (1.0)
13 that seems, n- u- not enough for people to
14 understand.
8. Jakobson (1950) notes that examples such as this will not fit within Saussurean speech circuit:
The whole sequence offers a typical example of an intricate process of communication.

It is very different from the trivial pattern of the speech circuit graphically presented
in textbooks: A and B talk face to face so that an imaginary thread goes from A’s brain
through his mouth to the ear and brain of B and through his mouth back to A’s ear
and brain.
144 Jack Sidnell
15 (1.5)
16 Nixon: Welwhat word would you express?
17 (2.0)
18 Frost: My goodness. That’s uh- I: think that there are three
19 things, since you asked me, I: would like to hear you
20 say. I think the American people would like to hear you
21 say. (0.2) One is:
22 (1.2)
23 There was probably mo:re (.) tha:n (0.4) mistakes,there
24 was (0.6) wrongdoing. Whether it was a crime or not.
25 Yes, it may have been a crime, too.
26 (1.0)
27 Secondly, (1.0) I did- (1.0) and I’m saying this without
28 questioning the motives, right. I did (.) abuse the power
29 I had as President. (0.6) or, ah, not fulfill the (.)
30 totality .hh of the oath of office.That-That’s the second
31 thing, (.) And thirdly, (0.6) I: put the American peup-
32 people through two years of needless agony, and I
33 apologize for that. (0.8) And I say that you’ve explained
34 your motives I think those are the categories.
35 (1.0)
36 And I know how difficult it is for anyone and most of all
37 you but I think (0.6) that people need to hear it (.)
38 and I think unless you say it, you’re gonna be haunted
39 for the rest of your life.
Frost’s use of “I” is clearly not self-referential (see Urban 1989). Rather, in a way similar
to reported speech, there is a transposition here such that “I” refers not to the current
speaker but to the speaker in an imagined scene. In this example then Frost animates
Nixon saying something that he has not yet said and, in fact, never did say.
Some shifts of footing are extremely subtle, even artful. Consider in this respect
the following case which begins with Rick offering Matt a bottle opener, accounting for
the rather enthusiastic summons in line 1 with an explanation that the beer bottle Matt
is holding is a “pop” and, therefore, cannot be opened with a bare hand (discussed in
Sidnell 2006). The opener is received and at line 3 Rick is thanked. After Matt jokes
that he was, “trying to be really macho ‘n stuff,” (by not using an opener), Rick begins
to tell a story about a person opening a bottle with his teeth. At line 16, Dave proposes
to tell about a similar event (see Sacks, 1995, for discussion of such “second stories”).
The act of opening the bottle is here referred to using “did that”. It is the result of open-
ing the bottle with one’s teeth that takes center stage in Dave’s story. Thus at line 18,
Dave tells how he looked at the one opening the bottle and at line 19 describes the
person’s teeth as “just all fucked up.”
Participation 145
(3) Rip your hand

1 Rick: ere ere Matt
2 don’t rip your hand off (.) it’s uh uh it’s uh pop
3 Matt: °oh man cool [thanks°
4 Dave: [hahaha
5 Matt: I was tryin to be really macho ‘n [stuff
6 ?: [hahhahha
7 Rick: wn I wuz in Mexico?
8 G: mmhmhm
9 Rick: an’ an’ the du::de that we wen- we wen on a crui::se?
10 (0.5)
11 an this guy opnzit with his mouth like this
12 (0.8)
13 Dave: yes
14 (1.5)
15 G: ouch
16 Dave: uh guy at my house a- um (.) fourth of july did that
17 (0.7)
18 ‘n I wuz all wha::t?=I Iooked at iz teeth and jstall
19 f:::ucked [uh:p
20 J: [hahaha
21 Rick: [hahahha, Oh Shi:t=
22 J: =haha
23 Rick: [I dunnit like this] y’know on the table (.)
24 J: [don’t drink much huh?]
What is remarkable in this example is the way the perspective shifts within the course
of Dave’s talk, from the person watching the events to one participating in them. Thus
when, at line 18, Dave says that he was “all what” he tilts his head slightly to the side in
a querying gesture at the same time widening his eyes (Figure 8). Although, it is dif-
ficult to see this in the figure provided, examination of the video record clearly shows
Dave realigning his head – cocking it to one side – and widening his eyes. Here the
speaker momentarily withdraws his gaze from the co-participants and re-enacts his
own surprise at seeing the result of opening the bottle with one’s teeth. By the time
Dave produces the word “teeth,” he has begun a gesture with his hand and arm. This
comes up over the mouth and makes a sudden slash across his face. At this point, then,
Dave describes the damage done to the teeth. So in the course of this single turn at talk,
Dave has presented to the recipients the events from two quite different perspectives:
First, he has shown them his own reaction as a witness to the events and, secondly, he
has shown them the event itself (a depiction of it). Here then a subtle repositioning of
the head and widening of the eyes marks a significant transformation of the partici-
pation framework. It conveys to the participants that what they are looking at is not
146 Jack Sidnell
simply Dave telling the story but Dave witnessing, with surprise, a person who has just
opened a beer bottle with his teeth and, in doing so, damaged them.
D: uh guy at my house a- um
(.)
fourth of july did that
(0.7)
D: ‘n I wuz all what::t?〉
D: =I looked at iz teeth
and jstall
D: f:::ucked [uh:p
J: [hahaha
R: [hahahha,
oh shi:t=
J: =haha
Figure 8. Footing shift

Participation 147
Consider now the reception roles. A single example will illustrate the bulk of the
categories Goffman introduces. For this we return to the back-yard barbeque known
as Auto-discussion. In this particular episode the participants are initially seated at the
picnic table as follows: on the left-side of the table Curt is in the foreground and Gary
is behind him. On the right-side of the table Carney is in the foreground and Mike in
the back, between them is Phyllis.
Figure 9. “She gonna g’m down here’n break those two up”
At the point where we pick up the action, Carney rises from the table, stands up
and walks around to other side (see Figure 9). As she approaches Gary (her spouse)
and Curt, Mike produces the utterance given as 4 below:
(4) Auto Discussion
06 MIK: Oh look-eh-she gonna g’m down here’n break those two u:[p.
07 CAR: [ehhhh!
08 MIK: se[e:?
09 CUR: [Aw[: ma:n,]
10 MIK: [hah hah] hah hah[hah.
11 GAR: [(You)talk about[j e alous.]
To whom is this addressed? Goffman (1981: 133) suggests that the addressed recipient is
‘the one to whom the speaker addresses his visual attention and to whom, incidentally,
he expects to turn over his speaking role’. While gaze surely plays an important role
here, the answer to the question of to whom an utterance is addressed should, I think,
148 Jack Sidnell
be sought first of all in the structure of the talk itself. In English at least, pronouns
and other forms of person reference are used in such a way as to define a “structure of
address”.9 Third person pronouns and other forms of reference are used to cast par-
ticipants as objects of talk and specifically not the addressed recipient of an utterance.
Here, Mike’s use of “she” to refer to Carney and “those two” to refer to Gary and Curt
casts these persons as non-addressed participants. By a process of elimination then,
this utterance appears to select Phyllis as its recipient.10 Goffman’s suggestion that the
addressed recipient is the one “to whom the speaker addresses his visual attention” is
not borne out: As he is producing this utterance, Mike seems to be looking at Curt and
Gary. It is not until he has finished his utterance that he brings his gaze to Phyllis.11
This example also casts doubt on Goffman’s suggestion that the addressed recipi-
ent is the one to whom speakership passes. In this case, it is Curt who speaks next
saying “Aw man” and after that Gary saying, apparently to Curt, “talk about jealous”.
It’s possible that Goffman put this too strongly and that he should have said that
the addressed recipient is “selected” to speak even if s/he does not do so. But the
evidence available does not support this suggestion either. Rather, as Sacks, Schegloff
and Jefferson (1974, see also Lerner 2003) showed, it would appear that, in order to
select next speaker, an utterance must combine address with some sequence, initiating
action (such as a question, a request, an invitation etc.). This is not to say that gaze is
not used in picking out a particular participant as the addressed recipient for an utter-
ance or even part of an utterance. Indeed as Goodwin (1979) demonstrated, to whom a
turn-at-talk is addressed is frequently worked through in the course of its production.
Not only can different components of an utterance target different participants as the
primary addressed recipient, but, moreover, a speaker may modify the utterance to
make it appropriate to the primary addressed recipient of the moment.
According to Goffman’s scheme Curt, Gary and Carney are undifferentiated as
“unaddressed recipients”. But, they are not solely unaddressed recipients, they are also
persons being talked about. Both Levinson and Goodwin note Goffman’s failure to
discriminate here. Levinson (1988), for instance, discusses the utterance produced by
Mark at line 34 (below), where Karen is a present participant.
9. On this point the classic statement is from Benveniste (1971) who describes the third person
pronoun as a “non-person”. Not all languages work in quite the same way in terms of which forms
are used for address and which for reference. In English for instance it is marked to use kin terms
like “mother” and “father” for self-reference. In Vietnamese, this is perfectly acceptable and
unmarked (see Luong 1990).
10. See Lerner (2003) on next speakers being selected by a process of elimination.
11. Goffman does not fully endorse this idea, see his qualifications in “Footing” (1981: 141).
Participation 149
(5) SN-4 p. 02–03

32 Sher: [You didn’t come t’ talk t’ Kerin?
33 (0.4)
34 Mark: No, Kerin: (·) Kerin ‘n I ‘r having a fight.
35 (·)
36 Mark: After-sh’ went out with Keith (the night be°fore.)
37 ?She: [( )
38 Ruth: [UH HUH HUH HUH HUH
39 (0.2)
40 Ruth: ˙huh ˙huh
41 (??): ((sni [ff))
42 Kar: [W’l Mark you never asked me ou:t.
In a case like this it is clearly problematic to categorize Ruth and Karen together as
“unaddressed recipients.” For though the utterance is explicitly addressed to Sherry as
a response to her question, it is “tacitly” addressed to Karen as a complaint. And notice
that Karen takes it up as a complaint by responding with a justification (which might
itself be heard as a counter-complaint). This and other examples from the literature
suggest the need for a category of “target” (e.g. Fisher 1976).
4. Elaborations and critique of footing
Levinson (1988) presents a highly sophisticated appreciation and critique of the argu-
ment in “Footing” in which he suggests that in seeking to decompose the notions of
speaker and hearer, Goffman was on the right track but that he did not go far enough.
With a nod to Prague school structural phonology, Levinson suggests that participant
roles can be decomposed using a kind of “distinctive feature” matrix which generates
a set of basic categories as well as “super-ordinate natural classes”. Table 2 reproduces
Levinson’s simple version of this for production roles.
Table 2. Feature matrix for production roles (from Levinson 1988)
Message origin Transmission

Author + +
Indirect source + –
Relayer – +
Super-ordinate (‘natural’) classes
Speaker = + transmission
Source = + origin
150 Jack Sidnell
Levinson goes on to provide evidence both from grammatical categories and

from the ethnography of communication that his finer discriminations are indeed
warranted. He notes, for instance, that though first and second person appear to be
basic, universal categories across the world’s languages, non-basic participant roles are
also reflected in a range of grammatical categories. For instance, as Boas (1911) long
ago noted, many languages have obligatory evidential marking which distinguishes
between different degrees of personal commitment. Hidatsa, for instance, includes a
quotative mood indicating “a second-hand but indefinite source” and a report mood
indicating “that the speaker was told the information given in the sentence by someone
else” (Levinson 1988: 184–185, from Matthews 1965: 99ff). As Levinson notes, English
glosses of such sentences are problematic precisely because they make the “quotative”
character of the utterance explicit where, in the original, it is completely embedded.
The point here is that these forms seem to grammatically encode the various different
production roles developed in the analysis of interactional footing.
Levinson discusses a number of grammatical codings which bear on reception
roles as well. He mentions, for instance, hortatives glossed with English “Let” construc-
tions, such as “Let him be welcome.” These are non-second person imperatives:
(6) wio ate awahu kuo axou-ka (Hidatsa, Matthews 1965: 107)
The woman hide in the house
“I demand that the woman hide in the house”
Levinson explains that such “third person imperatives” are of special interest in so
far as they are “a request to an addressee to get some third party to do something”
and thus presuppose two speech acts, a current one and a projected one. They thus
distinguish between an “illocutionary target” and an immediate recipient. Mention
may also be made here of elicitation routines characteristic of interaction between
caregivers and children (see, for example, Schieffelin 1990; Sidnell 1997; de Leon
2000). Levinson also shows that “avoidance registers”, documented for Guugu Yimdhirr
by John Haviland (1979), for instance, index and presuppose affinal “by-standers”
or “overhearers”.
Hanks (1990, 1996) develops ideas from Goffman’s footing as part of his masterful
study of deixis and lived space among the Maya. Like Levinson, Hanks argues that
participation frameworks constitute the foundation of deixis and thus, in one basic
sense, of context. It is often noted that deictic forms such as the English demonstra-
tives “this” and “that” make reference by virtue of an indexical connection to the event
of speaking. More specifically, though, they are tied to the participant roles of speaker
and hearer. On one analysis, “this” and “that” are organized via an asymmetric opposi-
tion in which marked “that” conveying “not close to speaker” is opposed to unmarked
“this” (see for instance Enfield 2003; Sidnell 2005). In this sense, participant categories
provide anchoring via an indexical origo for all other deictic forms.
Participation 151
Of course, participant categories are themselves instantiated through the use of

deictic, or at least indexical forms: pronouns, directed gaze, bodily orientation and
so on. As we’ve seen, in practice, these forms constitute emergent and unfolding
participant frameworks subject to moment-by-moment permutation and transposi-
tion. It thus would seem that the ground to which deixis is anchored is constantly
shifting under one’s feet. Given that situation, Hanks (1996) wonders how the par-
ticipants (the actual incumbents of participant roles that Goffman and Levinson
described) are able to keep it all straight (who is the principal or author now, who
is being addressed?) He goes on to suggest that participation frameworks might be
made more tractable if viewed from the perspective of some “superordinate units
such as events or genres of practice” which will allow for the incorporation of “native
understandings and habitual orientations toward speech, along with a less abstract
view of utterance production” (1996: 168). Hanks (1996) develops this argument in
an analysis of ritual performance.12
Irvine develops Goffman’s and Levinson’s ideas about footing along similar lines
arguing that a central issue has to do with the way in which complex participant role
structures are mapped onto actual incumbents. In this context, Irvine (1996: 136)
suggests that:
The best approach to the mapping problem, then, may be to resist the temptation
to try to arrive at a single, all-purpose solution that thrusts the problem into the
background as merely the rationale for a scheme of PRs. Instead, the mapping
problem itself – the process by which participation structures are constructed,
imagined, and socially distributed – is what should come to the fore. It is not only an
analytical problem; it is also a participants’ problem, to which there are creative, if
often evanescent, solutions.
Elaborating this important critique of the footing model, Irvine argues on the basis
of a consideration of Wolof insult poetry that complex participant mappings should
be understood diachronically. This genre of poetry is performed at weddings after the
wife has moved into the husband’s household. The event is sponsored by the new-
bride’s co-wives but the insult-poetry is delivered by a low-ranking (griot) woman.
Some days before the event the griot visits her patronesses and, together, they compose
the poems to be performed. So Irvine notes we have a split between author and prin-
cipal (the sponsoring/co-authoring co-wives) and the animator (the griot). Irvine
notes then that these footings, achieved in the event itself, presuppose and thus index
another occasion in which a different distribution of roles was in place. Specifically, in
the earlier event the griot was the addressee and the co-wives the speakers (including
the animators).
12. Hanks (1990) also develops the argument for more everyday/conversational contexts.
152 Jack Sidnell
In his discussion of footing, Goodwin (2007), like Irvine, questions the utility of
complex typologies of participants arguing that this in fact obscures the actual interactive
work that participants do in building complex participation frameworks moment-by-
moment in their talk. Goodwin, goes further, suggesting that an essential problem
with the footing model lies in its analytical separation of the speaker and hearer. This,
according to Goodwin and Goodwin (2004), has a number of consequences:
1. “Speakers and hearers inhabit separate worlds” so that while Goffman notes the
importance of mutual monitoring (see M.H. Goodwin 1980), he does not provide
analytic resources for examining the way speakers and hearers take each other
into account in the course of building an utterance.
2. As noted, Goffman’s analysis results in a typology and a set of static categories.
“No resources are offered for investigating how participation might be organized
through dynamic, interactively organized practices.”
3. There is an asymmetry in the analysis of different kinds of actors. The speaker
is a complex actor with an elaborate set of linguistic practices (of embedding,
reporting, contextualizing etc.) whereas all other participants are “left cognitively
and linguistically simple”, indeed, they are characterized solely “as points on an
analytic grid.”
4. The model privileges what is occurring in the stream of speech over all other
forms of embodied practice that, as we’ve seen, are implicated in the organization
of participation – from body orientation to gaze to gesture.
Goodwin (2007: 24) develops a compelling alternate view in which participation is

“analyzed as a temporally unfolding process through which separate parties demon-
strate to each other their ongoing understanding of the events they are engaged in by
building actions that contribute to the further progression of these very same events”
(see also Goodwin 1986, 1987, 1988, 2000; Goodwin & Goodwin 1987, 1990, 1992).
Goodwin demonstrates this through the analysis of an episode in which one par-
ticipant (Ann) tells a story about something her co-present partner (Don) said while
the couple was visiting the new house of friends. While touring the house, Don notices
some wallpaper and remarks “did they make you take this wallpaper or did you pick it
out.” Goodwin shows how Don organizes his body both to anticipate and prepare for
the retelling of this remark. In fact he does this twice. When Ann first starts the report
with “en D(h)o(h)n sa(h)ys,” Don adjusts his body so that his head and face are visually
accessible to the story-recipients. However, Ann momentarily discontinues the report
in order to provide some parenthetical background information. During this talk, Don
again adjusts his body, withdrawing slightly from the gaze of the recipients. When, at
the conclusion of the parenthetical background information, Ann comes again to the
Participation 153
report frame with “Do(h)n said,” Don resumes the postural alignment he had adopted
earlier, making his face visually accessible to the recipients. As Ann reports the faux
–pax about the wallpaper (and comments upon it through the inclusion of laughter), a
smile spreads over Don’s face.
5. Conclusion
In this chapter I’ve attempted to provide a review of a series of important ideas about the
organization of interaction. Although scholars have discussed the phenomena under a
range of quite different rubrics according to their own disciplinary background (speech
circuit, phatic communion, channel, contact and so on) I hope to have shown that the
term “participation” points to a highly organized domain of human conduct that can be
studied empirically. In this I follow one of the pioneers in this area, M.H. Goodwin, who
wrote in her classic study of talk among children that “activities align participants toward
each other in specific ways” and that “this process is central to the way that activities
provide resources for constituting social organization within face-to-face interaction”
(Goodwin 1990: 10). A focus on participation then highlights the fact that talk is as
Goffman insisted first and foremost a form of social organization quite irreducible to
anything else be it the psychology of the individuals involved, the language they employ
or the “culture” they share. Rather, talk-in-interaction is its own organization with its own
distinctive properties and features. Clearly, an analysis of the practices of participation
will be required if we are to continue making gains in the study of that organization.
References
Benveniste, E. (1971). The nature of pronouns. In E. Benveniste, Problems in General Linguistics.

University of Miami Press.
Boas, F. (1911). Introduction. In F. Boas, The Handbook of American Indian Languages 1: 3–83.
Bureau of American Ethnology.
Boddy, J. (1994). Spirit possession revisited: Beyond instrumentality. Annual Review of Anthropology
21: 407–434.
de Leon, L. (2000). The emergent participant: Interactive patterns in the socialization of Tzotzil (Mayan)
infants. Journal of Linguistic Anthropology 8(2): 131–161.
Du Bois, J. (1986). Self-evidence and ritual speech. In W. Chafe and J. Nichols (eds), Evidentiality: The
Linguistic Coding of Epistemology: 313–336. Ablex.
Egbert, M.M. (1994). Schisming: The collaborative transformation from a single conversation to multiple
conversations. UCLA Ph.D. dissertation.
Enfield, N.J. (2003). Demonstratives in space and interaction: data from Lao speakers and implications
for semantic analysis. Language 79(1): 82–117.
Fisher, L.E. (1976). Dropping remarks and the Barbadian audience. American Ethnologist 3: 227–42.
154 Jack Sidnell
Goffman, E. (1957). Alienation from interaction. Human Relations 10(1): 47–60.

Goffman, E. (1961). Encounters: Two Studies in the Sociology of Interaction. Bobbs-Merrill.
Goffman, E. (1963). Behavior in Public Places: Notes on the Social Organization of Gathering. Free
Press.
Goffman, E. (1964). The neglected situation. American Anthropologist 66(6, pt. II): 133–136.
Goffman, E. (1974). Frame Analysis: An Essay on the Organization of Experience. Harper and Row.
Goffman, E. (1981). Footing. In E. Goffman, Forms of Talk. University of Pennsylvania Press.
Gombrich, E.H. (1972). Action and expression in Western art. In R. Hinde, Non-Verbal Communi-
cation: 373–393. Cambridge University Press.
Goodwin, C. (1979). The interactive construction of a sentence in natural conversation. In G. Psathas,
Everyday Language: Studies in Ethnomethodology: 97–121. Irvington Publishers.
Goodwin, C. (1980). Restarts, pauses, and the achievement of mutual gaze at turn-beginning.
Sociological Inquiry 50: 272–302.
Goodwin, C. (1981). Conversational Organization: Interaction Between Speakers and Hearers.
Academic Press.
Goodwin, C. (1986). Audience diversity, participation and interpretation. Text 6(3): 283–316.
Goodwin, C. (1987). Forgetfulness as an interactive resource. Social Psychology Quarterly 50(2):
115–130.
Goodwin, C. (1988). Participation frameworks in children’s argument. In K. Ekberg & P.E. Mjaavatn,
Growing Into A Modern World: Proceedings from An International Interdisciplinary Conference
on the Life and Development of Children in Modern Society: 1188–1195. The Norwegian Centre
for Child Research.
Goodwin, C. (2000). Action and embodiment within situated human interaction. Journal of Pragmatics
32: 1489–1522.
Goodwin, C. (2002). Time in action. Current Anthropology 43(Supplement): 19–35.
Goodwin, C. (2007). Interactive footing. In E. Holt & R. Clift, Reporting Talk: Reported speech in
interaction: 16–46. Cambridge University Press.
Goodwin, C. & M. Goodwin (2004). Participation. In A. Duranti, A Companion to Linguistic
Anthropology: 222–244. Blackwell.
Goodwin, C. & M.H. Goodwin (1987). Concurrent operations on talk: Notes on the interactive
organization of assessments. IPrA Papers in Pragmatics 1(1): 1–52.
Goodwin, C. & M.H. Goodwin (1990). Interstitial argument. In A. Grimshaw, Conflict Talk: 85–117.
Goodwin, C. & M.H. Goodwin (1992). Context, activity and participation. In P. Auer & A. di Luzio,
The Contextualization of Language: 77–99. Benjamins.
Goodwin, M.H. (1980). Processes of mutual monitoring implicated in the production of description
sequences. Sociological Inquiry 50: 303–317.
Goodwin, M.H. (1990). He-Said-She-Said: Talk as Social Organization among Black Children. Indiana
University Press.
Goodwin, M.H. (1997). By-play: negotiating evaluation in story-telling. In G.R. Guy, J. Baugh,
D. Schiffrin & C. Feagin, Towards a Social Science of Language: Papers in Honour of William Labov:
77–102. John Benjamins.
Goodwin, M.H. & C. Goodwin (1986). Gesture and coparticipation in the activity of searching for a
word. Semiotica 62(1/2): 51–75.
Hanks, W.F. (1990). Referential Practice: Language and Lived Space Among the Maya. Chicago,
University of Chicago Press.
Participation 155
Hanks, W.F. (1996). Exorcism and the description of participant roles. In M. Silverstein & G. Urban,
Natural Histories of Discourse: 160–200. Chicago University Press.
Haviland, J.B. (1979). How to talk to your brother-in-law in Guugu Yimidhirr. In T. Shopen, Languages
and Their Speakers: 161–240. University of Pennsylvania Press.
Hockett, C.F. (1960). The origin of speech. Scientific American 203: 88–96.
Holt, E. (1996). Reporting on talk: the use of direct reported speech in conversation. Research on
Language and Social Interaction 29(3): 219–245.
Holt, E. & R. Clift, (eds.) (2007). Reporting Talk: Reported Speech in Interaction. Cambridge University
Press.
Irvine, J.T. (1993). Insult and responsibility: Verbal abuse in a Wolof village. In J.H. Hill & J.T. Irvine,
Responsibility and Evidence in Oral Discourse: 105–134. Cambridge University Press.
Irvine, J.T. (1996). Shadow conversations: The indeterminacy of participant roles. In M. Silverstein &
G. Urban, Natural Histories of Discourse: 131–159. Chicago University Press.
Jakobson, R. (1980 [1957]). Metalanguage as a linguistic problem. In S. Rudy, Contributions
to Comparative Mythology: Studies in Linguistics and Philology. Vol. 7 of Selected Writings:
113–121. Mouton.
Jakobson, R. (1987 [1949]). Language in operation. In K. Pomorska & S. Rudy, Language in Literature:
50–61. Belknap.
Jakobson, R. (1960). Linguistics and poetics. In T. Sebeok, Style in Language: 350–377. MIT Press.
Kendon, A. (1990). Conducting Interaction: Patterns of Behavior in Focused Encounters. Cambridge,
Lambek, M. (1981). Human Spirits : A Cultural Account of Trance in Mayotte. Cambridge University
Press.
Lerner, G. (2003). Selecting next speaker: The context sensitive operation of a context-free organization.
Levinson, S. (1988). Putting linguistics on a proper footing: explorations in Goffman’s concepts of
participation. In P. Drew & A. Wootton, Erving Goffman: Exploring the Interaction Order: 161–227.
Northeastern University Press.
Luong, H.V. (1990). Discursive Practices and Linguistic Meanings: the Vietnamese System of Person
Reference. J. Benjamins.
Malinowski, B. (1923). The problem of meaning in primitive languages. In C.K. Ogden & I.A. Richards,
The Meaning of Meaning: 296–336. Harcourt, Brace & World, Inc.
Matthews, G.H. (1965). Hidatsa Syntax. Mouton.
Ryle, G. (1949). The Concept of Mind. Barnes & Noble.
Sacks, H. (1995). Lectures on Conversation (2 Vols.). Basil Blackwell.
Sacks, H., E.A. Schegloff, et al. (1974). A simplest systematics for the organization of turn-taking for
conversation. Language 50: 696–735.
Saussure, F.D. (1959). Course in General Linguistics. Edited by Charles Bally & Albert Sechehaye,
in collaboration with Albert Riedlinger, translated from the French by Wade Baskin. Philosophical
Library.
Schegloff, E.A. (1997). Practices and actions: Boundary cases of other-initiated repair. Discourse
Schegloff, E.A. (1998). Body torque. Social Research 65(3): 535–596.
Schieffelin, B.B. (1990). The Give and Take of Everyday Life: Language Socialization of Kaluli Children.
156 Jack Sidnell
Sidnell, J. (1997). Organizing social and spatial location: Elicitations in Indo-Guyanese village talk.
Journal of Linguistic Anthropology 7(2): 143–165.
Sidnell, J. (2005). Deixis. In J. Verschueren et al., Handbook of Pragmatics. John Benjamins.
Sidnell, J. (2006). Coordinating gesture, talk, and gaze in reenactments. Research on Language and
Social Interaction 39(4): 377–409.
Sidnell, J. (frth a). “D’you understand that honey”: Gender and participation in conversation.
In S. Speer & E. Stokoe, Conversation and Gender. Cambridge University Press.
Sidnell, J. (frth b). Questioning repeats in the talk of four-year old children. In H. Gardner &
M. Forrester, Analysing Interactions in Childhood: Methods and Applications. Wiley.
Streeck, J. (1993). Gesture as communication I: Its coordination with gaze and speech. Communication
Monographs 60(4): 275–299.
Streeck, J. & U. Hartge (1992). Previews: gestures at the transition place. In P. Auer & A. di Luzio, The
Contextualization of Language: 135–157. Benjamins.
Tomasello, M. (1999). The cultural origins of human cognition. Harvard University Press.
Tomasello, M. (2003). Constructing a Language: A Usage-Based Theory of Language Acquisition.
Harvard University Press.
Tomasello, M. (2008). Origins of Human Communication. MIT Press.
Trevarthan, C. (1998). The concept and foundations of infant intersubjectivity. In S. Braten, Intersubjective
Communication and Emotion in Early Ontogeny: 15–46. Cambridge University Press.
Urban, G. (1989). The ‘I’ of discourse in Shokleng. In B. Lee & G. Urban, Semiotics, Self, and
Society: 27–51. Mouton de Gruyter.
Volosinov, V.N. (1973). Marxism and the Philosophy of Language. Translated by Ladislav Matejka &
I.R. Titunik. Seminar Press. (First Published 1929 and 1930.).
Wagner, A.R. (1939). Heralds and Heraldry in the Middle Ages: An Inquiry Into the Growth of the
Armorial Function of Heralds. Oxford University Press.
Politeness
Gabriele Kasper
University of Hawaii at Manoa
1. Historical overview
As a research object in linguistic pragmatics, politeness has a rather short history. The
impulse to the study of politeness came from H.P. Grice’s seminal paper ‘Logic and
conversation’ (1975), in which he proposed a Cooperative Principle and four maxims under-
lying transactional discourse (“maximally efficient information exchange”). Grice
noted that in order to account for other aspects of language use, additional maxims
may be needed, such as a politeness maxim. This suggestion was taken up and elaborated
in early accounts of linguistic politeness by Lakoff (1973, 1979), Leech (1977, 1983),
and Edmondson (1979, 1981). The common denominator of these proposals is that
they view politeness in terms of maxims or rules, thus underscoring their conceptual
link to the Gricean maxims.
A different approach to politeness was proposed by Brown & Levinson (1978/87),
who derive politeness from Goffman’s (1971) notion of face. To date, this has proven
to be the most influential proposal, extremely powerful in its generation of research
and controversy. Most theoretical contributions and empirical studies adopt Brown &
Levinson’s work as a reference point, debating in particular the universality of
their theory.
2. Approaches to politeness
2.1 Folk notion

Many though not all languages have a lexical item roughly equivalent to ‘politeness’
in English. The changing semantics of these lexical entries provide a window to the
perception of politeness in the speech community, from a diachronic and synchronic
perspective. For instance, the lexical element ‘li’ in Chinese ‘limao’ (=‘polite appearance’)
underwent important semantic changes which reflect such historic sensitivity: in
Confucius’ writings (551–479 B. C.), ‘li’ referred to the slavery-based social hierarchy
of the Zhou Dynasty, which Confucius advocated to be restored. The modern concept
of ‘li’ was first documented in the book Li Ji (attributed to Dai Sheng, ca. 200–100 B. C.),
where ‘li’ was equated with demonstration of self-denigration and respect to the other
158 Gabriele Kasper
person, especially in vertical relationships (Gu 1990). In English, ‘polite’ dates back
to the 15th century (=‘polished’); in the 17th century, a polite person was ‘of refined
courteous manners’ (The Oxford Dictionary of English Etymology). Politeness was
thus associated with the norms of social conduct extant in the upper classes, and this
sense of the term has survived in collocations such as ‘polite society’. The semantic
association of ‘polite’ with behaviors of the upper classes is even more obvious in German
‘höflich’, French ‘courtois’, Spanish ‘cortes’, all of which are adjectival derivatives of
nouns for court (G. Hof, Fr. cour, Sp. corte).
The semantics of the lexical entry ‘politeness’ thus sheds light on social mem-
bers’ perception and classification of politeness. Further sources of politeness as a folk
notion are key cultural concepts, etiquette manuals, the prescriptions and proscriptions
of socializing interaction, and metapragmatic comments on what is and is not polite
behavior. Since they are embedded in the cultural practices of a community, these
assessments and activities provide an emic perspective on politeness as an action-
guiding construct in the community. Politeness as a folk notion, representing the
‘social norm view’ of politeness in Fraser (1990), has also been categorized as ‘first
order politeness’. ‘Second order politeness’, by contrast, presents different theoretical
accounts of politeness (Watts, Ide & Ehlich 1992). First order politeness phenomena
constitute the empirical input to politeness theories.
2.2 Conversational maxim(s)

While all conversatinal maxim accounts of politeness are modeled on the Gricean
maxims, they differ in their relationship to the Cooperative Principle (CP). The
views reach from entirely subsuming politeness under the CP to affording the CP
and politeness equal status. Politeness, defined as ‘considerateness’, is often seen as
inherently face-threatening, and needs to be counterbalanced by appropriate doses
of politeness. Brown and Levinson distinguish a number of major politeness strate-
gies, ranging from avoiding the face-threatening act (FTA) altogether, to carrying it out
in different guises. Off-record FTAs are performed indirectly; due to their inher-
ent ambiguity, they have the greatest potential for negotiation (including denial).
On-record performance of FTAs can be achieved without redressive action, or by
adopting either or both of two kinds of redress: positive politeness, addressing the
hearer’s positive face wants, or negative politeness, addressing negative face wants.
Positive politeness strategies emphasize closeness between speaker and hearer by
confirming or establishing common ground, or by referring to desirable attributes
in the hearer (hence ‘solidarity strategy’, Scollon & Scollon 1983); negative politeness
strategies suggest distance by accentuating the hearer’s right to territorial claims and
freedom from imposition (hence ‘deference strategy’, Scollon & Scollon 1983).
Politeness 159
Objections against Brown & Levinson’s theory have centered on their notion of
face. While in Goffman’s (1967) original concept, face is a public property, ‘on loan’ to
the individual from society, and a negotiable outcome of social interaction, Brown &
Levinson’s construct is an individual’s unalienable possession, a trait that can be threat-
ened but not lost or taken away. It has been argued that the individualistic orientation
of Brown & Levinson’s face notion conflicts with cultural orientations outside the Anglo-
American community, where ‘face’ predominantly entails recognition of participants’
placement in social hierarchies (Ide 1989; Matsumoto 1988). In an attempt to reconcile
egocentric and sociocentric views of face, Mao (1994) proposed a ‘relative’ face
orientation: the ‘ideal social identity’ and the ‘ideal individual autonomy’. Societies
differ according to which of these two interactional ideals they favor; hence substantive
face constructs vary cross-culturally.
Writers who recognize a role for face in politeness have recently pointed out that
face can only be correctly understood in the context of notions of self, emphasizing
that such notions are necessarily informed by culturally varying perceptions of person-
hood and relationships between individual and society. While a comprehensive review
of studies on self concepts in different communities points to a consistent opposition
between interdependent and independent notions of self (Markus & Kitayama 1990), an
alternative view questions the adequacy of categorizing cross-culturally varying self-
orientations according to these categories, or even as ordered on a continuum between
these. Thus Rosenberger (1989) argues against the popular belief of a consistent
sociocentric self concept in the Japanese community, and for a dialectic model as one
of many maxims, on the same epistemological footing with the Maxims of Quality,
Quantity, Relevance, and Manner. Consequently, violating the Politeness Maxim gives
rise to conversational inference, just as in the case of any other maxim. While polite-
ness thus has a secondary status vis-à-vis the CP in Brown & Levinson’s (1987) theory,
Lakoff (1973) and, in a much elaborated version, Leech (1983) see politeness as a coor-
dinate construct to the CP. For Lakoff, pragmatic competence is constituted by two
major ‘rules’: “1. Be clear. 2. Be polite,” where clarity amounts to a condensed version
of the Gricean maxims, while politeness serves to avoid conflict between participants.
In Leech’s proposal of an ‘interpersonal rhetoric’, the CP is complemented by a Politeness
Principle (PP): “Minimize the expression of impolite beliefs” (1983, p. 79). Both CP
and PP are ‘second-order principles,’ each elaborated by a set of ‘contributory maxims’:
the Gricean maxims in the case of the CP, and six maxims of politeness — the Maxims
of Tact, Generosity, Approbation, Modesty, Agreement, and Sympathy — in the case of
the PP (p. 131ff). The ‘conversational maxim view’ (Fraser 1990) of politeness thus comes
in different versions, depending on how the relationship between the CP and polite-
ness is conceptualized. Less detailed versions of the conversational maxim approach
to politeness have been suggested by Lakoff (1973, 1979), Edmondson (1981), and
160 Gabriele Kasper
Kasher (1986), and for Chinese by Gu (1990). Myers-Scotton’s (e.g. 1993) markedness
model, developed as a conversational maxim approach to the sociopragmatic functions
of codeswitching in multilingual communities, can also account for politeness marking
in monolingual or bilingual interaction.
2.3 Redress to face-threat

The face-saving view of politeness, proposed by Brown and Levinson (1978/1987),
has been the most influential politeness model to date. Its fundamental assumptions
are a Weberian view of communication as purposeful-rational activity, combined
with Goffman’s concept of ‘face’ (1967), an individual’s publicly manifest self-esteem.
Social members are endowed with two kinds of face: ‘negative face’, the want of self-
determination, and ‘positive face’, the want of approval. Participants are assumed to
adopt as a global interactional strategy “the diplomatic fiction of the virtual offense, or
worst possible reading” (Goffman 1971: 138f), i.e., the working hypothesis that face is
constantly at risk. Consequently, any kind of linguistic act which has a relational dimen-
sion is seen which captures the switches of the Japanese self (jibun) between opposing
orientations: “group productivity, personal accomplishment, harmony or affection, and
pure impuls or gratification” (1989: 89f). Switching between these modes is brought
about through the flow of a person’s vital energy (ki) and actualization according to social,
spatial, and temporal context (also Lebra 1986). ‘Western’ and Japanese notions of self
differ not so much in their ‘independent’/‘egocentric’ vs. ‘interdependent’/‘sociocentric’
orientations but in their diverging beliefs about the unity of self: the ‘western’ ideal of a
consistent self that transcends conflicting contextual demands, and the Japanese ideal of
an accommodative self that optimally responds to varying contexts and purposes. The
ideological character of the ‘western’ construction of a consistent self has been illustrated
in an analysis of team sport in the US (Green 1992), demonstrating how athletes are
required to switch from the individualistic orientation prevalent in the society at large
to a strongly group-oriented, hierarchical subculture. The facility by which these adjust-
ments are performed suggests that ‘western’ selves are more contextually sensitive than
presumed by folk beliefs as well as some scientific models of self, e.g. in Freudian or
Jungian psychodynamic theory.
2.4 Social marking

According to some critics, the face-based politeness construct does not accommodate
politeness in the sense of social marking, e.g. by choosing a sociolinguistically
appropriate style and forms of person deixis. Whereas strategic politeness is called for
as an antidote to the face-threat involved in many linguistic acts, social marking is seen
as a pervasive normative feature of verbal interaction, unrelated to particular com-
municative goals. The social marking view thus represents a second-order approach to
politeness as social norms.
Politeness 161
Different proposals have been made to conceptualize the social marking position
relative to the face-saving view. According to Ide (1989), Brown & Levinson’s notions
of positive and negative face are universally valid but have to be complemented with
a component called discernment (Japanese wakimae), signalling social relationships.
Politeness in any society comprises a ‘volitional’ component (strategic politeness
attending to face concerns) and discernment, or social marking. These two components
are regarded as universal; communities differ in the emphasis they put on each. Thus
for Japanese interlocutors, ‘place’ purportedly takes precedence over ‘face’ (Ide 1989).
In contrast to Ide’s universalist position, Matsumoto (1988) rejects the notion of
negative face altogether as being inapplicable to Japanese culture. Her work emphasizes
the importance of social marking in Japanese language use and does not explore how
strategic politeness is encoded in Japanese interaction. Her position therefore represents
a ‘place instead of face’ view.
Ide’s proposal that any politeness theory has to account for relationship marking
and redress to face-threat, while being theoretically convincing, also has the advantage
of being empirically testable. It is more difficult to provide empirical support for her
contention that speech communities place differential premium on social marking and
face-saving strategies. In order to sustain this claim, etic criteria have to be found to
measure amounts and social significance of both politeness features across cultures; this
task remains to be accomplished by future data-based studies on linguistic politeness.
While conventional, or sociopragmatically unmarked social indexing affirms the
existent power structure in social relationships, unconventional, marked uses do occur.
Such deviations from expected choices (e.g. of personal deixis, styles, or indeed languages)
carry special social messages. In light of Myers-Scotton’s (1993) markedness model, they
may be seen as strategic exploitations of unmarked choices (e.g. Cook 1996a, b).
2.5 Conversational contract

The conversational contract view, proposed by Fraser (e.g. 1990), represents the most
global perspective on politeness. Rather than conceiving of politeness as something
additional to cooperation, such as redress in view of face-threat, polite conduct implies
acting in accordance with the requirements of the conversational contract at any given
moment of an encounter. At the outset, the terms of the conversational contract are
determined by participants’ rights and obligations; however these may change during
and as a result of the interaction itself. What exactly the current terms of the conversa-
tional contract are would depend on each participant’s assessment of relevant contextual
factors and of the conversational interchange itself. Acting politely, then, is virtually the
same as using language appropriately. In order to distinguish politeness in this broad
sense from speakers’ deliberate expression of respect and appreciation, Fraser follows
Goffman (1971) in referring to this aspect of linguistic activity as deference.
162 Gabriele Kasper
2.6 Politeness and politic behavior

Yet another line of criticism denies the role of face in politeness altogether. Watts
(1989) asserts that rather than being motivated by face concerns, politeness is located
in the wider context of politic behaviour, understood as (linguistic) activity serving to
establish and maintain interpersonal relationships. With special reference to Chinese
politeness, Gu (1990) argues that politeness is more appropriately seen as adherence to
social norms than attending to individual’s face wants. While these authors’ proposals
are thus at variance with Brown & Levinson’s individualistic notion of face, they are
quite compatible with the relative face concept proposed by Mao (1994).
2.7 Politeness and tact

Whereas Leech (1983) had proposed a tact maxim as one of the implementations of
his Politeness Principle, Janney & Arndt (1992) distinguish two coordinate types of
politeness, ‘social politeness’ and ‘tact.’ Social politeness refers to participants’ ability to
organize interaction smoothly by coordinating their activities along the conventions
holding for any given encounter. The main vehicle for social politeness is the routine
formulae serving to facilitate interaction in standard situations, such as structuring
and managing conversation and performing recurring speech acts. Social politeness
thus refers to observing the rules of the social-communicative game, while tact refers
to participants’ need to mutually save and enhance face. Hence, the function of tact is
to regulate interpersonal relationships by providing support, empathy, and avoiding
offense. In biological perspective, tact appears as a species-specific adaptive response,
allowing humans to produce affective display purposefully (emotive communication)
rather than as the mindless externalization of an affective state (emotional com-
munication). Both social politeness and tact are universals of human behavior; yet
communities differ in the ways they organize their social interaction, what counts as
tactful, and how tact is strategically and linguistically implemented.
3. Expression of politeness
3.1 Inherently polite speech acts?

Speech acts have been classified according to their inherent face-threat (threat to
speaker/hearer’s positive/negative face; Brown & Levinson 1978/1987) and interpersonal
impact (competitive, convivial, collaborative, conflictive; Leech 1983). Such classifications
are theoretically problematic because what is categorized as face-threatening or inter-
personally beneficial or disruptive would depend on the larger conceptual framework;
an empirical drawback is that abstract categorizations of speech acts may tell us little
Politeness 163
about as how polite or impolite participants perceive such linguistic acts in ongoing
interaction. On the other hand, at the level of first-order politeness, at least routinized
speech acts such as greeting, thanking, complimenting, or apologizing are categorized
as intrinsically polite in many communities. Whereas clearly any of these acts can be
used in a socially offensive way, this is perceived as highly marked and does not invalidate
their categorization as ‘polite’ in unmarked use.
Some types of linguistic action are carried out more frequently in some cultures
than in others, and this seems to reflect their politeness value as perceived in the com-
munity. Hearer-beneficial acts such as complimenting and thanking occur more regularly
in the US than in mainland China, reflecting both the strong positive politeness ori-
entation and reluctance to impose on others in mainstream American culture, on the
one hand, and the assumption, in China, that participants act according to their social
positions and associated roles and obligations, on the other hand. Also, hearer-costly
acts such as refusals are perceived as being more socially offensive by Japanese and
Chinese interlocutors and thus tend to be avoided, whereas it seems more consistent
with American interlocutors’ right to self-determination not to comply with another
person’s wishes. However, because refusing is threatening to the hearer’s positive face
and therefore problematic for American speakers, too, they tend to mitigate refusals
in various ways.
3.2 Conventions of means

Brown & Levinson (1978/87) offer rich evidence for the microstrategies available to
implement each of the major strategic choices for carrying out face-threatening acts.
Most of the subsequent studies have examined how specific speech acts are realized,
within a particular speech community or cross-culturally/cross-linguistically. For a num-
ber of speech acts, finite repertoires of realization strategies, or speech act sets, have been
identified, e.g. for requests (Blum-Kulka & House 1989), apologies (Olshtain & Cohen
1983), complaints (Olshtain & Weinbach 1993), compliments (Herbert & Straight
1989), refusals (Beebe & Cummins 1996), rejections (Bardovi-Harlig & Hartford
1990), disagreement (Sornig 1977), chastisement (Beebe & Takahashi 1989b), giving
embarassing information (Beebe & Takahashi 1989a), corrections (Takahashi & Beebe
1993), thanking (Eisenstein & Bodman 1993), suggestions (Banerjee & Carrell 1988),
and offers (Fukushima & Iwata 1987).
Current evidence suggests that the established speech act sets are cross-linguistically
robust. Thus for requests, modificatory dimensions include three major levels of directness
(direct, conventionally indirect, indirect), measured in terms of distance between locu-
tion and illocution; internal modification of the requestive act, mitigating or aggravating
impositive force; and external modification, expressed by ‘adjuncts’ supporting the
request proper. The speech act set of apology consists of an expression of regret,
164 Gabriele Kasper
admission of the offense, assumption of responsibility, account, minimizing the offense/

responsibility, and offers of compensation (Blum-Kulka, House & Kasper 1989). The
strategies composing specific speech act sets, and global modificatory dimensions such
as indirectness, minimization and maximization of pragmatic force (House & Kasper
1981; Held 1989) have been demonstrated to be valid across the languages and com-
munities studied so far. Yet the selections participants make from such repertoires vary
between speech communities, and these variations systematically reflect different cul-
tural orientations. Chinese (PRC) speakers not only compliment less, but also on differ-
ent attributes compared to American speakers. For instance, physical attractiveness is
not regarded a suitable topic for public praise by many Chinese people (Ye 1995). Dan-
ish interlocutors invested more effort into supporting their requests by justifications
than native speakers of English and German (Færch & Kasper 1989). Ritual refusal of
offers and invitations is traditionally required in Chinese culture (Gu 1990; Mao 1994)
but not everywhere else. Americans are more prone to preface corrections with positive
remarks than Japanese speakers (Takahashi & Beebe 1993) and to maximize compli-
ments, both of which reflect a positive politeness orientation. In refusals, Japanese and
Chinese speakers use proverbs and philosophical statements more often than American
speakers. While the indirect expression of pragmatic intent is a universally available
strategy, the relationship of indirectness and politeness is complex and varies cross-
culturally. In American English and Hebrew, conventional indirectness, rather than
non-conventional indirectness, was identified as the most polite request strategy; next
in politeness value were direct strategies in the perception of Israeli raters and noncon-
ventional indirectness according to American raters (Blum-Kulka 1987). In interaction
with familiars but not intimates, direct strategies (e.g. imperatives, statement of hearer’s
future action) seem more acceptable in the Slavic languages, Hebrew and German than
in any of the standard varieties of English examined so far. It has been suggested that
directness to speakers of these languages connotes sincerity, straightforwardness and
cordiality (Katriel 1988; Wierzbicka 1985) rather than imposition on their freedom of
action. However, in these languages direct strategies are often combined with mitigating
modal particles, diminutives and other linguistic means which do not alter the direct
expression of pragmatic intent but soften impositive force; politeness is thus explicitly
marked by different linguistic expressions, rather than dispensed with altogether.
3.3 Conventions of form
Languages differ in the conventionalized forms by which illocutionary force and polite-
ness can be implemented. For instance, formal equivalences of the structure ‘can/could
you do x’ serve to express conventional indirectness in the Germanic and Romance
languages as well as in Hebrew, Japanese, and Chinese, but not in the Slavic languages
Politeness 165
(Wierzbicka 1985). Repertoires of polite formulae index recurrent social events which
are most efficiently handled by language-specific pre-fabricated routines (Coulmas 1981).
It is part of the pragmalinguistic knowledge shared by members of a speech commu-
nity to recognize the contextual distribution of such frozen forms, whereas speakers do
not usually have an awareness of the semantic structure of politeness routines. To the
pragmaticist, such formulae are informative tokens of first order politeness since they
encapsulate the interpersonal meanings which are regularly conveyed in the community.
Stylistic choices, honorifics, and address terms function primarily as social indicators
rather than expressions of strategic politeness, although they can also be used for that
purpose. The assumption that social indexing may be more prevalent in some languages
than others is well supported by the fact that in Asian languages such as Japanese, Korean,
Thai, Javanese and others, relationship marking is grammaticized in highly complex
morphological systems, whereas such specialization is only rudimentary in European
languages (Coulmas 1992). Taxonomies of conventions of means and form have been
established crosslinguistically and for specific languages (e.g. Blum-Kulka 1989). What is
still not well understood, however, is the relative contribution of such conventions to the
overall politeness value of an utterance or discourse sequence.
4. Variables in politeness investment
It is an in-built sociolinguistic axiom of politeness theories that politeness investment

varies according to contextual factors. The two most elaborated theories, Brown &
Levinson (1978/1987) and Leech (1983), concur in this regard. First, they identify the
same factors as independent variables in politeness marking: social distance (Brown &
Levinson, Leech), social power (Brown & Levinson)/authority (Leech), and the degree
of imposition associated with a face-threatening act (Brown & Levinson)/costs and
benefits of an act (Leech). Secondly, both theories posit a linear relationship between
these factors and politeness investment. Thirdly, they both assume a positive correla-
tion between politeness and indirectness.
Data-based studies lend strong support to the identified context variables, whereas
the correlational issues are problematic. Each of the proposed factors represents a
composite construct which is culturally and contextually elaborated and weighted.
Social power includes factors such as
– interlocutors’ relative positions in social hierarchies (e.g. Morand 1991);
– age; e.g. in communication by and with the elderly (Coupland, Coupland & Giles
1991) and children (Kwarciak 1993, for overview);
– gender (Fasold 1990, for overview);
– language impairment (Bliss 1992; Stemmer 1994).
166 Gabriele Kasper
The impact on expended politeness of factors related to social power varies intra-
and interculturally. Intraculturally, the weight of each factor might vary contextually.
For instance, gender was identified as determining the amount of complimenting
and apologies received by New Zealand interlocutors, women receiving consistently
more, mostly from other women (Holmes 1988). Comparing politeness in requests,
American students did not vary politeness according to gender, whereas Hispanic stu-
dents were more polite towards the opposite sex (Rintell 1981). Different positional
roles in institutional hierarchies (university, work place) affected the politeness invest-
ment of American and Japanese interlocutors; however, Japanese speakers varied their
politeness strategies more dramatically, depending on whether speaker or hearer was
the status superior (Takahashi & Beebe 1989).
Social distance has been demonstrated to impact politeness in a more com-
plex way than theoretically predicted. Reviewing a number of studies on speech
act realisation, Wolfson (e.g. 1989) concluded that rather than correlating in a
linear fashion, social distance and politeness are related in a reverse bell-shaped
curve (‘Bulge’): most politeness appears to be expended in negotiable relation-
ships with familiars but nonintimates, such as co-workers and friends. In the more
fixed relationships at the opposite ends of the social distance continuum, intimates
and strangers, politeness is found to decrease. More recent evidence for the Bulge
Hypothesis comes from studies on complaining (Olshtain & Weinbach 1993) and
expressions of gratitude (Eisenstein & Bodman 1993).
While there is a comprehensive literature on the impact of social variables on
politeness implementation, much less research exists on the influence of psychological
factors. To some extent, this may simply reflect the fact that demographic variables are
easy to identify whereas (social-)psychological factors are not. Ciliberti (1993) argues
that interactional style is as much a product of participants’ cultural background as of
personality, and an analogous argument can be made for demographic profiles and
personal variables. Slugoski (1985) demonstrates that familiarity (= social distance)
has to be distinguished from affect (= psychological distance), a hypothesis supported
by historical evidence from a study of politeness in Shakespearian tragedies (Brown &
Gilman 1989; on the impact of affect also Boxer 1993).
In addition to these participant variables, features of linguistic acts themselves —
the ‘imposition’, or costs and benefits accruing from them — shape politeness enact-
ment. For several speech acts, the elements of the composite construct ‘imposition’
have been identified, for instance, in
– requesting: urgency, legitimacy (Hermann 1982), likelihood of hearer’s compliance

and speaker’s psychological difficulty in carrying out the request (Blum-Kulka &
House 1989)
Politeness 167
– apologizing: perceived severity of the offence, subsuming obligation to apologise

and likelihood of apology acceptance (Olshtain 1989)
– thanking: indebtedness, comprising the degree of received benefit and trouble
undergone by the benefactor (Ikoma 1993)
– complaining: magnitude of social obligation violated by the offender (Olshtain &
Weinbach 1993).
Participant factors and properties of contextualized linguistic action interact in com-

plex ways and vary crossculturally in their impact on linguistic politeness. For instance,
in request performance, Israeli speakers varied their strategy selection according to
requestive goal, age, and power (Blum-Kulka, Gerson & Danet 1985); Japanese and
German speakers modified their requests according to legitimacy and likelihood of
hearer’s compliance, but the German not the Japanese speakers made their strategy
selection also contingent on urgency (Hermann 1982; Morosawa 1990); Israelis,
Germans, and Argentinians differed in their perceptions of interlocutors’ rights and
obligations, likelihood of hearer’s compliance, and speaker’s difficulty in performing
the request (Blum-Kulka & House 1989); requestees’ obligation to comply was perceived
higher by American than by Japanese raters (Shimamura 1993).
While most studies have examined the impact of context variables on politeness
investment, Brown & Levinson’s theory also predicts a flow of pragmatic information in
the reverse direction: speakers’ choices of politeness strategies inform listeners about the
value of context variables. Perceived imposition in requesting, for instance, was found
to be contingent on actual imposition, request realization, and time of judgement.
Just as the relationship between context variables and politeness varies intra- and
interculturally, so does the relationship between patterns of linguistic action and their
politeness value. Negative politeness strategies were perceived as more polite by Japanese
residing in the US than by Japanese in Japan (Kitao 1990). Japanese and Americans
also gave different appropriateness judgements of requests with and without supportive
moves (Shimamura 1993). A particularly intriguing issue is the relationship between
indirectness and politeness. Contrary to theory derived predictions, it was conventional
indirectness (e.g. preparatory strategies such as ‘can/could you’) rather than nonconven-
tional indirectness (hinting) that was rated most polite by Israeli, American (Blum-Kulka
1987) and German (House 1986) judges. The preference for conventional indirectness
appears to be motivated by the balance struck between clarity and consideration and
in the low processing costs to the hearer. Consistent with this finding, Weizman
(e.g. 1993) has suggested that nonconventional indirectness in requesting is not moti-
vated by politeness at all but by the ‘deniability potential’ inherent in ambiguous
language use. Whether this re-interpretion of indirectness can be sustained across
speech communities is a question for future study.
168 Gabriele Kasper
5. Discourse perspective
Instead of isolating specific context variables and examining their impact on polite-
ness, a large body of literature explores politeness in different discourse contexts,
for instance
– in institutional discourse: courtrooms (Berk-Seligson 1988; Lakoff 1989), medical

practice (Aronsson & Rundström 1989), psychotherapy (Lakoff 1989), academic
advising (Bardovi-Harlig & Hartford 1990), counselling (Frederiksen & Shultz
1982; Fiksdal 1991), supervisory conferences (Robert 1992), special education
conferences (DuFon 1993), classroom discourse (Lörscher & Schulze 1988), lan-
guage proficiency interviews (Ross 1995), consumer service agencies (Johnson &
Fawcett 1987), sermons (Dzameshie 1992), citizen-bureaucracy interaction (Hero
1986), opinion poll interviews (Johnstone, Ferrara, & Bean 1992), church business
meetings (Pearson 1988);
– in workplace communication (Bryan & Gallois 1992) and other professional inter-
action, e.g. aviation discourse (Linde 1988), business negotiations (Yamada 1990),
sales negotiations (Lampi 1993), organisational interaction (Morand 1991), sports
teams (Jones 1992);
– in interpersonal discourse, e.g. family dinners (Blum-Kulka 1990), dinner enter-
tainment (Befu 1986), phatic talk (Coupland, Coupland, & Robinson 1992),
intimate conversation (Frank 1988), interpersonal decision-making (Scheerhorn
1991/92);
– discourse in different media, e.g. telephone conversations (Sifanou 1989), computer
games (Covato 1991), computer messages (Yasuhisa 1991), interpreted discourse
(Knapp-Potthoff 1992);
– written discourse, e.g. letters of request (Cherry 1988), business letters (Hagge &
Kostelnick 1989), narratives (Scollon & Scollon 1981), argumentative writing
(Zammuner 1991), scientific writing (Myers 1989), peer reviews (Johnson 1992).
A general outcome from these studies is that politeness is not only located in the con-
ventions of means and form by which linguistic action is carried out, but also in the
discourse structure. In conversation, politeness is jointly managed by the interlocutors
through activities such as turn-taking, backchanneling, preference organization, and the
principle of recipient design. The ongoing discourse itself constitutes, maintains and alters
participants’ rights and obligations, increases and reduces distance, and strengthens and
weakens affective bonds. Rather than viewing the relationship of discourse context and
politeness as one between dependent and independent variables, it is more appropriately
conceptualized as a dynamically evolving, dialectic interplay of figure and ground.
Politeness 169
6. Further reading
Several journals have published special issues on linguistic politeness: Multilingua 7:4
(1988), 8:2/3 (1989), 12:1 (1993), Journal of Pragmatics 14:2 (1990), 21:5 (1994),
International Journal of the Sociology of Language 27 (1981), 92 (1991). An edited volume
on the topic is Watts, Ide, & Ehlich (1992). For bibliographies on linguistic polite-
ness, see DuFon, Kasper, Takahashi, & Yoshinaga (1994); specifically on politeness
in Japanese: Yoshinaga, Maeshiba, & Takahashi (1992).
References
Aronsson, K. & B. Rundström (1989). Cats, dogs, and sweets in the clinical negotiation of reality: On
politeness and coherence in pediatric discourse. Language in Society 18: 483–501.
Banerjee, J. & P.L. Carrell P.L. (1988). Tuck in your shirt, you squid: Suggestions in ESL. Language
Learning 38: 313–347.
Bardovi-Harlig, K. & B.S. Hartford (1990). Congruence in native and nonnative conversations: Status
balance in the academic advising session. Language Learning 40: 467–501.
Beebe, L.M. & M.C. Cummings (1996 [1985]). Natural speech act data versus written questionnaire
data: How data collection method affects speech act performance. In S.M. Gass & J. Neu (eds.),
Speech acts across cultures: 65–86. Mouton de Gruyter.
Beebe, L.M. & T. Takahashi (1989). Do you have a bag?: Social status and patterned variation in
second language acquisition. In S. Gass, C. Madden, D. Preston & L. Selinker (eds.), Variation
in second language acquisition: Discourse and pragmatics: 103–125. Multilingual Matters.
Beebe, L.M. & T. Takahashi (1989). Sociolinguistic variation in face-threatening speech acts. In
M. Eisenstein (ed.), The dynamic interlanguage: 199–218. Plenum.
Befu, H. (1986). An ethnography of dinner entertainment in Japan. In T.S. Lebra & W.P. Lebra (eds.),
Japanese culture and behavior: 108–120. University of Hawaii Press.
Berk-Seligson, S. (1988). The impact of politeness in witness testimony: The influence of the court
interpreter. Multilingua 7: 411–439.
Bliss, L.S. (1992). A comparison of tactful messages by children with and without language impairment.
Language, Speech, and Hearing Services in Schools 23 (4): 343–347.
Blum-Kulka, S. (1987). Indirectness and politeness in requests: Same or different? Journal of Pragmatics
11:131–146.
——— (1990). You don’t touch lettuce with your fingers: Parental politeness in family discourse. Jour-
nal of Pragmatics 14: 259–288.
Blum-Kulka, S. & J. House (1989). Cross-cultural and situational variation in requestive behavior. In
S. Blum-Kulka, J. House & G. Kasper (eds.): 123–154.
Blum-Kulka, S., J. House, & G. Kasper (1989). Investigating cross-cultural pragmatics: An introductory
overview. In S. Blum-Kulka, J. House & G. Kasper (eds.): 1–34.
Blum-Kulka, S., J. House & G. Kasper (eds.) (1989). Cross-cultural pragmatics. Ablex.
Blum-Kulka, S., B. Danet, & R. Gerson (1985). The language of requesting in Israeli society. In
J. Forgas (ed.), Language and social situation. Springer.
170 Gabriele Kasper
Boxer, D. (1993). Social distance and speech behavior: The case of indirect complaints. Journal of
Brown, P. & S.D. Levinson (1978). Politeness: Some universals in language usage. In E.N. Goody
(ed.), Questions and politeness: Strategies in social interaction: 56–289. Cambridge University
Press.
——— (1987). Politeness: Some universals in language usage. Cambridge University Press.
Brown, R. & A. Gilman (1989). Politeness theory and Shakespeare’s four major tragedies. Language
in Society 18: 159–212.
Bryan, A. & C. Gallois (1992). Rules about assertion in the workplace: Effect of status and message
type. Australian Journal of Psychology 44 (1): 51–59.
Cherry, R.D. (1988). Politeness in written persuasion. Journal of Pragmatics 12: 63–81.
Ciliberti, A. (1993). The personal and the cultural in interactive styles. Journal of Pragmatics
20: 1–25.
Cook, H.M. (1996a). The use of addressee honorifics in Japanese elementary school classroom. In
N. Akatsuka, S. Iwasaki & S. Strauss (eds.), Japanese/Korean Linguistics 5: 67–81.
——— (1996b). Japanese verbal suffixes as indicators of distance and proximity. In M. Pütz & R. Dir-
ven (eds.), The construal of space in language and thought: 3–27. Mouton de Gruyter.
Coulmas, F. (1981). ‘Poison to your soul’. Thanks and apologies contrastively viewed. In F. Coulmas
(ed.): 273–288.
——— (ed.) (1981). Conversational Routine. Mouton.
——— (1981). Routine im Gespräch. Zur pragmatischen Fundierung der Idiomatik. Athenaion.
——— (1992). Linguistic etiquette in Japanese society. In R.J. Watts, S. Ide & K. Ehlich (eds.):
299–323.
Coupland, J., N. Coupland & J.D. Robinson (1992). “How are you?”: Negotiating phatic communion.
Language in Society 21 (2): 207–230.
Coupland, N., J. Coupland, & H. Giles (1991). Language, society and the elderly. Blackwell.
Covato, L.G. (1991). The design of an adventure game authoring tool for exploring polite requests in
an English as Second Language context. Dissertation Abstracts International 53: 430-A.
DuFon, M.A. (1993). Referential and relational meaning in interpreted discourse. Journal of Pragmatics
20: 533–558.
DuFon, M.A., G. Kasper, S. Takahashi, & N. Yoshinaga (1994). Bibliography on linguistic politeness.
Dzameshie, A.K. (1992). Motivations for the use of politeness strategies in Christian sermonic
discourse. Dissertation Abstracts International 53: 1143-A.
Edmondson, W.J. & J. House (1981). Let’s talk and talk about it: A pedagogic interactional grammar of
English. Urban & Schwarzenberg.
Eisenstein, M. & J. Bodman (1993). Expressing gratitude in American English. In G. Kasper &
S. Blum-Kulka (eds.): 64–81.
Færch, C. & G. Kasper (1989). Internal and external modification in interlanguage request realization.
In S. Blum-Kulka, J. House, & G. Kasper (eds.): 221–247. Ablex.
Fiksdal, S. (1991). The right time and pace: A microanalysis of cross-cultural gatekeeping interviews. Ablex.
Frank, J. (1988). A comparison of intimate conversations: Pragmatic theory applied to examples of
invented and actual dialog. The SECOL Review 12: 186–208.
Fraser, B. (1990). Perspectives on politeness. Journal of Pragmatics 14: 219–236.
Fukushima, S. & Y. Iwata (1987). Politeness strategies in requesting and offering. JACET Bulletin
18: 31–48.
Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. Anchor Books.
Politeness 171
——— (1971). Relations in public. Penguin.

Green, G. (1992). The universality of Gricean accounts of politeness: You gotta have wa. Unpublished
ms. University of Illinois.
Grice, H.P. (1975). Logic and Conversation. In P. Cole & J.L. Morgan (eds.), Syntax and semantics
Vol. 3: Speech acts: 41–58. Academic Press.
Gu, Y. (1990). Politeness phenomena in modern Chinese. Journal of Pragmatics 14: 237–257.
Hagge, J. & C. Kostelnick (1989). Linguistic politeness in professional prose: A discourse analysis of
auditors’ suggestion letters with implications for business communication pedagogy. Written
Communication 6: 312–339.
Held, G. (1989). On the role of maximization in verbal politeness. Multilingua 8: 167–206.
Herbert, R.K. & H.S. Straight (1989). Compliment-rejection versus compliment-avoidance: listener-based
versus speaker-based pragmatic strategies. Language and Communication 9: 35–47.
Hero, R. (1986). Citizen contacting and bureaucratic treatment-response in urban government:
Some further evidence. Social Science Journal 23: 181–187.
Herrmann, T. (1982). Language and situation: The pars pro toto principle. In C. Fraser & K.R. Scherer
(eds.), Advances in the social psychology of language: 123–158. Cambridge University Press.
Holmes, J. (1988). Paying compliments: A sex-preferential positive politeness strategy. Journal of
House, J. & G. Kasper (1981). Politeness markers in English and German. In F. Coulmas (ed.):
157–185.
Ide, S. (1989). Formal forms and discernment: Two neglected aspects of linguistic politeness.
Multilingua 8: 223–248.
Ikoma, T. (1993). “Sorry for giving me a ride”: The use of apologetic expressions to show gratitude in
Japanese. Unpublished MA thesis, University of Hawaii at Manoa.
Janney, R.W. & H. Arndt (1992). Intracultural tact versus intercultural tact. In R.J. Watts, S. Ide &
K. Ehlich (eds.): 21–41.
Johnson, D.M. (1992). Compliments and politeness in peer-review texts. Applied Linguistics 13 (1): 51–71.
Johnson, M.D. & S.B. Fawcett (1987). Consumer-defined standards for courteous treatment by service
agencies. Journal of Rehabilitation 53: 23–26.
Johnstone, B., K. Ferrara & J.M. Bean (1992). Gender, politeness, and discourse management in
same-sex and cross-sex opinion-poll interview. Journal of Pragmatics 18 (5): 405–430.
Jones, K. (1992). A question of context: Directive use at a Morris team meeting. Language in Society
21 (2): 427–445.
Kasher, A. (1986). Politeness and rationality. In J.D. Johansen & H. Sonne (eds.), Pragmatics and
linguistics. Festschrift for Jacob L. Mey on his 60th birthday 30th October 1986: 103–114. Odense
University Press.
Kasper, G. & S. Blum-Kulka (eds.) (1993). Interlanguage pragmatics. Oxford University Press.
Kitao, K. (1990). A study of Japanese and American perceptions of politeness in requests. Doshisha
Studies in English 50: 178–210.
Knapp-Potthoff, A. (1992). Secondhand politeness. In R.J. Watts, S. Ide & K. Ehlich (eds.): 203–218.
Kwarciak, B.J. (1993). The acquisition of linguistic politeness and Brown and Levinson’s theory.
Multilingua 12 (1): 51–68.
Lakoff, R.T. (1973). The logic of politeness, or, minding your p’s and q’s. In C. Corum, T.C. Smith-Stark &
A. Weiser (eds.), Papers from the ninth regional meeting of the Chicago Linguistic Society: 292–30.
Chicago Linguistic Society.
——— (1979). Stylistic strategies within a grammar of style. New York Academy of Science Annals
321: 51–78.
172 Gabriele Kasper
——— (1989). The limits of politeness: Therapeutic and courtroom discourse. Multilingua 8: 101–130.
Lampi, M. (1993). Discourse organization and power: Towards a pragmatics of sales negotiations. In
L.F. Bouton & Y. Kachru (eds.), Pragmatics and language learning vol. 4: 195–208. University of
Illinois at Urbana-Champaign.
Leech, G. (1977). Language and tact. Linguistic Agency University of Trier.
——— (1983). Principles of Pragmatics. Longman.
Linde, C. (1988). The quantitative study of communicative success: Politeness and accidents in aviation
discourse. Language in Society 17: 375–399.
——— (1988). Who’s in charge here? Cooperative work and authority negotiation in police helicopter
missions. Paper presented at the annual ACM conference on computer supported collaborative
work. (ERIC Document Reproduction Service No. ED301038).
Lörscher, W. & R. Schulze (1988). On polite speaking and foreign language classroom discourse.
International Review of Applied Linguistics in Language Teaching 26: 183–199.
Mao, L.R. (1994). Beyond politeness theory: ‘Face’ revisied and renewed. Journal of Pragmatics
21: 451–486.
Markus, H.R. & S. Kitayama (1991). Culture and the self: Implications for cognition, emotion, and
motivation. Psychological Review 98: 224–253.
Matsumoto, Y. (1988). Reexamination of the universality of face: Politeness phenomena in Japanese.
Morand, D.A. (1991). Power and politeness: A sociolinguistic analysis of dominance, deference, and
egalitarianism in organizational interaction. Diss. Abstracts International 52: 2207-A.
Morosawa, A. (1990). Intimacy and urgency in request forms of Japanese: A psycholinguistic study.
Sophia Linguistica 28: 129–143.
Myers, G. (1989). The pragmatics of politeness in scientific articles. Applied Linguistics 10: 1–35.
Olshtain, E. (1989). Apologies across languages. In S. Blum-Kulka, J. House, & G. Kasper (eds.):
155–173.
Olshtain, E. & A. Cohen (1983). Apology: A speech act set. In N. Wolfson & E. Judd (eds.), Sociolinguistics
and language acquisition: 18–35. Newbury.
Olshtain, E. & L. Weinbach (1993). Interlanguage features of the speech act of complaining. In
G. Kasper & S. Blum-Kulka (eds.): 108–122.
Pearson, B. (1988). Power and politeness in conversation: Encoding of face-threatening acts at church
business meetings. Anthropological Linguistics 30(1): 68–93.
Rintell, E. (1981). Sociolinguistic variation and pragmatic ability: A look at learners. International
Journal of the Sociology of Language 27: 11–34.
Roberts, J. (1992). Fact-threatening acts and politeness theory: Contrasting speeches from supervisory
conferences. Journal of Curriculum and Supervision 7: 287–301.
Rosenberger, N.R. (1989). Dialectic balance in the polar model of self: The Japan case. Ethos 17: 88–113.
Ross, S. (1995). Accommodation in oral proficiency interviews. Ph.D. Diss, University of Hawai’i at
Manoa.
Scheerhorn, D.R. (1990). Minding your P’s and Q’s in interpersonal decision-making: The politeness
of messages as a consequent of decisional and relational goals. Dissertation Abstracts International
51: 344-A.
——— (1991). Politeness in decision-making. Research on Language and Social Interaction 25: 253.
Scollon, R. & S. Scollon (1981). Narrative, literacy and face in interethnic communication. Ablex.
——— (1983). Face in interethnic communication. In J.C. Richards & R.W. Schmidt (eds.), Language
and communication. Longman.
Politeness 173
Shimamura, K. (1993). Judgement of request strategies and contextual factors by Americans and
Japanese EFL learners. MA diss. University of Hawaii at Manoa.
Sifianou, M. (1989). On the telephone again! Differences in telephone behaviour: England versus
Greece. Language in Society 18: 527–544.
Slugoski, B.R. (1985). Grice’s Theory of Conversation as a Social Psychological Model. Ph.D. Diss.
Oxford.
Sornig, K. (1977). Disagreement and contradiction as communicative acts. Journal of Pragmatics
1: 347–374.
Takahashi, T. & L.M. Beebe (1993). Cross-linguistic influence in the speech act of correction. In
G. Kasper & S. Blum-Kulka (eds.): 138–157.
Watts, R.J. (1989a). Relevance and relational work: Linguistic politeness as politic behavior. Multilingua
8: 131–166.
——— (1989b). Taking the pitcher to the ‘well’: Native speakers’ perceptions of their use of discourse
markers in conversation. Journal of Pragmatics 13 (1): 203–237.
Watts, R.J., S. Ide & K. Ehlich (1992). Introduction. In R.J. Watts, S. Ide & K. Ehlich (eds.): 1–17.
Watts, R.J., S. Ide & K. Ehlich (eds.) (1992). Politeness in language: Studies in its history, theory and
practice. Mouton de Gruyter.
Weizman, E. (1993). Interlanguage requestive hints. In G. Kasper & S. Blum-Kulka (eds.): 123–137.
Wierzbicka, A. (1985). Different cultures, different langauges, different speech acts. Journal of
Wolfson, N. (1989a). Perspectives: Sociolinguistics and TESOL. Newbury House.
———(1989b). The social dynamics of native and nonnative variation in complimenting behavior. In
M. Eisenstein (ed.), The dynamic interlanguage: 219–236. Plenum.
Yamada, H. (1990). Topic management and turn distribution in business meetings: American versus
Japanese strategies. Text 10: 271–295.
Ye, L. (1993). Complimenting in Mandarin Chinese. MA Diss. University of Hawaii at Manoa.
Yoshinaga, N., N. Maeshiba & S. Takahashi (1992). Bibliography on Japanese pragmatics. In G. Kasper
(ed.), Pragmatics of Japanese as native and target language: 1–26. Second Language Teaching &
Curriculum Center, University of Hawaii at Manoa.
Zammuner, V.L. (1991). Children’s writing of argumentative texts: Effects of indirect instruction.
European Journal of Psychology of Education 6 (2): 243–256.
Prosody
Elizabeth Couper-Kuhlen
1. Prosody defined
The term prosody is current both in the fields of literature and of linguistics. In liter-
ary studies it is used to refer to the metrical analysis of verse, while in linguistics its
denotation encompasses not only stress- and accent-related phenomena in language
use, but also diverse types of melodic or pitch-related, and dynamic or loudness-related
speech modulation. In addition, it is used for the temporal manipulation of syllables
and segments — whether with respect to their duration (stretching and clipping) or
with respect to their timing (rhythm and rate of speech). Prosodic features create audi-
tory effects in spoken language above and beyond those which identify the sound seg-
ments themselves. They are in one sense distinct to speech, but of course also virtually
present in many written forms of verbal communication.
However easy it may be to list single prosodic phenomena, it is no simple matter to
establish the upper and lower bounds of the category ‘prosodic’. Nor can this be done in
a language-independent fashion. At the lower end it is necessary to exclude phenomena
such as, for instance, lexically and/or syntactically determined pitch accents and tones. The
label ‘prosodic’ applies only to those suprasegmental auditory effects in speech which, in
Crystal’s words, have “an essentially variable relationship to the words selected” (1969: 5).
Crystal argues, on the other hand, that not all lexically variable auditory effects in speech
qualify as prosodic. He would restrict the category to only those auditory effects which
can be traced to the dimensions of pitch, loudness and duration, and whose domain is
minimally the syllable. Such features, he claims, are constantly present in speech, whereas
other auditory effects — such as, for instance, giggling (a ‘voice qualification’ in his termi-
nology) or whispering (a ‘voice qualifier’) — are not continuous with speech at all times
and therefore do not count as prosodic. One disadvantage of taking such a restrictive view
is that it relegates pauses to the margins of the category. Yet the manipulation of pausing
is generally recognized as an eminently prosodic phenomenon.
Other scholars have taken a more inclusive view. Firth (1957), for instance,
called all types of syntagmatic relations holding between syllables prosodies, includ-
ing syllable structure and number or nature of the syllables, e.g. as concerns stress,
tone, quantity or quality. In this more encompassing view, certain of Crystal’s ‘voice
qualifiers’ — e.g. breathiness, creak, nasalization, whisper — would also qualify
as prosodic, provided they are independent of the phonemic make-up of the single
Prosody 175
sounds. Indeed, the fact that such vocal devices are often deployed in speech to the
same effect as, say, rising pitch, increased volume or syllable stretching, argues in favor
of grouping them together — as early work did (Trager 1958). For present purposes,
the label ‘prosodic’ will be used in this more encompassing, paralinguistic sense.1
2. Prosody as a pragmatic phenomenon
It is a commonplace to say that the way we speak conveys as much meaning as the words
we use. Much of this additional meaning can be ascribed to prosodic effects in speech.
Yet how does prosody convey what it does? In the structuralist tradition it has been cus-
tomary to draw an analogy between prosodic phenomena and meaning-distinctive units
such as phonemes and/or meaning-bearing units such as morphemes. Pike (1945), for
instance, argued that pitch levels in American English distinguish meaning in much the
same way as phonemes do in minimal words. Other linguists have seen a parallel between
holistic intonational contours or melodies and ‘discourse’ meanings such as contradic-
tion or surprise, making prosodic patterns more similar to morphemes (Liberman & Sag
1974; Sag & Liberman 1975). Followers of this line of thinking speak of an ‘intonational
lexicon’, in which the semantics of particular tones or tone sequences can be represented
in a fashion similar to that for morphemes and words. Yet this view — if feasible at all —
applies to only a very small part of prosody, namely to a restricted set of tonal configura-
tions for single utterances. Larger-scale, non-compositional pitch patterns, e.g. at the level
of declination unit (Schuetze-Coburn, Shapley & Weber 1991) or paratone (Yule 1980;
Couper-Kuhlen 1983), cannot be captured. Moreover, dynamic and temporal variations
including e.g. rhythm, tempo and pause (see e.g. Auer, Couper-Kuhlen & Müller 1999)
are not amenable to the lexical analogy and would have to be disregarded altogether.
Prosodic phenomena are particularly recalcitrant when it comes to establishing dis-
crete categories, as is necessary in the structuralist enterprise. Bolinger was acutely aware
of this and argued persuasively against a view of intonation entailing discrete phoneme-
like pitch levels and predictable (i.e. semantically determined) accent placement (1965,
1972a). Instead, he emphasized that intonation is in important ways a gradient phenom-
enon: “whatever it is that we find with the raised […] peak, there is more of it as that peak
goes higher” (1961: 43). Bolinger compares intonation to gesture and calls it gesturing
with one’s larynx (1986). There is a great deal of insight in this remark, as we shall see.
It also transfers nicely to other aspects of prosody such as rhythm or pause, where —
although an organic source may be lacking — a gestural function is nevertheless discern-
ible. In short, the best answer to the question of how prosody conveys what it does is that
1. See also Kelly & Local (1989), who advocate the wider, Firthian denotation of ‘prosodic’ and
give it the general label ‘phonetics’.
176 Elizabeth Couper-Kuhlen
it does so in a fashion similar to mimetics, gesture and kinesics — all dimensions which
reflect the embodiment of language when used in actual (face-to-face) communication.
If one’s aim is to come to terms with prosodic phenomena as they function in
genuine language communication, the structuralistic framework is thus of limited
value (see also Couper-Kuhlen & Selting 1996). This is because, semiotically speaking,
prosodic effects, like gestures, are not symbolic signs but are rather indexical, and on
occasion iconic, signals. As indexicals, prosodic signals function pragmatically, cueing
inferences rather than symbolically representing aspects of extra-linguistic reality. In
contrast to the symbolic signs of language (words or morphemes), the meanings which
prosodic signals cue cannot be specified out of context. In fact, prosodic effects never
appear without words or word-like entities to carry them. These words and the dis-
course situation in which they are used provide a context for interpreting the prosody
which accompanies them — just as, vice versa, the prosody itself provides a context for
interpreting its carrier, the words. The relation between prosody and its context is thus
a reflexive one, as we might expect in general of indexical signs.2
Because prosodic signals hint at possible ways of understanding what is being said,
they have been called contextualization cues for language (Gumperz 1982; Auer 1986;
Auer & di Luzio 1992). As contextualization cues, one of their significant features is that
the interpretative frames they call up are open to negotiation in interaction. For instance,
although one interpretive frame may be cued by the speaker, the recipient may choose
to interpret according to another. What is ultimately understood to have been said is the
product of negotiating which frame is to be taken as the relevant one for the talk in ques-
tion. Related to this is a second feature of prosodic signals: they are not ‘accountable’ in
the same way that words are (Garfinkel 1967). Speakers can be held responsible for (i.e.
criticized, blamed, asked to apologize for, etc.) their choice of words, but it is difficult to
take them to task for their prosody. Someone who asks a friend in a complaining voice
‘Why do you leave the fridge door open all the time?’, when challenged, can always escape
blame by claiming ‘I’m not criticizing, I’m just asking’ (Günthner 1996). As Silverstein has
put it, prosody is often “beyond the limits of pragmatic awareness” (1976).
3. Prosody and early work on spoken discourse
Within modern linguistics, the pragmatic nature of prosody, in particular intona-

tion, was not recognized until relatively late.3 Among the first to do so was Halliday
2. Although they are indexical, prosodic signs are also partially conventionalized, as becomes
apparent from cross-cultural comparison (Gumperz 1982). This is what makes a systematic descrip-
tion of their inference-cueing potential in specified situations of use feasible.
3. Anthropologists and scholars of oral poetry had been appealing to the prosodic dimensions of
performance for longer (cf. Lord 1964; also, more recently, Tedlock 1977 & Woodbury 1985.)
Prosody 177
(1967, 1970), who saw in intonational distinctions such as tonality (division into tone-
groups), tonicity (placement of the tonic accent) and tone a means for speakers to
chunk information into units, to mark information foci and to distinguish new from
given information, respectively. At about the same time, Bolinger was warning against
the predictability of accent on syntactic grounds (1972a) and claiming that because of
its gradience and essentially expressive function intonation was ‘around the edge of
language’ (1972b).
A second wave of work on prosody, especially intonation, as a pragmatic phe-
nomenon was initiated when linguists enlarged their perspective from the sentence to
larger units of discourse. Yule (1980), for instance, identified on prosodic grounds a
phonological unit larger than the intonation phrase (often seen as corresponding to a
clause; cf. Crystal 1975) which he dubbed, in analogy to the paragraph, ‘paratone’ (see
also Couper-Kuhlen 1986: 189–200).4 Brazil, Coulthard & Johns (1980) also described
intonational contrasts beyond the intonation phrase and identified patterns of pitch-
level ‘harmony’ within classroom exchanges, which they then postulated as constraints
applying to the production of asymmetrical discourse in general.
At about the same time but in a different part of the world, Chafe was examining
the use of prosody in the organization of oral narrative and in particular of the Pear
Stories (1980). Chafe sees intonation as providing a window on consciousness via the
establishment of the intonation unit, which encompasses the information that is in
the speaker’s focus of consciousness at a given moment, and the accent unit, which
is the domain of activation for new, accessible and/or given information (1993). Also
within this tradition, DuBois et al. (1992, 1993) postulate a notion of transitional con-
tinuity between one intonation unit and the next, marked by different sorts of terminal
pitch contours. The term transitional continuity describes the extent to which “the
discourse business at hand will be continued or has finished” (1993: 53).
While all of the above work addresses intonation, or more generally prosody, at the
level of discourse, it tends to do so almost exclusively in terms of information. Moreover,
with the exception perhaps of Brazil, the focus is heavily on spoken text and monologic
discourse. To redress the balance, the following discussion will be geared primarily to
the pragmatics of prosody in more interactive forms of discourse. As conversation anal-
ysis has developed a methodology especially adapted to talk-in-interaction, the discus-
sion will be framed in this model.
4. Major paratones are said to begin with an extra high pitch on the first stressed syllable of an
intonation phrase and to end with extra low pitch and a noticeable pause at the end of a subse-
quent intonation phrase. They are thus roughly equivalent to what Schuetze-Coburn et al. (1991)
call ‘declination unit’.
4. Prosody in talk-in-interaction: Structural dimensions
What sorts of interpretations does prosody cue in interaction? Research so far has
suggested that it is implicated in both structural and interactional dimensions of talk-
in-interaction. Structurally speaking, participants in interaction construct minimal
and extended turns at talk; they organize successive turns at talk sequentially; and
they manage rights to the floor, all in systematic ways. Prosody can be seen as one of
the resources at their disposal for handling the tasks of turn construction, sequential
organization and floor management.
4.1 Turn construction

Consider the turn-constructional unit (TCU): this is the smallest unit which can form a
turn at talk (Schegloff 1996). According to Sacks, Schegloff & Jefferson (1974), TCUs are
constituted by different sorts of syntactic constituents, including words, phrases, clauses,
sentences. Now, since these syntactic constituents are hierarchically ordered with respect
to one another, it is not immediately obvious — on syntactic grounds alone — whether,
say, a single word is a TCU itself or instead only part of a phrase or clause which consti-
tutes the TCU. Yet, as recent research has shown, intonation helps resolve this problem.
It is melodic phrasing which contributes to deciding what piece of potentially complete
syntax is indeed a minimal unit of turn construction. A morpho-syntactically consti-
tuted lexeme, phrase or clause which can stand on its own is typically not considered
transition-relevant unless it is configured with a potentially complete intonation contour
(Ford & Thompson 1996; Ford, Fox & Thompson 1996; Selting 1996a; Schegloff 1998).
Prosody is also implicated in the construction of multi-unit turns. When several
TCUs make up a turn, they are typically formatted in a way which reflects their posi-
tion in the turn (see also Schegloff 1998). For instance, a TCU will be displayed as a
follow-up to a prior TCU if its first accented syllable, or onset, is lower than the onset
of the prior TCU. Signalling prosodic ‘subordination’ this way appears to be used in
interdependence with whatever ‘rhetorical’ relations — e.g. causality, concession, con-
trast — may hold between the TCUs in question (Ford 1993; Couper-Kuhlen 1996a;
Couper-Kuhlen & Thompson 1999; Ford 2000). Further, the fact that a particular TCU
is being constructed as the first in a set of TCUs can be displayed, for instance, by
positioning its first accented syllable (or onset) at a relatively higher position in the
speaker’s voice range. This strategy is often employed in ‘big packages’, for instance in
reason-for-the-call turns in radio phone-in talk (Couper-Kuhlen 2001).
4.2 Sequential organization

Prosody is furthermore crucial in displaying sequential structure, the relation between
successive turns at talk. For instance, the adjacency pair — a set of two consecutive
Prosody 179
turns, the first of which anticipates and makes conditionally relevant the second — is
characterized by ‘close ordering’: first pair-parts are ‘fitted’ to second pair-parts with
respect to the action being carried out (Schegloff & Sacks 1984). But the notion of ‘close
ordering’ also has prosodic correlates. Seconds to first pair-parts tend to be ‘fitted’ not
only in terms of action but also in terms of the way this action is done. For instance,
they are ‘fitted’ in terms of amplitude, the second pair-part being regularly matched in
volume to the first (Goldberg 1978). Moreover, seconds are typically produced — in
English — with rhythmic integration: the first accented syllable in the second pair-part
is timed so as to coincide with a ‘beat’ which has been established by the regular timing
of the last two accents of the immediately prior first pair-part (Couper-Kuhlen 1993;
Auer, Couper-Kuhlen & Müller 1999). And finally, seconds tend to be ‘fitted’ intona-
tionally, in the sense that the onset of the second pair-part will not be higher in pitch,
ceteris paribus, than the onset of the first pair-part (see §5 below).
But it is not only turns within an adjacency pair which are made cohesive prosodi-
cally. As the work of Schegloff and Fox has shown, adjacency pairs themselves ‘cohere’
in differing degrees with respect to the actions which are being carried out. Schegloff
(1990), for instance, shows how adjacency pairs may be expanded by pre-, insert and
post-sequences. Fox (1987) speaks of adjacency pairs being extended by follow-ups,
and of adjacency pairs which engage in the same activity as being ordered in a series.
One of the ways of cueing that the first pair-part of an adjacency pair is engaging in
the same activity as another is to format it prosodically like this other, for instance by
using identically high onset in its first pair-part (see §5 below). Prosodic parallelism
between adjacency pairs can be iconically suggestive of structural parallelism. Con-
versely, a noticeably lower onset in the first pair-part of an adjacency pair will suggest,
by comparison with the onset of the first in a prior pair, that the two adjacency pairs
are in a hierarchical relationship to one another. This can be found, for instance, in
sequence expansions (see §5 below) and in repair initiation by other.
4.3 Floor management

When speakers wish to place a turn-constructional unit on the floor, they rely on
prosodic cues in order to judge when this can be done legitimately. By monitoring
the prosodic development of the current turn, they can anticipate an upcoming
transition-relevance place (TRP), the moment at which a current speaker’s turn will
be potentially complete. Would-be next speakers of course monitor for a prosodically
complete contour, one of the constitutive features of a TCU. But as Schegloff (1998)
and Wells & Macfarlane (1998) have argued, they also listen for specific pitch accents
which help project when the next TRP is due. In English, for example, a ‘pitch peak’, i.e. a
pitch accent which begins higher than prior syllables and then falls, has been observed
to regularly precede the TRP (Schegloff 1998). According to Wells & Macfarlane
(1998), there is actually a set of distinct TRP-projecting accents in English which
characterize possible turn endings.5 Once such a pitch accent has occurred, a potential
transition space has been opened. Significantly, the notion of a (pitch-initiated) transi-
tion space has a rhythmic or temporal corollary: any overlap which occurs between the
last accented syllable of a current turn and the first accented syllable of an incoming
turn — what Jefferson (1984) has called ‘transitional overlap’ — generally goes unno-
ticed in English interaction. Participants do not orient to there having been a possible
‘interruption’ in these cases (Couper-Kuhlen 1993).
When, on the other hand, speakers have urgent business that needs addressing now
and they come in during an ongoing turn rather than at the next transition relevance
place, then those speakers may be heard as competing with a current speaker for rights
to the floor. As French & Local (1983) have shown, turn-competitive incoming talk is
regularly marked with high pitch and forte volume, cueing the message ‘This is urgent,
I need the floor here and now’. Incoming talk which lacks these prosodic features tends
to be treated as an aside, an ‘illegitimate’ contribution but not one which is compet-
ing for the floor. A speaker whose control of the floor is threatened by an illegitimate
incoming will be observed to react not only verbally — e.g. by abruptly cutting off or
by re-cycling — but also prosodically. French & Local (1983) describe two strategies:
either the current speaker yields to competition for the floor by reducing pitch and
volume, or the current speaker returns the competition by suddenly using higher pitch
and increasing the volume. The latter — a prosodically cued ‘return of competition’ —
tends to last until the floor struggle has been decided one way or the other.
In this section then prosody has been shown to be implicated in three form-related
tasks in conversation: constructing turns, organizing sequences and managing the floor.
5. Prosody in talk-in-interaction: Interactional dimensions
These prosodically cued structural dimensions engage in important ways with more
interactional, or sense-making, dimensions of talk. As speakers we deploy grammar
and prosody in routine ways to carry out particular actions. Moreover, we can designedly
exploit the routine formatting of our actions in order to convey special meanings. As
recipients, we interpret what actions our interlocutors are engaging in with reference
to our expectations for grammatical and prosodic design. If and when our expecta-
tions are foiled, we inferentially impute special meanings to our interlocutors’ talk. The
meanings conveyed by the deployment and exploitation of prosodic routines are more
than structural: they impinge upon our interpretations of what kind of action is being
accomplished and on affect-related dimensions such as attitude and stance.
5. Although these may vary from dialect to dialect, within one single dialect they are said to
contrast with non-TRP-projecting accents, which do not signal upcoming transition relevance.
Prosody 181
5.1 Prosodic routines for action

As resources for carrying out actions in interaction, grammar and prosody are deployed
in routine and typical ways. For instance, interrogative forms are frequently used to do
questioning (Freed 1994; Heritage & Roth 1995), evaluative adjectives are frequently
used to do assessments (Pomerantz 1984), oh is frequently used to receipt question-
elicited informings (Heritage 1984). Moreover, these lexico-syntactic forms have pro-
sodic routines associated with them. For instance, oh as a news receipt tends to be
produced with a dynamic rising-falling pitch accent if a recipient is displaying ‘surprise’
at the informing (Local 1996). It is in particular when the words used to carry out an
activity are functionally indeterminant in context that prosodic cueing becomes a dis-
tinctive feature. Consider the case of self-repair. Fox & Jasperson (1995) have shown
how self-repairing speakers of English rely on syntactic means to mark the repairing
segment as doing just that — repair — and not, say, adding on to prior material. This is
partly accomplished by using disjunctive lexico-syntax (see also Fox, Hayashi & Jasperson
1996). But prosodic signals too are deployed in self-repair to the same effect: abrupt
increases in pitch and volume as well as rhythmic breaks routinely mark the material in
question as replacive rather than additive (Couper-Kuhlen 1993).
Another action for which speakers often deploy routines is complaining. For
instance, Schegloff (1988) has shown that a noticing of something which has not
taken place, in the proper context, will come across as a subtle complaint. One lexico-
syntactic device frequently deployed to do complaining in German and English is the
wh-question. Yet here prosodic cueing is crucial. A question like ‘Why is it not in
the paper?’ with expressive lengthenings and accents realized as rising-falling glides
will be treated in the appropriate context as a complaint, whereas the same question
without such prosodic marking, although in a similar context, will be heard as simply
eliciting information (Günthner 1996). (Extreme formulations such as why on earth,
never or always will of course help to mark the wh-question as a complaint.)
Finally, take the case of lexical repetition. Within the context of dialect survey
interviews, Kelly & Local (1989) show that word repetitions by the interviewee with
high falling pitch to low and marked loudness are treated as understanding checks,
eliciting confirmation tokens from the interviewer, whereas word repetitions by the
interviewee which lack high falling pitch to low and have breathy, soft voice are heard
as cueing that the speaker is ‘mulling over’ the word he or she has been asked to gloss.
In a wholly different context — that of a riddle-guessing game on a radio phone-in —
Couper-Kuhlen (1996b) shows that lexical repetitions by the studio moderator of a
caller’s candidate answer come off as requests for confirmation if they are designed with
similar prosody, i.e. if the moderator uses the same relative pitch register as the caller.
But if the prosody is identical, i.e. the same absolute pitch register, then the moderator
will be heard as mimicking his caller and in a sense laterally assessing the proffered
guess as silly for the overhearing radio audience.
5.2 Prosodic cueing of stance and affect

Often it is relatively clear in a given sequential context which action is being carried
out. But the prosodic design of the turn in question may signal something special
about the way it is being done and thus cue the speaker’s stance towards the mes-
sage or its recipient. Speech rate, for instance, understood as the number of syllables
articulated within a given unit of time, is often deployed as a prosodic cue to relative
importance (Uhmann 1992). Within long turns, turn-constructional units which are
said with a sudden increase in speech rate compared to surrounding units will come
across as less important or less central to a speaker’s goal than ones which lack this
marking. Inversely, turn-constructional units with a noticeable decrease in speech rate
will be heard as the ‘main’ point.
Another action in which prosody is often implicated is other repair initiation.
Extra loudness and high pitch on wie bitte? in German, for instance, will not only
elicit a confirmation of the prior turn from one’s interlocutor but at the same time
tend to express the speaker’s astonishment over it, thus prompting the recipient to
provide some accounting (Selting 1996b). Pausing and rhythmic delay in other repair-
initiation have been shown to provide an extra opportunity for one’s interlocutor to
engage in self-repair (Schegloff, Jefferson & Sacks 1977). But they also ‘expose’ the
trouble source (Jefferson 1987), locating it in the interlocutor’s turn and providing for
interlocutor repair, thus subtly deflecting responsibility (Couper-Kuhlen 1992).
As a final example, consider the way news is responded to in talk-in-interaction.
Here too timing is crucial. A noticeable pause or rhythmic delay on next turn — be
it a news mark, a repeat or a news-receipting oh — will contribute to signalling that
the news was unexpected, a surprise (Auer, Couper-Kuhlen & Müller 1999). (Oh
alone, as Heritage 1984 has shown, does not carry this implication.) Moreover, the
pitch configuration will display whether the news is being receipted as good or bad.
Sharp pitch rises, wide pitch range and a lively speech rate tend to be associated with
receipts which have ‘positive’ valence, while stretched vowels slightly falling in pitch
together with nasalization and creaky or breathy voice tend to cue receipts with
‘negative’ valence (Freese & Maynard 1998).
In this section some of the ways in which prosody contributes to sense-making
in interaction have been discussed, in particular as regards the type of action which is
inferred to be in progress and the stance which a speaker is inferred to be taking.
6. Prosody in talk-in-interaction: A case study
To make some of these observations more concrete, let us examine the way prosody
is deployed in the following excerpt from a long-distance telephone conversation
Prosody 183
between Lesley and her mother.6 The exchange takes place approximately two minutes
into the conversation at a topical juncture within a section of the conversation where
participants are engaged in ‘bringing (each other) up to date’ since they last talked.
Ann lives in the same city as Lesley’s mum; Gordon is Lesley’s son.
(1) Holt I:1
1 Lesley: .hh (oh)=↑By the wa:y ANN hasn’t:: sent gordon anything.=
2 Mum: =yes she HA:S.
3 Lesley: well it hasn’t CO:ME,
4 Mum: oh well probably get it TOMORROW.
5 (0.3)
6 Lesley: OH: yes.= ((late))
7 =↑What is she SENDING,
8 do you KNOW,
9 (.)
10 Mum: (uh YES./well GUESS.)
11 (0.3) ((empty beat))
12 Lesley: oh Not one of those whi:te TEE shirts.=
13 Mum: =well i don’t know what sort of tee shirt it IS,
14 but it I:S.
15 Lesley: 〈l〉 OH:: mu::m::. ((slower))
16 (0.4) ((empty beats))
17 Lesley: it’s just a waste of MONEY.
Focusing first on turn construction, notice how each TCU corresponds to a single intona-
tion phrase7 (indicated by a separate line in the transcript). Words that could on principle
be TCUs — for instance, yes in line 2 or oh well in line 4 — are not heard as such in these
instances because they do not form independent pitch contours or intonation phrases of
their own. Yet there is nothing intrinsic which prevents them from being designed this
way, as line 6 demonstrates. Here oh yes is a TCU of its own; it is responsive to Mum’s prior
turn in line 4, whereas the next TCU what is she sending is the first part of an adjacency
pair projecting a next turn from Mum. Notice that the second TCU in this turn has high
onset (indicated by the upwards arrow at the beginning of line 7). In this case the second
6. This excerpt was transcribed by Gail Jefferson. To facilitate reading,‘eye dialect’ has been converted
to standard orthography. Furthermore, selected prosodic symbols have been replaced by the author —
after multiple listenings to the original tape — according to the list of conventions appended.
7. An intonation phrase is generally taken to be a unit of prosodic phrasing which displays inter-
nal rhythmic and melodic cohesion, with e.g. its pitch accents being regularly timed and forming
a holistic melody or contour. At the same time it is said to be delimited by rhythmic and melodic
breaks, with e.g. pauses, anacrustic syllables and pitch step-ups on unaccented syllables marking its
boundaries. For further details see Couper-Kuhlen (1986) and Cruttenden (1997).
TCU is designed in a way which suggests that it is not carrying on from the prior but is
instead engaging in a new activity. Compare this prosody now with that in Mum’s multi-
unit turn in lines 13–14. Here the second TCU but it is is designed so that its onset is no
higher than that of the prior TCU (it lacks an upwards arrow). This prosody underlines
the rhetorical relation of contrast between the two TCUs (see §3.1 above).
With respect to sequential organization, notice how successive turns are ‘fitted’
prosodically to one another. ‘Close ordering’ generally translates rhythmically as a lack
of empty ‘beats’ between turns.8 And indeed this is the case at all transitions between
Lesley and her mother except for two. The first rhythmic break comes in line 5, indicated
in transcription as a 0.3 sec. pause. Mum’s prior turn has established a regular beat
with the secondary accents on oh and probably and the primary accent on tomorrow.
Yet Lesley’s oh in line 6 is late with respect to the next pulse. In this case, where close
sequential ordering makes close timing expectable, the lack of rhythmic integration
cues special inferencing (see below). The second rhythmic break occurs in line 11 after
Mum’s next turn. Once again an empty beat goes by unfilled. On the reading of line 11 as
uh yes, Mum could be expected to elaborate on what Ann’s present is, and the next
beat is when she should do so. But this beat is empty — a missed opportunity. It may
be because of this that Lesley now realizes Mum is not going to be forthcoming with
details and so proceeds to proffer a guess herself (line 12).
Next, consider how successive adjacency pairs are designed prosodically. One
noticeable feature of Lesley’s first turn in this excerpt (line 1) is the extra high pitch on by
the way. The expression by the way has been called a ‘misplacement marker’ (Schegloff &
Sacks 1984): it does the work of signalling that an upcoming topic proffer is not moti-
vated by anything in prior talk. The high onset here — together with the relatively slow
rate of delivery — can be heard as a cue to the fact that the misplaced topic is being
proffered as a full-fledged contribution to the floor rather than as an aside, an implica-
tion which low onset and fast speech rate, by contrast, would have. Significantly, Lesley’s
next first pair-part (line 3) has lower onset than the prior topic-opening first pair-part
in line 1. This is a rather iconic signal of the fact that her next action is a follow-up to,
and thus in a sense dependent on, her prior action. By contrast, Lesley’s subsequent
what is she sending in line 7 — by virtue of its high onset — comes across as an action
on a par with that in line 1: the two adjacency pairs are thus cued as forming a series.
Finally, consider how interactional meaning is cued prosodically in the excerpt
above. Lesley’s first turn Ann hasn’t sent Gordon anything is, on one possible reading, an
informing action: it claims to tell Mum about something which Lesley knows, which
8. A beat is to be understood with reference to the regular or rhythmic timing of accented syl-
lables in talk. A rhythmic pattern, once established, tends to project continuation. If, however, a next
accented syllable does not come at the projected point in time, an ‘empty’ or silent beat is created.
For further details see Auer, Couper-Kuhlen & Müller (1999).
Prosody 185
Mum is assumed to be ignorant of but which Lesley believes Mum should know. Mum’s
response makes it clear not only that she does have knowledge about Ann’s sending a gift
but furthermore that she knows something different from Lesley, namely that Ann has
sent one (line 2). Lesley’s next turn is, on this same reading, another informing: well it
hasn’t come (line 3). Mum’s response oh well probably get it tomorrow (line 4) reconciles
Lesley’s information with Mum’s state of knowledge. On another reading, however,
Lesley’s informings are noticings, and indeed noticings of things which have not hap-
pened. In this sense they are hearable as complaints (see §4.1 above). Moreover, the fact
that the complaints are being addressed to Mum potentially implicates her as a responsi-
ble party. On this second reading, Mum’s responses can be heard as removing the grounds
for Lesley’s complaint about Ann. Moreover, if they are heard this way, then it is worth
noting that Mum carries out this action from a position of alignment with Ann. (This can
be seen by comparing other possible responses from Mum. Had she said, for instance, ‘Yes
I know, it’s awful — she’s always forgetting birthdays but believe or not, this time she actu-
ally remembered’, then Mum would be perceived as more in line with Lesley.)
Now it is the prosody of Lesley’s response in next turn (line 6) — rhythmic delay
and a low-pitched, slow downwards glide — which is a cue to her sense-making of
prior talk. Indeed, her delivery of oh is consistent less with the interpretation of prior
talk as an informing sequence, in which case we might expect a more enthusiastic
response to the ‘good news’ — than with an understanding of prior talk as a complain-
ing sequence. If Mum is heard as removing the grounds for a complaint from Lesley —
and indeed in alignment with Ann — then a prosodically cued ‘begrudging’ oh from
Lesley is fully consistent. Thus, prosody displays here what actions are being carried
out in talk — and indeed from participants’ (here, Lesley’s) perspective.
The subsequent development of talk confirms Lesley’s and Mum’s alignment with their
respective roles of ‘complainer’ and ‘supporter of Ann’. When Lesley now inquires what it is
that Ann is sending and appends a post-completer do you know, Mum — on one hearing
— responds to the post-completer: uh yes (line 10).9 But she also hearably withholds any
further information (see the empty beat in line 11), thus displaying once again a sensitivity
to Ann’s point of view as a gift-giver who is likely to want her gift to remain a surprise. This
prompts Lesley to proffer a ‘best guess’, expressed via the negation as a worst case: not one
of those white tee shirts (line 12). Since Mum is a knowing recipient, Lesley’s proffered guess
makes some answer on Mum’s part the next relevant action. Mum’s response once again
displays a sensitivity to Ann’s concerns by offering only minimal information: she claims
not to know what kind it is but concedes that it is a tee shirt (lines 13–14).
Lesley’s next turn oh mum is verbally indeterminate, but its prosodic design leaves
no doubt as to her stance. With low register, slow delivery, extreme lengthening on
9. The alternative hearing well guess, which Jefferson opts for, is to my ears less likely but does not
invalidate the analysis being given here.
both syllables and a downwards glide on oh, it cues the informing as ‘bad news’, sug-
gesting a negative reaction on Lesley’s part. By contrast, a high, sharp pitch jump on
oh in conjunction with quick delivery would have cued the news as ‘good’ and Lesley’s
reaction as enthusiastically positive. When Mum now withholds a response to Lesley’s
prior ‘appeal’ (see the empty beats in line 16), Lesley makes her stance concrete in next
turn: it’s just a waste of money (line 17).
The above analysis demonstrates how crucial the prosodic details of interaction
are for an understanding of participants’ sense-making in talk. It documents the fact
that the prosodic features of talk are not accidental but are designed and that their
decipherment, via careful application of the tools of conversation analysis, will enrich
our understanding of interaction.
7. Directions for future research
Among some of the more promising directions for future research are those dealing
with prosody in first and second language acquisition and in language-impaired com-
munication. Intonational patterns appear to be acquired extremely early in infant-
hood (see Crystal 1975 for an overview of the literature) — indeed before language,
as the phenomenon of ‘jargon intonation’ indicates (Cruttenden 1997: 167). Yet chil-
dren do not acquire the ability to produce reduced syllables and stress-timed rhythm
until much later: Allen & Hawkins (1980) found English-speaking children as old as
four still having difficulty using light or reduced syllables properly. Once prosody is
acquired, however, it enters the realm of meta-pragmatic unawareness, with the result
that its contribution to successful communication is often overlooked or neglected
in second language teaching and learning. Prosodically based misunderstandings are
consequently at the root of many problems in intercultural communication (Gumperz
1982). A ‘first in, last out’ phenomenon, prosody often remains intact even after severe
language loss (aphasia) — where, together with other gestural and body semiotics, it
can be crucial to the (co-)construction of meaning in interaction (Goodwin 1995).
Appendix: Selected transcription conventions
1 line of transcript 1 intonation phrase

(oh) Uncertain hearing
(uh yes/well guess) Alternatice hearings
= Latching
:: Lengthening
(.) Micropause
(0.3) Measured pause
Prosody 187
CAPS Major accented word

. Final pitch falling to low
, Final pitch rising to mid
≠ High pitch onset
<l> Low overall pitch
.hh Inbreath
(( )) Transcriber’s comment
References
Allen, G.D. & S. Hawkins (1980). Phonological rhythm: Definition and development. In
G.H. Yeni-Komshian, J.F. Kavanaugh & C.A. Ferguson (eds.), Child Phonology, Vol. 1: 227–25.
Academic Press.
Atkinson, M.J. & J. Heritage (eds.) (1984). Structures of Social Action. Studies in Conversation Analysis.
Auer, P. (1986). Kontextualisierung. Studium Linguistik 19: 22–47.
Auer, P. & A. Di Luzio (eds.) (1992). The Contextualization of Language. John Benjamins.
Auer, P., E. Couper-Kuhlen & F. Müller (1999). Language in Time: The Rhythm and Tempo of Spoken
Interaction. Oxford University Press.
Bolinger, D. (1965). The atomization of meaning. Language 41: 555–573.
——— (1961). Generality, gradience and the all or none. Mouton.
——— (1972a). Accent is predictable (if you’re a mind-reader). Language 48(3): 633–44.
——— (1972b). Around the edge of language: Intonation. In D. Bolinger (ed.), Intonation. Selected
readings: 19–29. Penguin.
——— (1986). Intonation and its Parts. Stanford University Press.
Brazil, D., M. Coulthard & C. Johns (1980). Discourse Intonation and Language Teaching. Longman.
Chafe, W. (ed.) (1980). The Pear Stories: Cognitive, Cultural, and Linguistic Aspects of Narrative
Production. Ablex.
——— (1993). Prosodic and functional units of language. In J.A. Edwards & M.D. Lampert (eds.): 33–43.
Couper-Kuhlen, E. (1983). Intonatorische Kohäsion: Eine makroprosodische Untersuchung. Zeitschrift
für Linguistik und Literaturwissenschaft 49: 74–100.
——— (1986). An Introduction to English Prosody. Edward Arnold & Max Niemeyer.
——— (1992). Contextualizing discourse: The prosody of interactive repair. In P. Auer & A. di Luzio
(eds.): 337–364.
——— (1993). English Speech Rhythm: Form and function in everyday verbal interaction. John Benjamins.
——— (1996a). Intonation and clause combining in discourse: The case of ‘because’. Pragmatics 6:
389–426.
——— (1996b). The prosody of repetition: On quoting and mimicry. In E. Couper-Kuhlen &
M. Selting (eds.): 366–405.
——— (2001). Interactional prosody: High onsets in reason-for-the-call turns. Language in Society
30(1): 29–53.
Couper-Kuhlen, E. & M. Selting (1996). Towards an interactional perspective on prosody and a prosodic
perspective on interaction. In E. Couper-Kuhlen & M. Selting (eds.): 11–56.
Couper-Kuhlen, E. & M. Selting (eds.) (1996). Prosody in Conversation. Interactional studies. Cambridge
University Press.
Couper-Kuhlen, E. & S.A. Thompson (1999). On the concessive relation in conversational English.
In F.-W. Neumann & S. Schuelting (eds.), Anglistentag 1998 Erfurt: 29–39. Wissenschaftlicher
Verlag.
Cruttenden, A. (1997). Intonation, 2nd edition. Cambridge University Press.
Crystal, D. (1969). Prosodic Systems and Intonation in English. Cambridge University Press.
——— (1975). The English Tone of Voice. Edward Arnold.
Du Bois, J.W., S. Schuetze-Coburn, D. Paolino & S. Cumming (1992). Discourse Transcription. Santa
Barbara Papers in Linguistics 4: 1–225.
Du Bois, J.W., S. Schuetze-Coburn, S. Cumming & D. Paolino (1993). Outline of discourse transcription.
In J.A. Edwards & M.D. Lampert (eds.): 45–89. Lawrence Erlbaum.
Edwards, J.A. & M.D. Lampert (eds.) (1993). Talking Data: Transcription and coding methods for
language research. Lawrence Erlbaum.
Firth, J.R. (1957). Sounds and prosodies. In Papers in Linguistics 1934–1951. Oxford University Press.
Ford, C.E. (1993). Grammar in Interaction. Cambridge University Press.
——— (2000). The treatment of contrasts in interaction. In E. Couper-Kuhlen & B. Kortmann (eds.), Cause,
Condition, Contrast, Concession: Cognitive and discourse perspectives: 283–311. Mouton de Gruyter.
Ford, C.E. & S.A. Thompson (1996). Interactional units in conversation: syntactic, intonational, and
pragmatic resources for the management of turns. In E. Ochs, E.A. Schegloff, & S.A. Thompson
(eds.): 134–184.
Ford, C.E., B.A. Fox & S.A. Thompson (1996). Practices in the construction of turns: the ‘TCU’
revisited. Pragmatics 6: 427–454.
Fox, B. (1987). Discourse Structure and Anaphora. Cambridge University Press.
Fox, B.A., M. Hayashi & R. Jasperson (1996). Resources and repair: A cross-linguistic study of syntax
and repair. In E. Ochs, E.A. Schegloff, & S.A. Thompson (eds.): 185–237.
Fox, B.A. & R. Jasperson (1995). A syntactic exploration of repair in English conversation. In
P.W. Davis (ed.), Alternative Linguistics. Descriptive and theoretical modes: 77–134. John Benjamins.
Freed, A. (1994). The form and function of questions in informal dyadic conversation. Journal of
Freese, J. & D.W. Maynard (1998). Prosodic features of bad news and good news in conversation.
French, P. & J.K. Local (1983). Turn-competitive incomings. Journal of Pragmatics 7: 17–38.
Garfinkel, H. (1967). Studies in Ethnomethodology. Prentice-Hall.
Goldberg, J.A. (1978). Amplitude shift: A mechanism for the affiliation of utterances in conversa-
tional interaction. In J. Schenkein (ed.), Studies in the Organization of Conversational Interaction:
199–218. Academic Press.
Goodwin, C. (1995). Co-constructing meaning in conversations with an aphasic man. Research on
Language and Social Interaction 28(3): 233–260.
Günthner, S. (1996). The prosodic contextualization of moral work: An analysis of reproaches in
‘why’-formats. In E. Couper-Kuhlen & M. Selting (eds.): 271–302.
Gumperz, J.J. (1982). Discourse Strategies. Cambridge University Press.
Halliday, M.A.K. (1967). Intonation and Grammar in British English. The Hague.
——— (1970). A Course in Spoken English: Intonation. Oxford University Press.
Heritage, J. (1984). A change-of-state token and aspects of its sequential placement. In M.J. Atkinson &
J. Heritage (eds.): 299–345.
Heritage, J. & A.L. Roth (1995). Grammar and institution: questions and questioning in broadcast
media. Research on Language and Social Interaction 28: 1–60.
Jefferson, G. (1984). Notes on some orderliness of overlap onset. In V. D’Urso & P. Leonardi (eds.),
Discourse Analysis and Natural Rhetorics: 11–38. Cleup Editore.
Prosody 189
——— (1987). On exposed and embedded correction in conversation. In G. Button & J.R.E. Lee
(eds.), Talk and Social Organization: 86–100. Multilingual.
Kelly, J. & J. Local (1989). Doing Phonology. Manchester University Press.
Liberman, M. & I. Sag (1974). Prosodic form and discourse function. Chicago Linguistic Society
10: 416–427.
Local, J. (1996). Conversational phonetics: Some aspects of news receipts in everyday talk. In
E. Couper-Kuhlen & M. Selting (eds.): 177–230.
Local, J., J. Kelly & W.H. Wells (1986). Towards a phonology of conversation: turn-taking in Tyneside
English. Journal of Linguistics 22: 411–437.
Lord, A.B. (1964). The Singer of Tales. Harvard University Press.
Ochs, E., E.A. Schegloff, & S.A. Thompson (eds.) (1996). Interaction and Grammar. Cambridge
University Press.
Pike, K.L. (1945). The Intonation of American English. University of Michigan Press.
Pomerantz, A. (1984). Agreeing and disagreeing with assessments: some features of preferred/
dispreferred turm shapes. In J.M. Atkinson & J. Heritage (eds.): 57–101.
Sacks, H., E. Schegloff & G. Jefferson (1974). A simplest systematics for the organization of turn-
Sag, I. & M. Liberman (1975). The intonational disambiguation of indirect speech acts. Chicago Lin-
guistic Society 11: 487–497.
Schegloff, E.A. (1988). Discourse as an interactional achievement II: An exercise in conversation
analysis. In D. Tannen (ed.), Linguistics in Context: Connecting observation and understanding:
135–158. Ablex.
——— (1990). On the organization of sequences as a source of “coherence” in talk-in-interaction.
In B. Dorval (ed.), Conversational Organization and its Development: 51–77. Ablex.
——— (1996). Turn organization: One intersection of grammar and interaction. In E. Ochs,
E.A. Schegloff, & S.A. Thompson (eds.): 52–133.
——— (1998). Reflections on studying prosody in talk-in-interaction. Language and Speech 41: 235–263.
Schegloff, E., G. Jefferson & H. Sacks (1977). The preference for self-correction in the organization of
repair in conversation. Language 53: 361–382.
Schegloff, E.A. & H. Sacks (1984). Opening up closings. In J. Baugh & J. Sherzer (eds.), Language in
Use. Readings in sociolinguistics: 69–99. Prentice-Hall.
Schuetze-Coburn, S., M. Shapley & E.G. Weber (1991). Units of intonation in discourse: A compari-
son of acoustic and auditory analyses. Language and Speech 34(3): 207–234.
Selting, M. (1996a). On the interplay of syntax and prosody in the constitution of turn-constructional
units and turns in conversation. Pragmatics 6(3): 357–388.
——— (1996b). Prosody as an activity-type distinctive cue in conversation: the case of so-called ‘astonished’
questions in repair. In E. Couper-Kuhlen & M. Selting (eds.): 231–270.
Silverstein, M. (1976). Shifters, linguistic categories, and cultural description. In K. Basso &
H.A. Selby (eds.), Meaning in Anthropology: 11–55. University of New Mexico Press.
Tedlock, D. (1977). Toward an oral poetics. New Literary History 8: 507–519.
Trager, G.L. (1958). Paralanguage: A first approximation. Studies in Linguistics 13(1): 1–12.
Uhmann, S. (1992). On some forms and functions of speech rate changes in everyday conversation.
In P. Auer & A.D. Luzio (eds.): 297–336.
Wells, B. & S. MacFarlane (1998). Prosody as an interactional resource: Turn-projection and overlap.
Language and Speech 41: 265–294.
Woodbury, A.C. (1985). The functions of rhetorical structure: A study of Central Alaskan Yupik
Eskimo discourse. Language in Society 14: 153–190.
Yule, G. (1980). Speakers’ topics and major paratones. Lingua 52: 33–47.
Reported speech
Elizabeth Holt
University of Huddersfield
1. Introduction
Typically the term ‘reported speech’ is used to refer to the presentation of discourse
that purports to be from a prior occasion, and may originate from another author.
It is often used as a synonym of ‘quotation’ or ‘citation’. The category includes, for
example, the replaying of the words of another speaker, or the same speaker on a
previous occasion, in spoken discourse. The term can also include less prototypical
instances such as reports of hypothetical locutions (see Irvine, 1996; Mayes, 1990).
Reports of discourse not covered by the term ‘speech’, for example, thought and writ-
ing, are closely related and sometimes subsumed by it. Thus, the portrayal of the words
of characters in a novel, or the quotation of a comment made by a public figure in a
newspaper, can be included within the category. Instances of reported speech are
recurrently preceded or followed by a reporting clause such as ‘he said’, ‘I thought’, ‘she
goes’, ‘he whispered’, ‘they were like’.
Reported speech has received attention from scholars in the fields of literary
theory, philosophy of language, linguists, sociolinguists, conversation analysis and
ethnographers of communication. Although much research focuses on reported
speech in English, there are a number of studies of the phenomenon in other lan-
guages (see, for example, Aikhenveld & Dixon, 2003; Cohen et al. 2002). Issues
receiving analytic attention include: the reflexive and metapragmatic capacity of
reported speech (D. Davidson, 1968–9, 1984; Lucy, 1993; Quine, 1960); the semantic
and syntactic differences between direct and indirect forms (Banfield, 1973, 1982;
Li, 1986; Partee, 1973; Wierzbicka, 1974); the relationship between the reporting
clause and the report (McGregor, 1994); the absence of the reporting clause and
alternative forms (Mathis & Yule, 1994; Tannen, 1986); the authenticity of reported
speech (Philips, 1986, 1992); its role in storytelling (Wierzbicka, 1974); and its use
in a wide range of different interactional environments and contexts (see below).
Reflecting the fact that the majority of authors have tended to focus on spoken
discourse, this article will be mainly concerned with reported speech in verbal interac-
tion. I begin by referring to a number of authors who have recognised the importance
and ubiquity of reported speech in discourse and who have been highly influential in
the field (Section 2). I then consider the structure and forms of the device (Section 3),
Reported speech 191
which leads on to consideration of its role in interaction and the environments of its
use (Section 4).
2. Influential figures
2.1 Vološinov and Bakhtin

Recognition of the incredible ubiquity of speech reporting has much to owe to the
work of Vološinov and Bakhtin. Considering both literary and spoken communi-
cation, Bakhtin (1986: 93) undermined the distinction between reported and non-
reported discourse by pointing out that any utterance contains the “half-concealed or
completely concealed words of others”. Furthermore, he highlighted how pervasive
reporting the words of others is in interaction:
…in real life people talk most about what others talk about – they transmit, recall,
weigh and pass judgement on other people’s words, opinions, assertions, information;
people are upset by others’ words, or agree with them, contest them, refer to them and
so forth.” (Bakhtin, 1981: 338)
The work of Bakhtin and Vološinov has subsequently inspired many authors in the
study of reported speech. They identified its functions across a wide range of types of
discourse and Vološinov (1971: 153) adumbrated the direction most research would
take when he wrote:
Earlier investigators of the forms or reported speech committed the fundamental
error of virtually divorcing the reported speech from the reporting context…the true
object of inquiry ought to be precisely the dynamic interrelationship of these two
factors, the speech being reported (the other person’s speech) and the speech doing
the reporting (the author’s speech).
Yet despite recognising the ubiquity of reported speech in discourse and urging inves-
tigators to study the relationship between the reported utterance and the reporting
context, Bakhtin and Vološinov drew most of their examples from literary narrative.
In this article it will become clear that research based on invented examples or ones
taken from literary contexts has revealed much about the forms of the device. But the
increasing body of research based on empirical analysis of its use across a wide range
of contexts has shed further light on both its design and uses.
2.2 Goffman
Like Vološinov and Bakhtin, the Sociologist, Erving Goffman recognised that reported
speech is part of a wider process that is pervasive in talk:
192 Elizabeth Holt
In daily life the individual ordinarily speaks for himself, speaks, as it were, in his ‘own’
character. However, when one examines speech, especially the informal variety, this
traditional view proves inadequate…When a speaker employs conventional brackets
to warn us that what he is saying is meant to be taken in jest, or as mere repeating of
words by someone else, then it is clear that he means to stand in a relation of reduced
personal responsibility for what he is saying. He splits himself off from the content
of the words by expressing that their speaker is not he himself or not he himself in a
serious way (1974: 512).
To Goffman, reported speech is a function of the fact that speakers constantly ‘shift
footing’. Footing is defined as ‘the alignment of an individual to a particular utter-
ance…’ (1981: 227). He shows that the role of speaker is in fact made up of a number
of constituent parts: the ‘animator’ – ‘the sounding box’; the ‘author’ – ‘the agent who
scripts the lines’; and the ‘principal’ – ‘the party to whose position the words attest’
(1981: 226). Although the speaker may play all three roles at the same time, commonly
they play only one or two. Goffman gives the example of a vice-president reading out
a speech on behalf of the president: he plays the animator but not the author or princi-
pal. Reported speech is, then, one of the many ways in which we can adopt a footing by
changing roles within the ‘production format’. (See C. Goodwin, 2007; and Levinson,
1988; for further discussion of the production format.)
Goffman (1981) pointed out that one reason for shifting footing in order to play
the animator of an utterance that was authored by someone else on another occasion
is that it reduces personal responsibility. Thus, a speaker can have greater licence when
using reported speech. He notes how taboo utterances and curses may be used with
greater freedom when speakers shift footing to distance themselves from the author-
ship and principal of the locution.
3. Forms of reported speech
3.1 Categories and terminology

Many terms have been used to refer to the phenomenon reflecting both the diversity
of the category and concerns with the term ‘reported speech’. These include: quota-
tion, quoted direct speech (Macauley, 1987), oratio recta/oratio oblique, represented
speech/thought, direct/indirect speech, summary reported speech (Buttny & Williams,
2000), reported dialogue (Günthner, 1997), discourse presentation (Semino & Short,
2004) and constructed dialogue (Tannen 1986, 1989).
This range of terms reflects growing awareness of the varied forms that can be
considered members of the category; however, much research has focused on dis-
tinctions between direct and indirect reported speech (henceforth DRS and IRS).
In DRS the writer/speaker claims to produce an authentic version of the prior, for
Reported speech 193
example (taken from conversation and transcribed using conversation analytic tran-
scription conventions):
(1) [Holt:C85:4:2–3]
1 Lesley: AND uh ↑we were looking rou-nd the ↓stalls ‘n poking about
2 ‘n he came up t’me ‘n he said Oh: hhello Lesley, (.) ↑still
3 trying to buy something f ’nothing .tch! .hhhahhhhhhhhh!
Here Lesley purports to reproduce the locution. It begins with the discourse particle or
turn initial ‘Oh’ followed by the greeting. These clearly indicate that she is purporting to
reproduce words said on a previous occasion. Inclusion of the term of address ‘Lesley’
also indicates that this was an utterance made to her by another speaker. A marked
shift in prosody, represented by the underlining (for emphasis) and the arrows (indi-
cating a marked upward or downward shifts in intonation), also helps to portray this
as the words of another being replayed by Lesley now.
In IRS the writer/speaker does not purport to reproduce the actual words used:
(2) [Holt:X(C)1:1:6:8]
1 Lesley: ih↓Ye::s uhm-:-: u-Mark said eez not surpri:zed
2 that he:hh that he behaved like tha:t?
In this instance Lesley does not claim to reproduce the locution. The pronoun ‘he’ is
not changed to ‘I’ and there is no noticeable shift in intonation or inclusion of a turn
initial (as in, for example, ‘he said ↑well I’m not surprised he behaved like that’).
According to Coulmas (1986: 2) DRS “evokes the original speech situation and
conveys, or claims to convey, the exact words of the original speaker” while IRS
“adapts the reported utterance to the speech situation of the report”. In DRS the
speaker purports to portray not only the words used, but also the way the utterance
was made. Thus, in DRS pronominalization, place and time deixis, and verb tense
are all from the point of view of the ‘original’ speaker (Li, 1986). Further elements
of DRS such as turn initials, terms of address and marked changes of intonation or
voice quality can be used to contribute to the impression that the speaker is portray-
ing an utterance from another occasion. In IRS inclusion of the complementizer
‘that’ may contribute towards indicating that the reported utterance is being adapted
to the current speech situation.
According to Vološinov (1971) DRS is characteristic of the ‘linear style’ and reflects
a concern for maintaining the “integrity and authenticity” (p. 153) of the original. IRS
is characteristic of the ‘pictorial style’ and facilitates infiltration by the author to com-
ment on or interpret the meaning of the quotation.
However, the distinction between direct and indirect forms is not always clear-
cut. Speakers can, for example, recall utterances using IRS, while including words or
phrases that appear to be directly reported.
194 Elizabeth Holt
(3) [NB:II:2:10)
1 Nancy: …I only had o:ne (0.3) .hhhhhh (0.4)
2 dero:gatory rema:rk? if: you c’d call it
3 tha:t a:nd ah,h (0.6) u-it ca:me from a
4 gi:rl (0.2) and she said she fe:lt thet
5 → I: would of gott’n more out’v the cla:ss
6 → if I hed not been en eVOIder, h w’tever
7 → sh’meant by tha:t, .hhhhh u-but that
8 → ah:::, (0.5) I will c’ntinue t’remember
9 → th’class en gro:w from it. Er sump’n (.)
The majority of Nancy’s report here is indirect (for instance, the pronouns are from
the point of view of the current speaker), but “en eVOIder” appears to be directly
reported. Elements of the last part of the reported speech – “will c’ntinue t’remember
th’class en gro:w from it” – may also to be directly reported. Forms that amalgamate
elements associated with indirect and direct reporting have been termed ‘free indirect’
or ‘quasi-direct’ speech and have received attention in spoken and written form (Banfield,
1973, 1982; Coulmas, 1986; Semino & Short 2004; Thompson, 1984; for a survey see
McHale, 1978).
The division of the category into direct and indirect speech (sometimes with the
inclusion of free-indirect), and concentration on these forms, may derive, at least in
part, from the fact that early analysis was often based on invented examples or ones
taken from literary texts (e.g. Coulmas, 1986; Haiman, 1985; Halliday, 1985; Li, 1986;
Wierzbicka, 1974). However, recent research on reported speech in interaction has
demonstrated that divisions based on the features mentioned above can gloss over the
complexity of the device. Günthner (1997), in an analysis of reported speech in German,
found that speakers can convey affect using indirect forms and, thus,
…the simple dichotomy between direct reported speech as conveying the “message
plus the form” of the quoted utterance, and indirect reported speech as just conveying
the “message”, unduly reduces the complexities of restaging past dialogues in everyday
interactions. (Günthner 1997: 267)
Other recent research on reported speech in spoken interaction has begun to reveal the
range of forms that may be seen as being included within the category. For example,
Haakana (2007) explored reported thought in complaint sequences in Finnish, and
Holt (2007) considered ‘enactments’ of hypothetical conversations (see also Semino &
Short, 2004).
3.2 The reporting clause

Reported speech usually involves two elements: a reporting or framing clause involving
a pronoun or name and a verbum dicendi, such as ‘he said’ or ‘I thought’, and the
Reported speech 195
report itself. Linguists have been interested in the relationship between these two, and
it has variously been viewed as involving the embedding of the report as a complement
clause, as one of dependence (Foley & van Valin, 1984; Halliday, 1985;), and of framing
(McGregor, 1994). On occasions reported speech may be introduced without a prefa-
tory component or framing clause, these are sometimes known as zero-quotatives (see
Holt, 2007; Klewitz & Couper-Kuhlen, 1999; Longacre, 1985; Mathis & Yule, 1994;
McGregor, 1994; Romaine & Lange, 1991; Tannen, 1989). Linguists and sociolinguists
have explored the variety of phenomena that can be used to introduce the reported
speech. Certain elements such as ‘be like’, ‘be’ and ‘like’ appear to be undergoing
change in the frequency of their use. Thus, according to Romaine and Lange (1991)
the increasingly widespread use of ‘like’ in American English blurs the boundaries
between direct and indirect forms (see also Cukor-Avila, 2002; Ferrara & Bell, 1995;
Macauley, 2001; and Tagliamonte & Hudson, 1999). Whilst analytic attention has been
paid to the various means by which the start of a reported utterance is signalled, there
has been less concern with indications that the speaker has reached the end of a report.
However, in a study focusing on reported speech in Russian, Bolden (2004) found
that a range of ‘repositioning’ devices, such as producing a comment on the quoted
utterance(s), the presence of disfluencies in the talk, and using indexical expressions
co-referential with the reporting context, as well as changes in prosody, can act as an
‘unquote’. But speakers can also introduce ambiguity about whether they are continu-
ing to report the words of another, or reverting back to their own, by ‘fading out’.
4. Reported speech in discourse
4.1 The authenticity of reported speech

Because reported speech purports to be a replaying of a prior utterance, it has com-
monly been viewed (by analysts and speakers [see Mayes, 1990: 330–31]) as an accurate
rendition of previous speech. Quotations were seen as ‘demonstrations’ by Clark and
Gerrig (1990) because, unlike descriptions of a prior comment, they depict their refer-
ents. DRS, in particular is considered to be an authentic reproduction of a prior locu-
tion. In an early study of the device, Bally (1914) described DRS as “a phonographic
reproduction of the thoughts and words of the original speaker” (quoted in Clark &
Gerrig, 1990). Li (1986) states: “A direct quote communicates a more authentic piece
of information than an indirect quote in the sense that a direct quote implies a greater
fidelity to the source of information than an indirect quote” (p. 41). According to
Mayes (1990: 331), direct quotation is assumed to be a verbatim ‘re-enactment’ the
prior utterance that may include the original intonation and gestures. That DRS is gen-
erally judged to be an accurate replaying can be used by participants to interaction, for
196 Elizabeth Holt
example, in courtroom talk lawyers exploit the assumption that DRS is seen as more
accurate and consequently more reliable (Philips, 1986). In religious rituals -such as
trances- and medium-client consultations, speakers use DRS to convey the authentic-
ity of their message (Du Bois, 1986; Wooffitt, 2007). According to Besnier (1993) on
the Polynisian Island of Nukulaelae Atoll, where it is believed that messages should be
conveyed with great accuracy, there is a preference for direct forms.
An upshot of the fact that reported speech (particularly DRS) is often seen as
an authentic rendition of a prior locution means that it not only conveys what was
said but also offers evidence. In research on speakers recounting paranormal experi-
ences Wooffitt (1992) discovered that DRS is frequently employed to make their claims
robust. Here is an instance from his corpus:
(4) [HD:223]
((The speaker has just finished recounting an experience that occurred to her husband
while he was living in a particular hut on the Samoan Islands.))
A: And, well, what is even more fascinating about the story is, that he’s telling
the experience to other people and they said “Oh, that wasn’t too strange an
experience” because they had heard it before about this particular hut.
According to Wooffitt (1992: 159) “The reported talk in this account serves to confirm
the objectivity of her husband’s experience: if others have heard similar reports from
people staying in the same place, the husband’s account is, in part, substantiated”. In
conversation DRS is also used to provide evidence of a former locution. In the follow-
ing extract from a telephone conversation, DRS is employed to provide evidence when
speakers disagree over a former comment.
(5) [Holt:SO88:1:11:11]
((Vanna is staying with Lesley’s Mum. During a call between Mum and Lesley, Mum
asks if she would like to talk to Vanna.))
1 Mum: [W’l y’talk tuh][ [Vanna,
2 Les: [hh[hhhh
3 Mum: Lesley=
4 Mum: =Mm. ( [ )
5 Les: [((sniff!))
6 (0.5)
7 Van: H’llo Lesle[y.
8 Les: [Hello:, Sh[e getting confu:sed,.h[hhhh
9 Van: [( ) [Pardon?
10 Les: Is she getting confu:se[d?h
11 Van: [↑No:: she’s not[gett↓ing=
12 Les: [.hhhh
13 Van: =confu[sed.↓
Reported speech 197
14 Les: [ih-↑She called you v- me Va:nna heh heh

15 Les: he[h ↑.eh .h h h h h h h
16 Van: [No:: she said would you like tuh talk tuh Va:-nna.=
17 Les: =.hh ↑N- Oh eh I think she meant the other way round.
Lesley’s question in line 8 appears to stem from her belief that her mother has just
called her Vanna. Vanna disagrees emphatically in lines 11 and 13 and uses DRS
(line 16) to portray Mum’s “actual” words (Holt, 1996). Thus, in the face of disagreement
(see lines 14–15), Vanna purports to reproduce the disputed locution. (Interestingly,
although it was uttered only a short while beforehand, Vanna does not reproduce the
turn at line 1 entirely verbatim.)
A related feature of reported speech (particularly direct forms) is that it purports
to provide the recipient with “access” to the locution (or thought) (Golato, 2002; Holt,
1996). An upshot of this is that recipients can assess a reported comment. This is par-
ticularly useful in complaints, as Extract (1) exemplifies. Here is the extract again with
the following turns where the recipient reacts to the story.
(1) [Continuation]
1 Lesley: AND uh ↑we were looking rou-nd the ↓sta:lls ‘n poking
2 about ‘n he came up t’me ‘n he said Oh: hhello Lesley,
3 (.) ↑still trying to buy something f ’nothing,
4 Joyce: hh[hahhhhhh!
5 Lesley: [↑ hhohhh!
6 (0.8)
7 Joyce: Oo[ : : : ]: ( )]
8 Lesley: [↑Oo:.]ehh heh ↑heh]
9 (0.2)
10 Joyce: ↓I:s[n ‘ t ] [↓he
11 Lesley: [↑What]do ↑y[ou ↑sa↓:y.
12 (0.3)
13 Joyce: ↓Oh isn’t he ↓drea:dful.
In recalling this put-down Lesley portrays the man’s word’s without overtly evaluating
them (in introducing the story she says she is “broiling about something”, but she
then proceeds to give a number of details leading up to the DRS without any overt
evaluation). In line 4 Joyce does a shocked sounding in-breath followed by an exas-
perated sounding “Oo::::” ( the colons representing a stretched sound). She begins
an assessment in line 10 which appears to be completed in line 13. Thus, based on
the story Lesley has told, culminating in the report of the man’s words, Joyce offers a
negative assessment (Holt, 1996, 2000). The capacity to provide access to locutions or
thoughts renders reported speech a useful device in many interactional environments,
for example, when speakers wish to portray a series of utterances or thoughts. Thus,
198 Elizabeth Holt
DRS is recurrently used to convey several turns that constituted a conversation, giving
the recipient an insight into the stances of the reported speakers, or to portray a series
of thoughts that accompanied some actions or occurrences (Holt, 1996).
But reported speech is characterized by paradox. Despite the fact that it is used to
provide evidence of a former locution (or thought), and that recipients often treat it
as accurate, it is unlikely to be so on most occasions. Psychologists have demonstrated
that verbatim recall is not often possible: participants tend to remember the mean-
ing rather than the form of locutions (see Lehrer, 1989). In an analysis of reported
speech in interaction Mayes (1990) found that the authenticity of at least fifty percent
was doubtful. Further, Sternburg (1982: 108) observed that, no matter how accurately
discourse is reproduced, “tearing a piece of discourse from its original habitat and
recontextualizing it within a new network of relations cannot but interfere with its
effect”. Increasing awareness of the unlikelihood of reported speech being the accurate
portrayal of prior locutions or thoughts has lead many researchers to reject the term
and substitute others that avoid the implication of authenticity. For example, Tannen
(1986) referred to it as ‘constructed dialogue’.
4.2 Reported speech in storytelling

In the prior section aspects of the interactional uses of reported speech have begun
to become clear. This section continues this theme by considering research that has
focused on reported speech in discourse. The recent move towards empirical analysis
has both lent further weight to claims made by linguists and other authors based on
invented examples, and revealed new aspects of the device.
Reported speech commonly occurs in the environment of storytelling, so researchers
have tended to concentrate on identifying its uses in terms of its contributions to this
activity. According to Labov (1972) quoted direct speech can provide ‘embedded’ or
‘external’ evaluation. The former refers to occasions where the speaker explicitly states
their justification for telling a story in his or her role as a narrator; the later refers to
conveying the point of the story implicitly within the telling by, for example, reporting
the word’s of another.
The use of DRS at the peak of a narrative has been reported in a number of lan-
guages (see Glock, 1986; Larson, 1978; Li, 1986). Based on his analysis of the use of
reported speech in narratives in a Scottish dialect, Macaulay (1987: 29) stated:
The use of quoted direct speech adds liveliness to a narrative, it provides the possibility
of a different perspective and it can give the impression of authentic recollection of
an event…
DRS can display the speaker’s ‘involvement’ in the story (Chafe, 1982; Tannen, 1982)
and according to Wierzbicka (1974) DRS has a ‘theatrical’ nature. Mayes (1990: 346)
attributes the theatrical nature of DRS to the fact that its deictic centre is that of the
Reported speech 199
original event, thus “the speaker assumes the role of the person she/he is quoting, and
whether the quote is real or invented makes little difference as long as it is presented
effectively and the listeners believe the story”. While DRS occurs at the peaks of narratives,
can convey the point of a story, and gives an insight into the teller’s stance towards the
reported utterance, IRS is recurrently used to convey factual and background information
(Holt, 1996; Mayes, 1990).
4.3 The interactional environments of reported speech

The recent move towards exploring the use of reported speech in interaction (and
written [non-narrative] discourse, see Baynham & Slembrouck, 1999; Semino & Short,
2004) has resulted in analysis of the use of reported speech in a wide range of interac-
tional environments and contexts. Some of these will be considered in this section in
an attempt to provide an insight into the diversity of methods of speech and thought
presentation, the environments of its use, the interactional role it plays, and how con-
clusions based on invented examples (or from literature) are born out or refuted by
empirical analysis. I will begin with consideration of research exploring the device in
informal conversation, then turn to institutional contexts including the media.
In informal interaction reported speech is recurrent in many environments. One
particularly common one has already been mentioned, that of complaints. That DRS
can be employed to provide access to, and evidence of, a reprehensible comment
coalesce to render the device particularly useful at the climax of complaint stories
(Drew, 1998; Holt, 2000). An upshot of these characteristics is that it can be used to
provide the recipient the opportunity to be the first to explicitly assess the complained
about comment/conversation, thus enabling him or her to support the speaker’s implicit
evaluation (Holt, 2000). Hakanna (2007) showed that reported thought is also recur-
rently used in a collection of Finnish complaint stories. It can present a silent criticism
of the complained about person, thus helping to create a multi-layered portrayal of
the ‘original’ conversation and convey the current speaker’s evaluation, consequently
guiding the current recipient in offering assessment.
An upshot of the fact that DRS can provide implicit insight into the speaker’s stance
towards a reported utterance, and the level of fine ‘granularity’ (Schegloff, 2000) on
which it operates, contribute towards explaining its recurrent use in another conversa-
tional environment. Reported speech commonly occurs in amusing stories and joking
sequences. In telling a story a speaker may use laughter or smile voice at the delivery
of a reported utterance to convey their (and sometimes the reported speaker’s) stance
towards the locution. This can also serve to invite laughter from the current recipi-
ent, thus helping to facilitate a coordinated response by the participants in the light
of the report (Holt, 2000). In joking sequences involving ‘enactments’ (Holt, 2007)
of (often hypothetical) scenarios, speakers can use a form of DRS to collaboratively
200 Elizabeth Holt
produce a series of turns in which they play the roles of characters or actual people.
That these enactments often contain improprieties, leads us to another feature of DRS:
the speaker has greater licence when suggesting that the words are not their own. In
a study of joking sequences in which Western Apaches portray ‘the whiteman’ Basso
(1986) found that they were able to take moral liberties that would have been highly
controversial under different circumstances.
These features of DRS also coalesce to help explain its recurrence in a further
conversational environment – gossip. Bergman’s (1993) analysis of reported speech in
German ‘gossip’ demonstrates that speakers can delegate responsibility for the reported
words to the reported speaker. According to Bergman (1993: 113),
The fact that in quoting the speaker has greater freedom to use forbidden expressions
than in his own discourse makes quotations a particularly suitable means of presenting
obscene jokes – but also for gossip which is often concerned with sexual themes too.
Much of the research considered up until now has tended to focus on reported speech
in story-telling. However, analysis of interaction has revealed recurrent patterns
underlying the use of the device in non-narrative contexts, and has demonstrated that
it can be used to perform different actions. Based on analysis of a corpus of interviews
in French, Vincent and Perrin (1999: 293) concluded that non-narrative reported
speech can have “an appreciative function, if it reproduces a distinct point of view in
order to highlight an event related by the speaker…”; “a support function, if it tends
to illustrate a metadiscursive comment by the speaker…”; and “an authority func-
tion, when the speaker personally communicates what is expressed in the quote…’’.
Conversation analytic research by Couper-Kuhlen (2007) and Clift (2007) also demon-
strated that reported speech recurrently occurs in non-narrative contexts, and that a
thorough understanding of the device should consider other environments in which
it is used. Couper-Kuhlen (2007) found that reported speech is recurrently associated
with assessments and accounts and serves to heighten evidentiality. Relatedly, Clift
(2007) discovered that reported speech is used when there is some disagreement over
rights to assess.
Research on group interaction has highlighted similar uses of reported speech. In
an analysis of data drawn from focus group discussions, Myers (1999) built on Clark
and Gerrig’s (1990) argument that reported speech is a kind of demonstration to show
that it is employed to convey both ‘detachment’, when the report is distinct from what
the speaker says for him or herself, and ‘direct experience’ when it depicts what was
said and carries an immediacy, bearing an indexical connection to the reported con-
text. An upshot is that reported speech can be used in quite diverse ways including
to offer evidence in the light of actual or expected challenges, or to convey a stance
that the speaker may not agree with. In a study of discussions of racial attitudes on
a college campus, Buttny (1997) and Buttny and Williams (2000) found evidence of
Reported speech 201
the evidential use of reported speech, as well as its association with assessment and
complaints. Reported speech, in the data, is used to evaluate the speakers and others,
mainly involving criticising others.
That reported speech can be used to introduce a distance between the speaker’s
stance and the words or utterance portrayed has been found to be crucial in the use
of reported speech by news interviewers in ‘displaying neutrality’ (Clayman, 1988).
Building on Goffman’s theory of footing, Clayman (1988, 2007) shows that news inter-
viewers use reported speech and related forms to attribute a statement (often one that
is potentially contentious) to a third party (or to people en mass) in order to maintain
an appearance of neutrality. Wooffitt (2007), based on analysis of a corpus of public
demonstrations of mediumship in theatres, as well as private sittings, found that medi-
ums regularly use DRS to claim to reproduce the words of a spirit. Wooffitt attributes
this to the engaging nature of the device which also offers evidence of the presence of
the spirit, and to the contribution it makes to the ongoing sequence whereby it can be
used to manage potentially damaging episodes, thus supporting the finding that DRS
plays an important evidential role.
It is clear, then that much of this research follows Volšinov’s exhortation to explore
the relationship between the reported speech and the reporting context. The funda-
mental significance of the use of the reported speech on the ongoing action sequence
has been documented in a variety of settings. For example, Matoesian (2000) showed
how playing sections of a police/witness interview within the Kennedy Smith rape
trial was used in subtle and powerful ways to undermine the witness during cross-
examination. M.H. Goodwin (1990), in a study of American children, explored how
the reporting of potentially controversial comments by other children could lead
to confrontational events. That the use of reported speech may have a fundamental
impact on the recipient was suggested by Noy (2007) who found that listening to
accounts by Israeli backpackers changes the status of the listener to that of a ‘second
order witness’. (See also Wortham [1994] for an analysis of the impact of enactments
during classroom discussion).
5. Conclusion
In sum, recent interest in the use of reported speech in interaction has extended
our knowledge of the device and tested many of the claims made based on invented
examples or ones drawn from literature. In terms of forms of the device it has shown
that DRS is common and that speakers do orient to distinctions between it and other
forms, such as IRS. However, it has also highlighted the complexity of the category,
demonstrating that speech, thought and writing may be represented in a range of
ways and that all need to be considered for a full understanding of the device. It has
202 Elizabeth Holt
also demonstrated that we must be wary of seeing rigid distinctions between forms or
making assumptions about their design or uses without empirical analysis.
However, bearing this in mind, it is clear that much research on DRS has attested
to a group of overlapping and interconnected uses which stem from the fact that the
speaker purports to be simply reproducing the words (and sometimes prosody) of a
prior speaker, thus acting as a ‘demonstration’ or ‘showing’ of a previous locution. This
underlies the evidential use of the device, its ability to give the recipient ‘access’ to an
‘original’ utterance, exchange or thought, and that a speaker can distance themselves
from the report thus inviting the recipient’s assessment or making use of the greater
licence afforded in these circumstances. The complexity, multi-layered and sometimes
paradoxical use of the device is exemplified by its use as an evidential. Research has
shown that it is likely that it is rarely verbatim, yet speakers often treat it as providing
evidence. Further, although it may appear to be the reporting of a former utterance
etc, it contributes to a range of sequential environments in the reporting sequence,
and is, therefore designed to contribute to this action sequence as much as to report a
previous one.
References
Aikhenvald, A.Y. & R.M.W. Dixon (eds.) (2003). Studies in Evidentiality. John Benjamins.
Bakhtin, M.M. (1981). The Dialogic Imagination (M. Holquist, ed., and C. Emerson & M. Holquist,
Trans.). University of Texas Press.
——— (1986). Speech Genres and Other Late Essays (M. Holquist, ed., and C. Emerson & M. Holquist,
Trans.). University of Texas Press.
Banfield, A. (1973). Narrative style and the grammar of direct and indirect speech. Foundations of
Language 10: 1–39.
——— (1982). Unspeakable Sentences: Narration and Representation in the Language of Fiction.
Routledge & Kegan Paul.
Basso, E. (1986). Quoted dialogues in Kalapalo narrative discourse. In J. Sherzer & G. Urban (eds.),
Native South American Discourse: 119–68. Mouton de Gruyter.
Baynham, M. & S. Slembrouck (1999). Speech representation and institutionaldiscourse. Text, 19–4:
439–458.
Bergman, J.R. (1993). Discreet Indiscretions: The Social Organization of Gossip. Walter de Gruyter.
Besnier, N. (1993). Reported speech and affect on Nukulaelae Atoll. In J.H. Hill & J.T. Irvine (eds.),
Responsibility and Evidence in Oral Discourse: 161–181. Cambridge University Press.
Bolden, G. (2004). The quote and beyond: defining boundaries or reported speech in conversational
Russian. Journal of Pragmatics 36: 1071–1118.
Buttny, R. (1997). Reported speech in talking race on campus. Human Communication Research
23: 477–506.
Buttny, R. & P.L. Williams (2000). Demanding respect: The use of reported speech in discursive
constructions of interracial contact. Discourse and Society 11: 109–133.
Reported speech 203
Chafe, W.L. (1982). Integration and Involvement in Speaking, Writing and Oral Literature. In D. Tannen
(ed.), Spoken and Written Language: Exploring Orality and Literacy: 35–44. Ablex.
Clark, H.H. & R.J. Gerrig (1990). Quotations as demonstrations. Language 66: 764–805.
Clayman, S.E. (1988). Displaying neutrality in television news interviews. Social Problems 35(4):
474–492.
——— (2007). Speaking on behalf of the public in broadcast news interviews.In E. Holt & R. Clift
(eds): 221–243.
Clift, R. (2007). Getting there first: non-narrative reported speech in interaction. In E.Holt & R. Clift
(eds): 120–149.
Cohen, D., M-C. Simeone-Senelle & M. Vanhove (2002). The grammaticalisation of ‘say’ and ‘do’:
An areal phenomenon in East Africa. In T. Güldemann & M. von Roncador (eds.), Reported
Discourse: A Meeting Ground for Different inguistic Domain: 227–251. John Benjamins.
Coulmas, F. (1986). Reported speech: Some general issues. In F. Coulmas (ed.), Direct and Indirect
Speech: 1–28. Mouton de Gruyter.
——— (ed.) (1986). Direct and Indirect Speech. Mouton de Gruyter.
Couper-Kuhlen, E. (2007). Assessing and accounting. In E. Holt & R. Clift (eds.): 81–119.
Cukor-Avila, P. (2002). She say, she go, she be like: Verbs of quotation over time in African American
Vernacular English. American Speech 77, 1: 3–31.
Davidson, D. (1968–9). On saying that. Synthèse 19: 130–46.
——— (1984). Quotation. In D. Davidson (ed.), Inquiries into Truth and Interpretation: 79–92.
Pergamon.
Drew, P. (1998). Complaints about transgressions and misconduct. Research on Language and Social
Interaction 31: 295–325.
Du Bois, J.W. (1986). Self-evidence and Ritual Speech. In W. Café & J. Nichols (eds.), Evidentiality:
The Linguistic Coding of Epistomology: 313–331. Ablex.
Ferrara, K. & B. Bell (1995). Sociolinguistic variation and discourse function of constructed dialogue
introducers: The case of be+like. American Speech 70: 265–289.
Glock, N. (1986). The use of reported speech in Sacamaccan discourse. In G. Huttar & K. Gregerson
(eds.), Pragmatics in Non-Western Perspective (Publications in Linguistics 73): 35–61. Summer
Institute of Linguistics and University of Texas.
Goffman, E. (1974). Frame analysis: An essay on the Organization of Experience. Harper and Row.
——— (1981). Footing. In E. Goffman (ed.), Forms of Talk: 124–159. Blackwell.
Golato, A. (2002). Self-quotation in German: reporting on past decisions. In T. Güldemann &
M. von Roncador (eds.), Reported Discourse: A Meeting Ground for Different Linguistic Domains:
Goodwin, C. (2007). Interactive footing. In E. Holt & R. Clift (eds.), Reporting on Talk: Reported Speech
in Interaction: 6–46. Cambridge University Press.
Goodwin, M.H. (1990). He-said-she-said: Talk as social organization among black children. Indiana
University Press.
Günthner, S. (1997). The contextualization of affect in reported dialogues. In S. Niemeier & R. Dirven
(eds.), The Language of Emotions: Conceptualization, Expression and Theoretical Foundation:
Haiman. J. (1985). Natural syntax. Cambridge University Press.
Haakana, M. (2007). Reported thought in complaint sequences. In E. Holt & R. Clift (eds.), Reporting
on Talk: Reported Speech in Interaction: 150–178.
Halliday, M.A.K. (1985). An Introduction to Functional Grammar. Arnold.
204 Elizabeth Holt
Holt, E. (1996). Reporting on talk: the use of direct reported speech in conversation. Research on
Language and Social Interaction 29, 3: 219–45.
——— (2000). Reporting and reacting: Concurrent responses to reported speech. Research on Language
and Social Interaction 33: 425–454.
——— (2007). ‘I’m eyeing your chop up mind’: reporting and enacting. In E. Holt & R. Clift (eds.):
47–80.
Holt, E & R. Clift (eds.) (2007). Reporting on Talk: Reported Speech in Interaction. Cambridge University
Press.
Irvine, J.T. (1996). Shadow conversations: The indeterminacy of participant roles. In M. Silverstein &
G. Urban (eds.), Natural Histories of Discourse: 131–159. University of Chicago Press.
Klewitz, G. & E. Couper-Kuhlen (1999). Quote-unquote: The role of prosody in the contextualization
of reported speech sequences. Pragmatics 9: 459–485.
Labov, W. (1972). Language in the Inner City. University of Pennsylvania Press.
Larson, M.L. (1987). The Functions of Reported Speech in Discourse. The Summer Institute of Linguistics.
Lehrer, A. (1989). Remembering and presenting prose: Quoted speech as a data source. Discourse
Levinson, S.C. (1988). Putting linguistics on a proper footing: Explorations in Goffman’s concepts of
participation. In P. Drew & A. Wootton (eds.), Erving Goffman: Exploring the Interaction Order:
29–45. Mouton de Gruyter.
Li, C. (1986). Direct and indirect speech: A functional study. In C. Coulmas (ed.), Direct and Indirect
speech: 29–45. Mouton de Gruyter.
Longacre, R. (1985). Sentences as combinations of clauses. In T. Shopen (ed.), Language typology and
syntactic description: 235–286.Cambridge University Press.
Lucy, J.A. (ed.) (1993). Reflexive Language: Reported Speech and Metapragmatics. Cambridge University
Press.
Macaulay, R. (1987). Polyphonic monologue: Quoted direct speech in oral narrative, IPrA Papers in
Pragmatics 2: 1–34
——— (2001). You’re like ‘why not’: The quotative expressions of Glasgow adolescents. Journal of
Sociolinguistics 5/1: 3–21.
Mathis, T. & G. Yule (1994). Zero quotatives. Discourse Processes 18, 63–76.
Matoesian, G. (2000). Intertextual authority in reported speech: Production media in the Kennedy
Smith rape trial. Journal of Pragmatics 32: 879–914.
Mayes, P. (1990). Quotation in spoken English. Studies in Language 14: 325–363.
McGregor, W. (1994). The grammar of reported speech and thought in Gooniyandi. Journal of
Linguistics 14: 63–92.
McHale, B. (1978). Free indirect discourse: A survey of recent accounts. PTL: A Journal for Descriptive
Poetics and Theory of Literature 3: 249–287.
Myers, G. (1999). Functions of reported speech in group discussions. Applied Linguistics 20(3): 376–401.
Noy, C. (2007). A narrative community: Voices of Israeli backpackers. Wayne Sate University Press.
Partee, B.H. (1973). The syntax and semantics of quotation. In S.R. Anderson & P. Kiparsky (eds.),
A Festschrift forMorris Halle: 410–418. Holt Rinehart & Wilson.
Philips, S.U. (1986). Reported Speech as Evidence in an American Trial. In D. Tannen, J.E. Alatis
(eds.), Languages and Linguistics: The Interdependence of Theory, Data and Application: 154–179.
Georgetown University Press.
Philips, S.U. (1992). Evidentiary standards for American trials: just the facts. In J.H. Hill & J.T. Irvine
(eds.), Responsibility and Evidence in Oral Discourse: 248–259. Cambridge University Press.
Reported speech 205
Quine, W.V.O. (1960). Word and Object. MIT Press.

Romaine, S. & D. Lange (1991). The use of like as a marker or reported speech and thought: A case of
grammaticalization in progress. American Speech 66, 3: 227–279.
Semino, E.M. Wynne & M. Short (2004). Corpus Stylistics: Speech, Writing and Thought Presentation
in a Corpus of English Writing. Routledge.
Schegloff, E.A. (2000). On granularity. Annual Review of Sociology 26: 715–720.
Sternberg, M. (1982). Proteus in quotation-land: Mimesis and the forms of reported discourse. Poetics
Today 3: 107–156.
Tagliamonte, S. & R. Hudson (1999). Be like et al. Beyond America: The quotative system in British
and Canadian youth. Journal of Sociolinguistics 3: 147–172.
Tannen, D. (ed.) (1982). Spoken and Written Language: Exploring Orality and Literacy. Ablex.
——— (1986). Introducing constructed dialogue in Greek and American conversational and literary
narrative. In F. Coulmas (ed.), Direct and Indirect Speech: 311–332. Mouton de Gruyer.
——— (1989). Talking voices. Repetition, Dialogue and Imagery in Conversational Discourse. Cambridge
University Press.
Thompson, G. (1984). Voices in the text: Discourse perspectives on language reports. Applied Linguistics
17: 501–530.
Vincent, D. & L. Perrin (1999). On the narrative vs non-narrative functions of reported speech: a
socio-pragmatic study. Journal of Sociolinguistics 3: 291–313.
Volosinov, V.N. (1971). Reported speech. In L. Matejka & K. Pomorska (eds.), Readings in Russian
Poetics: Formalist and Structuralist Views: 149–175. MIT Press.
Wierzbicka, A. (1974). The Semantics of Direct and Indirect Discourse. Papers in Linguistics 7, 3/4:
267–307.
Wooffitt, R. (1992). Telling Tales of the Unexpected: the Organisation of Factual Discourse. Harvester
Wheatsheaf.
——— (2007). The dead in the service of the living. In E. Holt & R. Clift (eds.): 244–269.
Wortham, S.E.F. (1994). Acting out participation examples in the classroom. John Benjamins Publishing
Company.
Harvey Sacks
Rod Watson
University of Manchester
Harvey Sacks was educated at Columbia, Yale and Berkeley. He qualified in Law as
well as Social Science and came to be Professor of Sociology at the University of
California, Irvine. He died in a car accident in 1975 at the age of 40. By that time, he
had devised the analytic instruments for a methodologically-radical respecification of
the sociology of knowledge and, derivatively, of sociology itself. He treated knowledge
as being, primordially, of lay, commonsense status and gave us ways of analysing it,
which — unlike conventional sociologies — did not conflate lay society-members’
concerns with those of analysts.
This approach contrasted with the sociology of knowledge and culture as pro-
pounded by analysts such as Karl Mannheim, who, for instance, treated truth value
as something to be arbitrated by the professional analyst rather than lay members:
for Sacks, truth value was, first and foremost, a practical matter to be interactionally
managed by members, (Sacks 1975, 1992: 549–67). Instead of analytically stipu-
lating the so-called ‘connectives’ between knowledge and social groupings, Sacks
examined commonsense cultural knowledge as incorporated into social actions —
practices such as laic sociological description (Sacks 1963, 1972). Sacks refrained
from ‘second guessing’ or ‘trumping’ either lay members’ descriptions or their
descriptive work, and, importantly, treated lay descriptive practices as characteris-
tically being linguistically transacted.
Sacks’ respecification of the sociology of knowledge and culture might thus be
conceived in terms of a number of ‘turns’: (i) a turn focalising the lay determinations of
knowledge, (ii) a praxiological turn, focalising members’ practices, and (iii) a linguistic
turn, concentrating upon the conjoint, in situ linguistic transaction of these practices.
In addition — by virtue of his early and continuing association with the founder
of ethnomethodology (EM), Harold Garfinkel — Sacks took a procedural or methodic
turn, treating ordinary commonsense knowledge of social structures as ‘knowledge
how’ or ‘know-how’ rather than just ‘knowing that (something is the case)’: on this
distinction and its implications, see also Ryle 1971a. This knowledge of social structures
was not only employed by interlocutors in putting their conversations together as
assembled objects, but was also exhibited in and through the talk thus produced — in
‘what was being talked about’. These intricately interpenetrating concerns formed the
basis of his Lectures on Conversation (1992), (though the title of that collection does
scant justice to the intellectual scope of its contents). One of the best examples of the
Harvey Sacks 207
confluence of these concerns is to be found in Sacks’ work on membership categorisations,

to which I shall later return.
These turns resulted in a variety of analytic trajectories, especially in Sacks’ earlier
work, but the predominant one instituted a body of work known as ‘Conversation(al)
Analysis’ (CA). CA was not primarily intended to designate a sociological (or linguistic)
‘research technique’ as such, let alone a ‘perspective’, but instead was an indicatory title,
pointing to a topic for analysis, one that conventional sociologies had by their very
nature overlooked, ignored or otherwise neglected.
That topic was the methodic conversational practices conjointly undertaken
by lay members in conducting their everyday lives, e.g. in co-producing recognisably
adequate lay sociological descriptions of the situations to which they were party,
and from within these selfsame situations. Phenomena such as ‘description’, ‘recog-
nition’ and ‘familiarity’ are of generic significance as they point to members’ conver-
sational practices as, most centrally, possessing a sense-making dimension — again,
we see an early and strong affinity with Garfinkel’s EM. This affinity was to take pub-
lished form in an article co-authored with Garfinkel (Garfinkel & Sacks 1970), in
which the authors assembled conceptual resources for the naturalistic, praxiological
analysis of the reflexively accountable features of members’ practices. The paper was
notable for the authors’ insistence upon the centrality of members’ mastery of natural
language practices, e.g. glossing practices: such mastery was, indeed, considered to be
constitutive of membership.1
Sacks laid claim to the invention of CA (Sacks 1992: 549), but how he arrived at
it is not so easy to specify. To the missing topic of conversational order, Sacks did not
bring a research technique but what one might best describe as an ‘analytic mentality’,
as Schenkein (1978: 1–6) termed it, and this mentality had very many facets. In it,
one can, most assuredly, see Garfinkel’s ethnomethodology but also Goffman’s quasi-
ethnographic orientation (Sacks 1972), sensibilities deriving from his legal training
(Sacks 1997), an eclectic mix of philosophies, e.g. A. Schütz, J.L. Austin, and the range
of cognitive-anthropological approaches collected under the title ‘Componential
Analysis’ as espoused by Frake, Conklin, Sturtevant et al. However, from the earliest
phases of his work, we see a distinctly Sacksian ‘take’ on all these influences: they were
not models for his approach.
Instead of an aprioristic reliance upon analytic models, Sacks began with an empir-
ical puzzle. He himself attested that he had tape-recorded some naturally-occurring talk
and then began to conjecture how one might make sociological sense of these record-
ings were he to treat them as data. To conceive of recorded talk as ‘data’ in itself was
1. Sacks told me and some others in John Lee’s house in 1973 that the paper was essentially
Garfinkel’s. That much seems clear, yet one can discern many of Sacks’ early preoccupations in the
paper, especially the issues he raised through his ‘Commentator Machine’ metaphor (Sacks 1963).
208 Rod Watson
a big reach for the vast majority of sociologists, and still is. What members said was
typically adumbrated, edited down and re-written by sociologists for the purposes of
corroboration of some analytic claim or other. This selection, paraphrasing, condensing,
‘clarifying’, etc. exemplified what Dorothy Smith (1974) later analysed as ‘the ideological
practices of sociology’: a key example of these practices vis-à-vis linguistic data was
to be found in the work of sociolinguist Basil Bernstein. At least, though, Bernstein
noted the centrality of language in society. Most sociologists simply ignored it or took
it to be merely a transparent conduit to the great empirical complexes of, say, social
stratification, religion, occupational structure and the rest. Even linguists of the TG
persuasion treated conversational utterances as linguistic detritus.
Sacks made sure that his own work changed all that. Instead of treating language
in such a cavalier fashion, he considered it to be a sociological datum in its own right,
and not just another topic to be added to the litany of others. He treated language as
a generic phenomenon through which these other phenomena of social order were
constituted: again, see his ‘commentator machine’ metaphor (Sacks 1963). Rather than
treat language as theory-iterative, he let the tape- recorded, transcribed data set the
terms of his analysis, and did so in the most radically naturalistic way. He at least
endeavoured to transcribe the data in ways that maximised the retention of its
phenomenological integrity — including its socially-organised nature. Thus, instead
of establishing a methodologically ironic stance that involved a competitive, down-
grading, scientistic or otherwise relativising approach to members’ knowledge-based
language uses, Sacks proposed the following respecification (referring to some strands
of componential analysis):
Fields like ethnobotany, ethnoastronomy, etc. …are basically directed to a cross-cultural

situation in which one wants to be able to map onto a universal grid the various
conceptions of botany that one would find in given cultures, where what one finds
is treated as a version of scientific botany. [Or], for example, one would map color
conceptions onto our scientific organization of the possibilities and then you could
compare various societies.
Our relation to such an enterprise is perhaps better clarified by an alternative kind of

consideration. A curious fact becomes apparent if you look at the first paragraph — it
may occur in the third paragraph — of the reportedly revolutionary treatises back to
the pre-Socratics and extending up to at least Freud. You find they all begin by saying
something like this, ‘About the thing I’m going to talk about, people think they know
but they don’t. Furthermore, if you tell them it doesn’t change anything. They still
walk around like they know even though they are walking in a dreamworld’. Darwin
begins this way, Freud begins in a similar way. Bloomfield’s analysis of language begins
in a similar way and I could provide a much larger list. What we are interested in is,
what is it that people seem to know and use? Here, what people know and use is not
to be mapped for each area onto what it is that science turns out to know but is to be
Harvey Sacks 209
investigated (in) itself. How does ‘what people know and use’ work? How could it be
enforced? What are its properties? …
The problem is that, since each major treatise that has set up scientific fields starts out
by saying that what people know and use is wrong, obviously it would not be a way
to find out about what people know and use by using considering ‘science’ proposals,
what we want to do is see if we can look at the enforceable and usable procedures for
whatever knowledge persons happen to have. What procedures do people use to do
going about knowing about the world? In that way, the aim is not to map what we
learn onto some pre-given grid … it does not turn on … the existence of a science of
sociology against which an ethnosociology for America or for the Burundi would be
mapped. (Sacks, 1968: 12–14)
This early quotation amounted to a rejection of methodologically-ironic approaches

and its replacement by a conception of social (including, focally, conversational)
order as commonsensically describable and, as such, incorrigible by the analyst qua
analyst: instead, lay descriptive practices are transformed into an object of analysis
(again, Sacks 1963). The contemporary evolution of CA is the major — though not
the only2 — derivative of Sacks’ concern, though whether Sacks would licence some of
these contemporary developments is open to question. Still, the early incarnations of
CA recognisably bear his stamp, and it is to these that I shall now turn.
Many of Sacks’ lectures exhibit a profound ethnographic sensibility and might
even be seen as a dialogue with the ethnography of communication and ‘Chicago
School’ ethnographies. He was a great admirer of the latter, considering them to be
about the only form of sociology that was worth critically considering: he showed
considerable disdain for the rest. His analysis of police assessment of moral character
(Sacks 1972) manifests Chicago-style concerns as elaborated by Goffman. He showed
considerable interest in anthropologists such as Ethel Albert, and he himself contributed
to the extension of techniques of ethnographic discovery. As Garfinkel (2002: 182)
attests, Sacks had semi-conjecturally arrived at a quasi-Austinian conceptual distinction
between ‘possessables’ and ‘possessitives’, and said to Garfinkel:
Now, Harold, what do I mean by that distinction? That is what I want to find out. I
don’t want you to tell me. I don’t want to settle it like that. I could go to the UCLA
Library…. but that’s not the way I want to learn what I mean … I don’t want to write
definitions, I don’t want to consult authorities. Instead, I want to find a work group,
somewhere, perhaps in Los Angeles, who, as their day’s work, and because they know
about it as their day’s work, will be able to teach me what I could be talking about as
they know it as their day’s work.
2. For example, Sacks occasionally considered textual data on, say, membership categories from
newspapers, etc., and indeed wrote an early example of what came to be known as textual analysis
in EM and beyond (eventually published in Sacks 1999).
210 Rod Watson
Sacks did find such a setting in the L. A. Police Department’s identification of lost
cars, and he spoke more generally about the distinction, partly with reference to
conversational data in a lecture (Sacks 1992, vol. 1, pp. 605–9). The elaboration of
ethnographic technique, however, became part of the armoury of contemporary EM
rather than CA, where, say, work groups would ‘teach the ethnographer what he was
talking about’ (Garfinkel 2002, p. 186).
When Sacks gave a series of lecture-seminars at Didsbury College, Manchester in
1973, he again showed his ethnographic sensibilities in citing a transcript of a woman
speaking about a child who, as part of a charitable operation, had been given a coat
which was much too big for him. ‘He ran up and down in it and wasn’t he pleased!’
declared the woman. Sacks noted that there was a two-way, not just a one-way, transfer
of benefits operating here — a transfer of goods to the child and a transfer in the
opposite direction of moral credit and associated definitional rights to the woman.
This proceduralised ethnographic description was built into Sacks’ observations of the
story format of the woman’s account and struck me at the time as being an exemplary
instance of the ethnography of communication.
As the above ‘police’ example suggests, Sacks regarded ordinary persons as being
‘practical ethnographers’, and one cultural resource at their disposal was what he termed
‘membership categorisation activities’. Membership categorisations were one locus of
the distribution of persons’ commonsense knowledge of social structures. These cate-
gorisations comprised ordinary-language references to persons and were employed in
conversation. Sacks submitted some componential-analytic work to a radically praxi-
ological respecification, and the sphere of membership categories was a major focus
of his early up to mid-point CA. He inspected the lay methods persons employed in
using such categorial references — the co-selection of categories, the prediction of
activities upon categories, the achieving of referential adequacy, the selection between
natural collections (‘devices’) of membership categories, and the incorporation of cat-
egories into conversational sequencing (see, e.g. Sacks 1972b, 1972, and vol. 1 of his
Lectures, passim). This ‘mix’ of sequential and categorial concerns is to be found in
some canonical papers by Sacks (e.g. Sacks 1975) and in those of some of his earliest
collaborators. For instance, such a mix is to be found in what many consider to be
E.A. Schegloff ’s finest paper (Schegloff 1972).
The analysis of membership categories, always a shadowy presence in later CA,
has latterly become an object of fierce debate, with some arguing that sequential and cat-
egorial concerns ought to be pursued under a single rubric (Watson 1997) and others who
believe that membership categorisation analysis should become a separately-pursued
form of CA (as was mooted by Carolyn Baker and others, at the I. I. E. C. A. Conference
in Manchester, 1973).
What cannot be denied is that as Sacks’ work evolved and became more of a
collaboratively controlled enterprise, it increasingly came to focalise (a) sequen-
Harvey Sacks 211
tial analysis and (b) the ‘systematics’ of sequencing. The notion of (a) ‘systematics’,
like that of ‘mechanism’ or ‘apparatus’, is here best understood as an allusion to the
methodic, procedural organisation not only of the talk itself but also of the ‘knowl-
edge how’ that is integral to that organisation. The notion of ‘systematics’ comprises
the CA determination of what, for Garfinkel, is the methodic practices that comprise
the locally-embedded production procedures for specifically-situated patterned out-
comes (as recognised by interlocutors themselves).
Like Garfinkel, Sacks produced procedural descriptions of these outcomes, where,
increasingly, the emphasis was on the turn-organised nature of the outcomes. An
utterance may be treatable by interlocutors as some single speaker’s responsibility but
Sacks and his associates observed that each utterance was multilaterally formed, not
only by the speaker but also those who hand the floor to him/her, recipients who listen
and monitor for projected utterance completion and so on: the baseline resource for
this multilateral production is the turn-taking system itself. The utterance is treated
as a serial object, as integrated (by all parties) into a turn-organised sequence. The
turn-taking system involved is one that is specifically, minutely adapted to natural
conversation and for nothing else: queues, for instance, have a differently-adapted
turn organisation. Each system is self-organising — brought about by parties to that
specific system from within the system.
In addition to his work on membership categorisation practices, Sacks’ Lectures on
Conversation analyse adjacency-paired utterances, pre-sequences and other expansions
of adjacency pairs, pronouns as transformational and tying techniques, turn-allocation
techniques, preference organisation in relation to repair and agreements, topic and
story organisation, conversational openings and closings, the design of utterances as
sequential objects, recipient design (orientation to co-participant), repeat utterances
and transformations, puns, paradoxes, the use of numbers and spatialised references
considered as produced sequential objects, and very many other phenomena — all
phenomena for co-participants, not just the professional analyst, and phenomena
whose orderliness was available only at a level of fine detail.
Sacks showed both exquisite craftsmanship and intellectual profundity in his
empirically grounded yet deeply conceptual analyses of this detail. He identified a proof
procedure for the characterisation of utterances, one that was grounded in interlocutors’
own laic proof procedures. These procedures involved the inspection of how a given
utterance is treated by next speaker in the immediately succeeding utterance. This
treatment may, in turn, be monitored by the first speaker as displaying an understanding
(or otherwise) of her/his utterance: this is most highly delineated in adjacency-
paired utterances.
By an adjacently positioned second, a speaker can show that he understood what a
prior aimed at, and that he is willing to go along with that. Thus, by virtue of the
occurrence of an adjacently produced second, the doer of a first can see that what he
212 Rod Watson
intended was indeed understood, and that it was or was not accepted. Also, of course,
a second can assert his failure to understand, or a disagreement, and inspection of a
second by a first can allow the first speaker to see that while the second thought he
understood, indeed he misunderstood … (Schegloff & Sacks 1974: 240)
Here we see how close CA was to espousing the EM project of focalising sense-making
practices, and how conversational sequencing was, at least in part, analysed in that way.
It shows that ‘next (second) utterance analysis’ comprises an intersubjectively-grounded
‘check’ on the characterisation of what an utterance is doing in just this sequential
location, here and now. This contrasts with the more arbitrary, and often more sche-
matic, attribution of action or ‘function’ to utterances that is characteristic of so much
Discourse Analysis.
One corollary of the insistence on the finely-tuned adaptation of turn-taking
systems to the work they have to do — co-ordinating people’s conversational activi-
ties, producing an order of service, etc. — is the development of a sub-branch of CA
focusing upon speech exchange in contexts that were ‘institutional’ rather than those
of ordinary conversation: these include doctor — patient interaction, courtroom
cross-examinations, school classroom talk, etc. This preoccupation has its origins in
the closing section of Sacks’ and his colleagues’ powerful summary statement of the
concepts and findings of early CA (Sacks, Schegloff & Jefferson 1974). These ‘insti-
tutional’ speech exchanges were seen as each involving variable ‘mixes’ of ordinary-
conversational features and features that were specific to that particular setting-type. It
is too easy, however, to reify these systems: after all, Sacks’ foundational observations
on ordinary (‘natural’) conversation came, to a significant extent, from ‘institutional’
settings — telephoned calls to a ‘crisis intervention centre’ and group therapy sessions.
There was, in Sacks’ analyses of these data, precious little reference to the ‘institutional’
nature of the talk.3
Some final remarks on Sacks’ own conception of his work. Whilst — as the quota-
tion on p. 4 of this paper indicates — Sacks opposed scientism and the methodologi-
cally ironic invocation of idealised versions of ‘science’, he nonetheless conceived of
his work as founding a ‘primitive natural science’, one that was resolutely empirical
in nature. Its conceptualisations would thus be grounded in empirical phenomena,
where these phenomena were, without exception, intersubjectively available within
the practical attitude of members/interlocutors themselves and were displayed as
such. Along with Garfinkel (e.g. Garfinkel 2002: 186) Sacks referred to ‘reproducible’
solutions to questions as found, usually, in single instances (though, more rarely, in
3. In addition, we must take into account Ryle’s argument (1971b) that things that get done
linguistically are not themselves necessarily linguistic in nature.
Harvey Sacks 213
collections of data too). It must be said, however, that Sacks’ Lectures and publications
seldom make reference to the scientific nature of his work, and ‘science’ hardly seems
to have been for him an omnirelevant concern or organising principle of his analyses.
He certainly eschewed sociology’s proclivity toward ungrounded, unconstrained
formal abstraction (what Garfinkel 2002, calls Formal Analysis, FA): like Garfinkel,
he sought to find order in the concrete, specifically identifying phenomenal detail of
actions and settings. We can speculate that he might have found an irony in the arrogation
of so much contemporary CA by linguistics; CA has apparently cast off one disci-
plinary kind of FA only to take on another! More generally, one might observe that,
impressive as the current corpus of CA is, it is debatable as to whether it expresses
the sheer scope of Sacks’ intellectual concerns and ways of dealing with members’
sociological descriptions.
References
Garfinkel, H. (2002). Ethnomethodology’s Program: Working Out Durkheim’s Aphorism. Rowman and
Littlefield.
Garfinkel, H. & H. Sacks (1970). On Formal Structures of Practical Actions. In J.C. McKinney &
E.A. Tiryakian (eds.), Theoretical Sociology: Perspectives and Developments: 337–66. Appleton-
Century-Crofts.
Ryle, G. (1971a). Knowing How And Knowing That. In Collected Papers: 212–225. Hutchinson.
——— (1971b). Use, Usage and Meaning. In Collected Papers, vol. 2: 407–413. Hutchinson.
Sacks, H. (1963). Sociological Description. Berkeley Journal of Sociology 8(1): 1–16. [reprinted in
J. Coulter (ed.) (1975). Ethnomethodological Sociology. Edward Elgar Publishers.]
——— (1972a). Notes on Police Assessment of Moral Character. In D. Sudnow (ed.), Studies in Social
Interaction: 280–293. The Free Press.
——— (1972b). On the Usability Of Conversational Data For Doing Sociology. In D. Sudnow (ed.),
Studies in Social Interaction: 31–74. The Free Press.
——— (1972c). On the Analysability Of Stories By Children. In J.J. Gumperz & D. Hymes (eds.),
Directions in Sociolinguistics: The Ethnography of Communication: 325–345. Holt, Rinehart and
Winston. [reprinted in R. Turner (ed.) (1974). Ethnomethodology: 216–232. Penguin.]
——— (1975). Everyone Has To Lie. In M. Sanches & B. Blount (eds.), Sociocultural Dimensions of
Language Use: 57–80. Academic Press.
——— (1992). Lectures on Conversation, 2 vols. [ed. G. Jefferson]. Academic Press.
——— (1997). The Lawyer’s Work. In M. Travers & J.F. Manzo (eds.), Law in Action: Ethnomethodological
and Conversation Analytic Approaches to Law: 43–51. Ashgate-Dartmouth.
——— (1999). Max Weber’s “Ancient Judaism”. Theory Culture and Society 16(1): 31–40.
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A Simplest Systematics For The Organisation Of
Turn-Taking For Conversation. Language 50(4): 696–735.
Schegloff, E.A. (1972). Notes on a Conversational Practice: Formulating Place. In D. Sudnow (ed.),
Studies in Social Interaction: 75–119. The Free Press. [reprinted in P.P. Giglioli (ed.) (1972). Language
and Social Context: 95–135. Penguin.]
214 Rod Watson
Schegloff, E.A. & H. Sacks (1974). Opening Up Closings. In R. Turner (ed.) (1974). Ethnomethodology:
233–264. Penguin.
Schenkein, J. (ed.) (1978). Studies in the Organisation of Conversational Interaction. Academic Press.
Smith, D.E. (1974). Theorizing as Ideology. In R. Turner (ed.), Ethnomethodology: 41–44. Penguin.
Watson, R. (1997). Some Reflections on “Category” and “Sequence” in the Analysis of Conversation.
In S. Hester. & P. Eglin (eds.), Culture in Action: 49–76. University Press of America.
Sequence
Jack Sidnell
University of Toronto
1. Introduction
In conversation, actions are not arranged serially, one after the other, like so many march-
ing penguins but rather come grouped together in various ways. An answer, for instance,
responds to a question and the two form together a paired unit. Some of the ways in
which actions are grouped together and related to one another so as to form sequences
are discussed in what follows. This leads to a consideration of the distinctive way in which
understanding is achieved and sustained in conversation. It also leads to an examination
of “preference” – a set of biases which operate across a wide range of sequence types.
Let’s begin by looking at a very brief recorded telephone conversation. Here Janet
has called Anne who apparently has Janet’s daughter at her house.
(1) xtr. 1
〈〈ring〉〉=
1 A: =Hello::
2 J: Oh=hi:_=it’s Janet_ [Cathy’s mo]m
3 A: [hi: Janet]
4 How eryou(h) .h hh
5 J: I’m goo:d,how are y[ou
6 A: [I’m fi:ne.h [we’re
7 J: [°good°
8 A: actually: uhm (0.2)
9 we’re running bit late=but we’re
10 (.)
11 on our wa(h)y:
12 J: a→ Do you want me to come an’ get her?
13 A: Uhm:, it doesn’t matte:r, like(hh)
14 (0.4)
15 .hhhhh
16 J: a→ I- I could.it’s very easy.
17 so rather than you h:av(h)e(h)
18 (.)
19 you know (.) tuh get everybody ou[t
20 A: [.hhh
21 J: I’ll justa
22 (0.2)
216 Jack Sidnell
23 come dow:n.=
24 A: =we:ll,my mom’s here, so I don’t have to uhm::
25 (.)
26 I don’t have to put my kids in the ca:r.
27 J: a→ I’ll ju[s- I’ll just come down.
28 A: [bu
29 b→ ↑Oka(h)y(h)=
30 J: =yeah
31 A: ok(h)ay
32 J: uh what’s yo[ur number again?=
33 A: [.hhhhh
34 =It’s seven-seven eight.
35 J: Okay great.
36 A: ↑Oka:y [thanks]
37 J: [thanks]
38 A: bye.
There are many ways this brief conversation could be analyzed. For instance, one could
focus on the topics of the conversation i.e. what these people were talking about (Anne’s
Mom, Anne’s kids, the car?). An alternative would be to examine the talk for the iden-
tities which the participants enact (e.g. “mother”) or the particular face arrangements
which their talk implicates. What distinguishes the conversation analytic approach
from these alternatives is a focus on the participants’s own understandings as these are
revealed in the talk itself. So the first thing we need to ask is what these people under-
stood themselves to be doing here. We need to begin, that is, by asking what actions
are being accomplished in this conversation. Let’s begin with talk at line 12 reproduced
here for convenience as (2):
(2) XTR 1 – Fragment
14 (0.4)
The turn at line 12 could be characterized in several different ways. If we look at the
linguistic form of the turn, we find that the auxiliary verb “do” occurs before the pro-
noun “you” and thus the sentence involves subject-auxiliary inversion (see Quirk
1985). Moreover, the turn at line 12 is produced with a markedly rising intonation
contour indicated by the question mark at its completion. These are two ways in which
speakers of English convey that they are asking a question (see Schegloff 1984). If the
talk in line 12 is a question, it seems likely that Anne’s response is an answer to it.1
. In fact, there are good reasons to suppose that responses such as this do not count as answers
for the participants.
Sequence 217
The question and answer pair together form a small sequence, what we will come to
call below, an adjacency pair. Or do they?
That Janet’s talk at line 12 poses a question is clear enough, but she seems also to
be doing something else. Specifically, Janet seems to be offering to come and pick up
her daughter from Anne’s house. In a moment we will review several types of evidence
for this claim that Janet is making an offer. However, before doing so, it may be useful
to briefly consider the relationship between the question and the offer in this example.
Describing this turn as a question captures something important about its design –
specifically about the use of recurrent and stable features of English grammar (and into-
nation) which distinguish “Do you want me to come over and get her?” from “I’m com-
ing to get her” etc.. Because these are recurrent and relatively stable features which recur
across a wide range of utterance types and actions, we call them “practices of speak-
ing” or, simply, “practices.” Describing this turn as an offer, on the other hand, captures
something important about what this speaker, on this occasion, is doing. An offer then
is an action. Analysis typically involves attention to both the actions which are being
accomplished by a turn and the practices of speaking which make them happen.2
So what evidence, beyond our own intuitions, do we have for the claim that Janet
is making an offer when she says “Do you want me to come an’ get her?” First, we can
quite readily find other examples of just this construction used in apparently similar
ways. For instance:
(3) Heritage 1:3:4
1 Ile: =Wellnow look.
2 → d’you want me ti[h come over’n get her?=or wha:t.
3 Lis: [°( )°
4 Lis: ↑Just please yerself dear we were g’nna t-bring’er
5 ↑back b’t chor very wel[come
6 Ile: [No well when you- when you
7 going to bring her ↓ba:ck.=
Or
(4) Holt 5/88:2:4:
1 Dee:→ So: are you going to go -back that night or d’you want
2 (.) a bunk bed or somethi[ng
3 Mar: [nNo we’ll go back thank you
4 very much it’s not very far
2. This is not to say that “questioning” cannot itself be the action with which a turn is occupied.
For that reason it is useful to distinguish between interrogative “practices of speaking” which
are recurrent features of grammar and questioning as an action. Notice that a turn can do ques-
tioning even if it has no interrogative features (See Sidnell frth).
218 Jack Sidnell
5 (.)
6 Dee: Are you su:re becuz we’re not having anybody to stay
7 here I mean we (.) we c- (.) we could
8 accomoda[te you:=
9 Mar: [.hhhhhh
10 Dee: =uh:[m
11 Mar: [n:No:: (.) no it’s alright it’s not (.) not
12 particu’rly fa:r a couple of hours ‘n we’re
13 home again.
Here are two examples in which speakers use the “do you want X” construction to
make offers. Notice that this kind of evidence involves a focus on the “practices of
speaking” involved in our example and requires that we look across a collection of
instances. A second kind of evidence is provided by what comes next in the sequence.
That is, we can look to see how the participants themselves treated the utterance in
responding to it. Consider Example 4 in this respect. Here Deena has invited Mark,
her cousin, to stay overnight after her daughter’s wedding. She does this by offering “a
bunk bed or something.” Mark’s response answers the question which is the vehicle for
the offer (“no”) and, at the same time both registers the offer and rejects it with “we’ll
go back.” Moreover, Mark’s “thank you very much” responds to Deena’s turn explicitly
in terms of its status as an offer.
We can access a third type of evidence by looking at the relation between the focal
turn and the previous one. That is, we can look at the talk to which the focal turn was
itself designed as a response. In this respect, it can be observed that offers are frequently
produced in response to a description of speaker-trouble (e.g. “my car is stalled,”). Notice
then that Janet’s offer to “come and get her” follows directly on the heels of Anne’s “we’re
running a bit late.” To summarize, we can draw on three kinds of evidence, beyond our
own intuitions, in making claims about what a given utterance is doing. First, we can
collect other instances of the same construction or, more generally, the same practices
of speaking to see if they do the same or similar jobs in those other cases. Secondly, we
can look at the response which the focal turn elicits from a recipient. Thirdly we can
examine the interactional circumstances which occasion the focal turn and, in particu-
lar, the immediately preceding turn to which it is built as a response.
In Example 4 we saw an offer rejected. In Example 1, Anne seems neither to
accept nor reject Janet’s offer in the immediately subsequent turn. Rather, Anne’s
“Uhm: it doesn’t matter” withholds a response to the offer. Moreover, after coming
to completion of her turn, Anne starts again only to trail-off the production of “like”
in this way providing Janet with another opportunity to speak. Thus, a sequence is
initiated by Janet’s offer and the recipient withholds a response to that offer in the
position where it is due. With, “it doesn’t matter” Anne puts the ball back in Janet’s
court so to speak.
Sequence 219
(5) XTR 1 – fragment 3

14 (0.4)
15 .hhhhh
16 J: a→ I- I could.it’s very easy.
18 (.)
20 A: [.hhh
21 J: I’ll justa
22 (0.2)
23 come dow:n.=
24 A: =we:ll,my mom’s here, so I don’t have to uhm::
25 (.)
26 I don’t have to put my kids in the ca:r.
27 J: a→ I’ll ju[s- I’ll just come down.
28 A: [bu
29 b→ ↑Oka(h)y(h)=
30 J: =yeah
31 A: ok(h)ay
Rather than launch a new course of action at line 16, Janet continues what was begun in
line 12. There are various ways in which an offer may be declined. Although declination
may be accomplished by a simple “no thank you,” often it involves the recipient giving a
reason or account for why the offer is not being accepted (see below). For instance, the
recipient may claim not to want or need the thing being offered as in Example 4. Alterna-
tively, the recipient of an offer may decline it on the basis of the burden acceptance would
create for the one making it. The one making the offer may themselves orient to potential
grounds for declining it. Here, in line 16, Janet’s “I- I could.it’s very easy” pushes the offer
forward by characterizing the task as a minimal inconvenience.3 Janet then goes on to
contrast the course of action which she is proposing with the alternative of Anne coming
to Janet. She concludes with, “I’ll just come down,” which, in this context, transforms
the offer such that it no longer hinges on Anne’s wants. In this sense it is strongly built
towards acceptance which it subsequently elicits from A with “okay.”
We can thus see how this talk forms a sequence of action initiated with an offer by
Janet to Anne and eventually brought to completion with an acceptance of that offer
by Anne. Now that we have seen something of this organization in action, we can turn
to describe its technical features.
. See also the talk at lines 6–8 of Example 4 for a parallel case.
220 Jack Sidnell
2. The adjacency pair
A great deal of talk is organized into sequences of paired actions or adjacency pairs.
For instance, a question creates a ‘slot’, ‘place’, or ‘context’ within which an answer is
relevant and expected next. Schegloff (1968: 1083) noted that such paired actions pose
two basic problems for analysis. First,
How can we rigorously talk about two items as a sequenced pair of items, rather than
as two separate units, one of which happens to follow the other?
And secondly,
How can we, in a sociologically meaningful and rigorous way, talk about the “absence”
of an item; numerous things are not present at any point in a conversation, yet only
some have a relevance that would allow them to be seen as “absent.”
Recognition of these problems led Schegloff (1968: 1083) to introduce the concept of
conditional relevance:
By the conditional relevance of one item on another we mean: given the first, the
second is expectable; upon its occurrence it can be seen to be a second item to the first;
upon its nonoccurrence it can be seen to be officially absent – all this provided by the
occurrence of the first item.
So questions are not always followed by answers. However, the conditional relevance
that a question establishes ensures that participants will inspect any talk that follows
a question to see if and how it answers that question. In other words, the relation-
ship between paired utterance-types such as question and answer is a norm to which
participants themselves orient in finding and constructing orderly sequences of talk
(see Goodwin & Heritage 1990; Heritage 1984). Adjacency pairs are, then, frameworks
of accountability which provide for the intelligibility of action. Schegloff and Sacks
(1973) identified four defining characteristics of the adjacency pair. It is composed of
two utterances that are:
i. Adjacent.4
ii. Produced by different speakers.
iii. Ordered as a first pair part (FPP) and second pair part (SPP).
iv. Typed, so that a particular first pair part provides for the relevance of a particular
second pair part (or some delimited range of seconds, e.g. a complaint can receive
a remedy, an expression of agreement, a denial as its second).
4. See the discussion of insert sequences.

Sequence 221
Discussing the last feature in this list, Schegloff (2007: 13) writes:
the components of an adjacency pair are pair-type related. That is, not every second
pair part can properly follow any first pair part. Adjacency pairs compose pair types;
types are exchanges such as greeting-greeting, question-answer, offer-accept/decline,
and the like. To compose an adjacency pair, the FPP and SPP come from the same pair
type. Consider such FPPs as “Hello,” or “Do you know what time it is?” or “Would you
like a cup of coffee?” and such SPPs as “Hi,” or “Four o’clock” or “No thanks.” Parties
to talk-in-interaction do not just pick some SPP to respond to a FPP: that would yield
such absurdities as “Hello,” “No thanks,” or “Would you like a cup of coffee?” “Hi.” The
components of adjacency pairs are “typologized” not only into first and second pair parts,
but into the pair types which they can partially compose: greeting-greeting (Hello,” Hi”),
question-answer (“Do you know what time it is?” “Four o’clock”), offer-accept/decline
(“Would you like a cup of coffee?” “No thanks,” if it is declined).
What kind of organization is the adjacency pair? It is not a statistical probability and
clearly not a categorical imperative. Rather, the organization described is a norm to
which people hold one another accountable. The normative character of the adja-
cency pair is displayed in participants’ own conduct in interaction. For example, as
the principle of conditional relevance implies, when a question does not receive an
answer, questioners treat the answer as “noticeably” absent. Questioners’ orientation
to a missing answer can be seen in three commonly produced types of subsequent
conduct: pursuit, inference and report.
In the following example (from Drew 1981) mother asks the child, Roger, what
time it is.
(6) Drew 1981: 249
1 M: What’s the time- by the clock?
2 R: Uh
3 M: What’s the time?
4 (3.0)
5 M: (Now) what number’s that?
6 R: Number two
7 M: No it’s not
8 What is it?
9 R: It’s a one and a nought
After Roger produces something other than an answer at line 2, mother repeats the
question at line 3. Here then a failure to answer prompts the pursuit of a response (see
Pomerantz 1984b). When this second question is met with three seconds of silence,
Mother, transforms the question now asking, “what number’s that?” Note that the first
question, “What’s the time?” poses a complex task for the child. The child must first
identify the numbers to which the hands are pointed and subsequently use those num-
bers to calculate the time. Here, in response to a failure to answer, mother takes this
222 Jack Sidnell
complex task and breaks it down into components. She has apparently inferred that
the child did not answer because the question, as initially put, posed a task that he
could not accomplish. Mother’s subsequent conduct is informed by an inference that
the child did not answer because he was not able to do so.
Although it does not happen here, questioners may also report an absent answer
saying such things as “you are not answering my question,” or “he didn’t answer the
question”, or “she didn’t reply” etc.. In public inquiries, lawyers commonly suggest that
the witness is not answering the question that has been asked of them (see Ehrlich &
Sidnell 2006; Sidnell 2004; Sidnell frth.).
It is important to see the difference between an absence on the one hand and an
“official,” “noticeable” or “accountable” absence on the other. An infinite number of
things can be accurately described as absent after the occurrence of a first pair part.
The next speaker did not blow his nose, scratch his head, jump up and down, sing “O’
Canada” etc.. The point here is that the first pair part of an adjacency pair has the
capacity to make some particular types of conduct noticeably or relevantly absent such
that their non-occurrence is just as much an event as their occurrence.
Would-be answerers also orient to a missing answer. Thus, the non-occurrence
of an answer typically occasions an account for not answering. One particularly com-
mon account for not answering is not knowing. The following examples from Heritage
(1984) illustrate this:
(7) From Heritage 1984 (W:PC:1:MJ(1):18)
1 J: But the trai:n goes. Does th’train go o:n th’boa:t?
2 M: °h °h Ooh I’ve no idea:. She ha:sn’t sai:d.
(8) From Heritage 1984 (Rah:A:1:Ex:JM(7):2) – simplified
1 M: Is he alri:ght?,
2 J: Well he hasn’ c’m ba-ack yet.
In the first example the speaker accounts for not answering by saying she has “no
idea” and subsequently accounts for not knowing by saying “she hasn’t said.” In the
second example, the speaker produces only an account for not knowing and by this
implies an account for not answering. To summarize, at the heart of many sequences
are adjacency pairs which organize two turns via a relation of conditional relevance.
An orientation to the normative character of adjacency pairs is revealed in various
aspects of the participants’ own conduct (questioners’ pursuit, inference and report
and answerers’ accounts).
Many explanations of human conduct involve people learning or otherwise inter-
nalizing patterns, scripts, procedures, ideologies, cultural values etc.. Scholars in a
number of traditions have recently pointed to the difficult questions about “structure”
and “agency” to which such explanations inevitably lead. This is the old philosophical
problem of determination and freewill. The question arises as to how much of what
Sequence 223
we do is a matter of self-conscious decision-making and how much is a matter of

shaping by a hidden hand. As long as the questions are posed in this way it seems as
though we must either be puppets or gods. The preceding consideration of adjacency
pairs hints at the distinctive way in which this issue is conceptualized within conversa-
tion analysis. CA is not deterministic since people may or may not do the normatively
called for action. On the other hand, it is not just a matter of people doing whatever
they want either since in making sense of what others are up to conversationalists are
also holding them accountable to the dictates of normative structures. Looking at the
sequential organization of talk we see that, from the perspective of conversation analy-
sis, “each speaker in conversation participates in an organization whose contingent
outcomes are only partially under his or her control” (Heritage 1984) and further that,
“conversationalists are … not so much constrained by rules or sanctions, as caught in
a web of inferences” (Levinson 1983: 321).5
3. “A context of publicly displayed and continuously up-dated

intersubjective understandings”
The relatedness of the two turns in an adjacency pair has both a prospective and retro-
spective dimension. Thus, as discussed above, the occurrence of a first pair part creates
a slot for a specific second pair part. At the same time, a second pair part displays its
speaker’s understanding of the first to which it responds. Adjacency pairs allow then
for a framework of understanding that is constructed and sustained turn-by-turn. If a
speaker responds to an FPP with an inappropriate SPP, the speaker of the FPP can see
that they have not been understood. They can then go back and initiate repair. This is
precisely what happens in Example 9.
(9) Third turn repair – from Schegloff 1992.

1 Annie: Which one:s are closed, an’ which ones are open.
2 Zebrach: Most of ‘em. This, this, [this, this ((pointing))
3 Annie: → [I ‘on’t mean on the
4 → shelters, I mean on the roads.
5 Zebrach: Oh!
6 (0.8)
7 Zebrach: Closed, those’re the ones you wanna know about,
8 Annie: Mm[hm
9 Zebrach: [Broadway…
5. Schegloff suggests that conditional relevance has the status of a social fact.
224 Jack Sidnell
Here, Annie and Zebrach are looking at a map together and, at line 1, Annie asks
Zebrach “which ones” are closed and which are open. In his response at line 2, Zebrach
indicates which of the “shelters” are open. This turn reveals a misunderstanding of the
question and specifically the reference with “which ones” in line 1. Zebrach’s response
reveals to Annie that he has understood her to be asking about the shelters when in
fact she meant to be asking about the roads. The problem is addressed via third posi-
tion repair at lines 3–4 (see Schegloff 1992).
Participants in conversation look to a next turn to see if and how they have been
understood. As analysts we can exploit the same resource. This is sometimes called
the next turn proof procedure. Consider the following fragment from one of Sacks’
recordings of the Group Therapy Sessions.
(10) (Sacks 1995 vI: 281).
1 Roger: On Hollywood Boulevard the other night they were
2 giving tickets for dirty windshields ((door opens))
3 Jim: hh
4 Therap: Hi, Jim [c’mon in.
5 Jim: [H’warya
6 Therap: Jim, this is uh Al,
7 Jim: Hi
8 Therap: Ken,
9 Jim: Hi
10 Ken: Hi
11 Therap: Roger.
12 Roger: → Hi
13 Jim: Hi
14 Therap: Jim Reed.
Sacks (1995 [1966]) draws attention to “the prima facie evidence afforded by a subse-
quent speaker’s talk” in his analysis of the therapist’s turns at 8 and 11 as recognizable
introductions. Thus, when, at line 12, Roger responds to the
utterance with his name (…) not with “What” (as in an answer to a summons), indeed
not with an utterance to the therapist at all, but with a greeting to the newly arrived
Jim, he shows himself (to the others there assembled as well as to us, the analytic
overhearers) to have attended and analyzed the earlier talk, to have understood that
an introduction sequence was being launched, and to be prepared to participate by
initiating a greeting exchange in the slot in which it is he who is being introduced
(Schegloff 1995: xliii).
Thus a response displays a hearing or analysis of the utterance to which it responds. As

noted, such a hearing or analysis is “publicly available as the means by which previous
speakers can determine how they were understood” (Heritage 1984). The third position
in a sequence is then a place to accept the recipient’s displayed understanding or, alter-
natively, to repair it. Heritage (1984: 259) writes:
Sequence 225
By means of this framework, speakers are released from what would otherwise be
an endless task of confirming and reconfirming their understandings of each other’s
actions … a context of publicly displayed and continuously up-dated intersubjective
understandings is systematically sustained… Mutual understanding is thus displayed
… ‘incarnately’ in the sequentially organized details of conversational interaction.
There is, of course, significant room to maneuver within this framework. Thus, Goodwin
and Goodwin write that, “rather than presenting a naked analysis of the prior talk, next
utterances characteristically transform that talk in some fashion – deal with it not in its
own terms but rather in the way in which it is relevant to the projects of the subsequent
speaker” (Goodwin & Goodwin 1987 cited in Heritage 1984). Consider for instance the
following in which Dick is reporting trouble in getting his family over for the holidays.
(11) Deb and Dick
22 Deb: [ s]o don’t you have all your
23 family coming today?
24 Dick: Well: they’re coming around two and I °hhh left
25 messages with Brian an:d mydad to(uh) see if
26 they wanted to come but=ah
27 (0.2)
28 °hh that’s all I could do was leave messages.
29 Deb: → owh
30 (0.4)
31 Dick: °Gotsome° °hhhh five pound lasagna thing to(hh)
32 throw in the oven=an
33 Deb: → o(h)h(h)=huh (.) well: I’m sure you’ll have a
34 good time.
35 Dick: [oh
36 Deb: [〈at least it’s inside. And it didn’t rain
37 yesterday so we were lucky [l- looking at it
38 Dick: [mmhm yeah
39
40 Deb: today god. woulda been awful.
At line 22–23 Deb asks whether Dick’s family is coming to visit. In response, Dick indi-
cates that although some members of the family are expected, he has been unable to get a
hold of “Brian” and his “Dad.” The answer concludes with Dick saying, “that’s all I could
do was leave messages.” The selected extreme case formulation (see Pomerantz 1986)
might well have elicited an expression of sympathy. Instead, however, Deb registers the
report with a prosodically non-committal “owh”. After a pause, Dick continues saying,
“°Gotsome° °hhhh five pound lasagna thing to(hh) throw in the oven=an” . The selec-
tion of “some”, rather than “a”, and “throw”, rather than “put”, endow Dick’s description
with a sense of pathos. By adding “thing” to the description of the food, Dick does not
even allow it to be unequivocally categorized as “lasagna”. Moreover, Dick’s description
of the lasagna in terms of its weight characterizes it as something to be eaten but not
226 Jack Sidnell
necessarily enjoyed. Taken together these choices in description convey Dick’s negative
stance towards what he is saying and suggest that he may have “troubles” to tell Deb,
however, blocks the projected troubles-telling by saying, “I’m sure you’ll have a good
time” and “at least it’s inside.” The “at least” here is a device for retrieving the good in
otherwise bad news. Used here, it also allows Deb to move step-wise into her own topic.
In this example, then, Deb resists being aligned as a “troubles recipient” (see Jefferson
1988) by dealing with Dick’s talk, as Goodwin and Goodwin say, “not in its own terms
but rather in the way in which it is relevant to the projects of the subsequent speaker.”
Since this is a point of crucial importance, one further illustration will be considered.
In the following, Dee has invited her cousin Mark to her daughter’s wedding. She
has decided not to invite Mark’s mother (her aunt) after whom Dee is asking in line 1.
(12) Holt 5/88:2:4:3

1 Dee: You doh-uh how is she[Mark.
2 Mark: [.hhh We:ll she’s-:-: r-
3 physically n-: she is: gettin’ a bit o:ld you
4 kno:w, uh::m .hhhhhh she can’t get around quite
5 so much although she still does ‘er shopping
6 still looks a:fter ‘erse:lf I mean it’s amazing
7 really, u-uh::: .hhhhhh She drives us nu:ts but
8 stihhhll[h-h .hu .hu
9 Dee: [Oh I know love I do know the feeling
9 you’ve got my every sy[mpathy
10 Mark: [phhh! ↑heh heh e-[he
11 Dee: [( ).
12 Mark: .hhhhhh[hh
13 [We: had it,
14 (.)
15 believe you me we had[it.
16 Mark: [hI: kno:w.[hYe:s
Dee has phoned to tell Mark that she has decided not to invite his mother to her
daughter’s wedding ostensibly so as not to inconvenience Mark and his wife. As such,
Dee has something invested in hearing, in response to her query at line 1, that she
(Mark’s mother) is not well, this having been presented as grounds for not inviting her
to the wedding. In his response, Mark seems initially to align with Dee’s evaluation
of his mother’s health (with e.g. “bit old,” “can’t get around”) but at line 4 he moves in
the opposite direction detailing the extent to which she is, in fact, able to get around
and look after herself. Note that there are several points within the turn at which Dee
might have treated Mark’s report about his mother’s heath as possibly complete. Dee,
however, holds off responding until the point at which Mark remarks that his mother
“drives us nuts.” “Drives us nuts,” confirms in no uncertain terms Dee’s already expressed
Sequence 227
conviction that the mother would hamper Mark and his wife’s enjoyment of the wed-
ding. It is here that Dee responds with an empathetic “I know…”. A report of this kind
then can be seen as the product of an interaction between speaker and recipient. The
recipient plays an active role in the production of the report in at least two ways. In the
first place, Dee selects a particular place in the course of Mark’s talk at which to begin a
response and thereby helps to construct its completion. Secondly, of the many ways in
which she might have responded – e.g. with optimism, with surprise, with skepticism,
etc. – Dee’s expression of sympathy (“Oh I know love I do know the feeling you’ve got
my every sympathy”) treats Mark’s report about his mother’s health as a complaint.
Dee hears in what Mark says here a description of the ways in which the mother has
become a burden and it is this hearing that she displays in her response to it.
Heritage (1984), from whom I have borrowed a great deal for the discussion in
this section, notes that “observations concerning the way a turn’s talk displays an
analysis, appreciation or understanding of a prior turn do not simply apply to …
responses or ‘reactive’ second utterances. They also apply to ‘first’ or initiatory actions
… which, in their own various ways, also display analyses of the ‘state of talk’. ” As
Heritage notes a speaker who initiates the pre-closing of a call with “well::” or some
other “possible pre-closing” displays an analysis of that “ ‘there and then’ as an appro-
priate place for that to occur.” Indeed, any utterance can be heard as, in some sense,
exhibiting its speaker’s sense of the current state of talk and the context in which that
talk is produced.
4. Preference
Schegloff (2007) writes, “In the vast majority of sequence types, there are not only
alternative responses which a first pair part makes relevant and a recipient of a first pair
part may employ; there are alternate types of response. These embody different align-
ments toward the project undertaken in the first pair part.” Thus, a request can either
be granted or it can be rejected. Of course, there are many ways in which a request may
be granted (e.g. “sure”, “I guess”, “of course, you needn’t ask” etc.) or rejected (“I don’t
think that would be a good idea”, “It’s just not possible”, “no way!”) but as Schegloff,
again, points out “accepting and declining, granting and rejecting, are fundamentally
different types of responses and alignments.” These are not “symmetrical alternatives”
(Schegloff & Sacks 1973: 14), rather, the response to the first pair part that promotes
accomplishment of the activity is, typically, the preferred one.
Preference, in this context, refers not to psychological states of the participants
but, rather, structural regularities in the talk. Consider a situation in which you
are invited to a dinner party at the house of someone whose company you assidu-
ously avoid. In such a situation, it is likely that you would prefer, in the individual
228 Jack Sidnell
or psychological sense, to decline the invitation but this does not alter the fact that
acceptance is the preferred alternative in terms of the organization of the talk. A
decline will likely require an accompanying explanation, for example you are too
busy, you have a rare and highly contagious fungus etc., whereas an acceptance will
not. In order to see that this is the case, consider the following example:
(13) SBL, T1/S1/C10, simplified
1 A: Uh if you’d care to come and visit a little while
2 this morning I’ll give you a cup of coffee.
3 B: hehh Well that’s awfully sweet of you. I don’t
4 think I can make it this morning. .hh uhm I’m
5 running an ad in the paper and-and uh I have to
6 stay near the phone.
In this extract from a telephone conversation, A’s turn at 1–2 is an invitation to B

to come and visit “a little while” and have a cup of coffee. In this context, B’s talk at
lines 3–5 clearly constitutes a decline of the invitation. Notice, however, that B never
says “no thanks”. How then does this talk nevertheless do declining – what features
of the talk make it clear to A that B does not accept the invitation? In fact, this turn
collects together many of the features characteristic of dispreferred responses (see
also Kitzinger 2000).
i. Delays: B delays the production of the decline by prefacing the turn with audible
breathing (“hehh”) and “well” and “that’s awfully sweet of you”. In terms of
positioning, dispreferred responses are often delayed both by interturn gap and
turn-initial delay.
ii. Palliatives: It is typical for dispreferred responses to contain some kind of appre-
ciation, apology and/or token agreement by which the overwhelmingly negative
valence of the turn is mitigated. Here the dispreferred response is accompanied by
the appreciation “that’s awfully sweet of you”. The declination is further mitigated
by being framed with “I don’t think…”.
iii. Accounts: Dispreferred responses typically contain explanations or justifications
indicating why the invitation is being declined.
Here B explains that she is prevented from accepting the invitation by the
fact that she is “running an ad in the paper” and thus must “stay by the phone.”
With accounts of this sort, people typically suggest that they are unable to accept
the invitation (or grant the request) not simply unwilling. Note, further, that the
account is designed to describe the relevant details of B’s situation which prevent
acceptance of the invitation (B has to stay by the phone (see Drew 1984)).
Returning to the distinction between structural preferences of talk-in-interaction

and personal preferences of individuals we can see from this example that they are
Sequence 229
quite different. There is no way for us to know whether B would “prefer” to accept or
decline the invitation. On the other hand, it is clear that she builds her decline of the
invitation as a dispreferred rather than preferred response.
5. Structural consequences of preference organization
The features of dispreferred turns mentioned above, particularly delay, are resources
which the speaker of a first pair part can use to project or anticipate the imminent pro-
duction of a dispreferred response. Anticipating a dispreferred response, the speaker
of the first pair part may take measures to prevent it from being produced. Thus, the
features of dispreferred second pair parts play a role in minimizing the chance that
such a response will, in fact, ever be articulated. Consider,
(14) Levinson 1983
1 C: So I was wondering would you be in your office
2 on Monday (.) by any chance?
3 (2.0)
4 Probably not
C’s talk at lines 1–2 is building towards some kind of request (although the exact
nature of that request is not available to us). Moreover, the design of the question is
constructed so as to prefer an affirmative answer – compare, “I don’t suppose you’ll be
in your office on Monday?” (see Sacks 1987[1973]). Note that, after coming to possible
completion after “Monday”, the request is extended a little bit with “by any chance?”.6
When the recipient does not respond immediately allowing a sizeable two seconds of
silence to develop, C treats this as foreshadowing a negative answer and partially with-
draws the (pre)-request by reversing the valence of the question with “probably not”.
Alternatively, rather than withdraw the action in the face of a projected dispre-
ferred response, the speaker can add further talk in an effort to make it more appealing
or easier to accept. Consider:
(15) Levinson 1983
1 A: C’mon down he:re,=it’s oka:y,
2 (0.2)
3 I got lotta stuff,=I got be:er en stuff.
Here a slight pause after the invitation reaches possible completion provides an oppor-
tunity for its speaker to ‘upgrade’ the action being implemented – here by adding ‘beer’.
6. This “by any chance” looks very much like what Davidson (1984) called a “monitor space.”
230 Jack Sidnell
In most cases the preferred response is the one which advances or aligns with the
action launched by the first pair part. However, there are at least two exceptions to this
general rule. First, there are actions that, by their very nature, inherently prefer dis-
aligning responses. For instance, both self-deprecating assessments and compliments
appear to prefer disagreeing rather than agreeing responses (Pomerantz 1978).
In addition to particular actions that prefer non-aligning responses, we must
be alert to the potential for general preferences to be defeated or even reversed in
certain contexts. For instance, although offers seem to generally prefer acceptances,
they can also be built to prefer declines. An offer such as, “why don’t I come over
and pick her up” is built towards acceptance whereas one such as “do you want me to
come an’ get her” may not be. The emphasis placed on the recipient’s needs or wants
in the latter appears to invite special treatment. Consider the example with which
we began:
(16) XTR 1 -
12 J: Do you want me to come an’ get ↑her?
14 (0.4)
15 .hhhhh
16 J: I- I could.it’s very easy.
18 (.)
20 [.hhh
21 I’ll justa
22 (0.2)
23 come dow:n.=
Here Anne’s turn at lines 13–15 is clearly built as a dispreferred response. The turn
begins with delay. Moreover, as noted above, Anne’s talk in 13 does not contain either
an acceptance or declination of the offer. Janet’s talk at lines 16–23 suggests that she
hears Anne’s dispreferred turn as adumbrating an acceptance rather than declination.
This then suggests that the offer is built to prefer a decline.
Preference is also implicated in the organization of sequences in so far as partici-
pants routinely work to avoid the production of dispreferred first actions as well. In
order to see how this is accomplished we need to consider the various ways in which
sequences can be expanded.
6. Sequence organization
So far we have considered sequences composed of only two turns – a first and second
pair part. Clearly, sequences can be much more complex than this. Much of this added
Sequence 231
complexity is the product of expansion. An adjacency pair can be expanded prior to

the occurrence of its first part, after the occurrence of its first part but before the occur-
rence of its second, or after its second pair part.
Expansions are typically built out of paired actions and can themselves serve as the
bases upon which further expansion takes place. It does not take much to see how this
could result in sequences of considerable complexity and length (see Schegloff, 2007).
As the name implies, pre-expansions involve an expansion of a sequence prior to the
occurrence of a base first pair part. Pre-expansions are, in a basic sense, preparatory to
some other, projected work to be done in the sequence (the action of the first pair part).
Some pre-expansions are type-specific, that is, they project a specific base first
pair part, e.g. they are pre-invitations “hey, are you busy tonight?”, pre-announcements
“Guess what happened to me?”, pre-requests “You wouldn’t happen to be going my
way would you?” (see Levinson 1983; Schegloff 1988; Terasaki 2004 [1976]). Such
type-specific pre-expansions typically check on a condition for the successful accom-
plishment of the base first pair part. So, for instance, pre-invitations may check on the
availability of the intended invitee. Consider the following:
(17) HS:STI,1
1 A: Judy?
2 B: Yeah,
3 A: John Smith
4 B: Hi John
5 A: How ya doin’=
6 → =say what’r you doing.
7 B: → Well, we’re going out. Why.
8 A: Oh, I was just gonna say come out
9 and come over here an’ talk about
10 this evening, but if you’re going
11 out you can’t very well do that.
Judy’s “why” at line 7 displays an orientation to the preceding turn as something

more than an information-seeking question. John’s answer at lines 8–11 confirms this
inference. Pre-requests typically check on the availability of the item to be asked for.
Consider the following example in which the participants have been talking about
some fish tanks belonging to Vic. At lines 1–2, Mike inquires about one particularly
large tank.
(18) US. 24
1 M: Wuhddiyuh doing wh dat big bow-puh-tank.
2 Nothing?
3 (0.5)
4 V: ((cough))
5 Uh-h-h
232 Jack Sidnell
6 (1.0)
7 V: I’m not intuh selling it or giving it. That’s it.
8 M: Okay
9 (1.0)
10 M: → Dat wz simple. Khhhh huh-huh-heh=
11 V: → =Yeh.
Vic hears Mike’s inquiry as building towards a request (or possibly an offer to
buy the tank) and responds by indicating that it will not be granted. In both cases
(Examples 17, 18), the second part of the pre-expansion blocks the doing of a
projected first pair part (an invitation in 17 and a request in 18). In this way, pre-
expansions work to prevent the occurrence of dispreferred second actions (e.g.
declinations, rejections). Presequences may also be seen to contribute to the non-
occurrence of dispreferred first actions. For instance, requests are dispreferred
relative to offers.7 Consider the following:
(19) Levinson 1983: 343
1 C: Hullo I was just ringing up to ask
2 if you were going to Bertrand’s party
3 R: yes I thought you might be
4 C: Heh heh
5 R: Yes would you like a lift?
6 C: Oh I’d love one
Here, C asks whether R will be going to Bertrand’s party. This is hearable as leading
to a request such as “can I get a ride with you?” Indeed, it could be said to check on
the availability of the service to be requested. R clearly anticipates this and, before the
request is produced, makes an offer.
These participants, then, treat questions such as “Wuhddiyuh doing wh dat big
bow-puh-tank” as preliminary moves to other, projected, actions. Recipients of such
questions are thus given an opportunity to show the way in which the action projected
(request, invitation) will fare. Recipients of such pre-requests or pre-invitations can
give a response which either encourages the other to go-ahead with or to abort the
projected action. Alternatively they can respond to the action as if it had already been
articulated – as the recipients in both 17 and 18 do.
Announcements and story-tellings are also frequently initiated with pre-expansions:
pre-announcements or story prefaces. It was suggested earlier that pre-expansions
typically check on a condition for the successful accomplishment of the base first
pair part. In the case of announcements and other tellings, perhaps the most important
7. Showing that this is, in fact, the case would take us too far afield (see Schegloff 2007).
Sequence 233
and basic of such conditions is that the recipient does not already know the thing to be
told or announced (this is the principle of recipient design, see Sacks et al 1974; Sacks
1995 [1971]). It should thus come as no surprise to find that pre-announcements often
take the form of questions about what the recipient knows.
(20) HG
1 Hyla:→ D’you know w’t I did t’day
2 I wz so proud a’my[s e l]f,=
3 Nan: [What.]
4 Hyla: =?hh I we:nt- (0.2) A’right like I get off
5 et work et one,=
6 Nan: Uh hu:h,=
In 20, Nancy responds to Hyla’s pre-announcement “D’you know w’t I did t’day” with
a go-ahead response which suggests that she does not know the thing to be told. In the
following fragment, Vivian’s pre-announcement does not elicit an audible response
from the co-participants but Vivian nevertheless produces the announcement in the
following turns.
(21) Chicken Dinner
16 SHA: [Uh wz goi:n
17 crazy tihday uh on th’on the roa:d
18 (0.2)
19 VIV:→ We’ yih know w’t he di[↑:d?
20 SHA: [Wen’outta my
21 fuckin’mi[:nd.
22 VIV: [He maHHde
23 (.)
24 VIV: a right- it wz- in Sanna Monig’yihknow
One type of pre-expansion, the summons-answer sequence, is generic and checks on

the precondition of any interactional project – the availability of a projected recipient
to attend the talk for which s/he is selected as recipient.
(22) Atkinson and Drew 1979: 46

1 Ch: → Mummy
2 M: Yes dear
3 (2.1)
4 Ch: I want a cloth to clean the windows
As Schegloff (1968) noted, a summons projects further talk by the same speaker.
Participants orientation to this is seen in deferring (e.g. “wait a minute”), go-ahead (e.g.
“what?”) and blocking (e.g. “I’m not interested in what you have to say,” “I know, take
out the trash.”) responses. Go-ahead responses, in so far as they orient to imminent
234 Jack Sidnell
further talk, are minimal contributions (similar in this respect to continuers) that claim
as little turn-space as possible. Taken to the extreme, this may result in the collapse of
a sequence into a single turn with an adjustment in gaze direction being treated as
equivalent to a summons’ answer (see Schegloff 2007).
An adjacency pair consists of two adjacent utterances, with the second selected
from some range of possibilities defined by the first. As discussed above, this account
of adjacency pairs is meant to describe members’s own normative expectations.
However, on some occasions, the two utterances of an adjacency pair are not, in fact,
adjacent. In some cases this is because a sequence has been inserted between the first
and second pair part of an adjacency pair (see Schegloff 1972). Such insert expan-
sions can be divided into post-firsts and pre-seconds (Schegloff 2007) according to
the kind of interactional relevancy they address. The most common form of post-
first insert-expansion consists of a next turn repair initiator (NTRI) and its response.
Pre-second insert expansions are oriented not to trouble with the FPP but rather to
trouble with an expected SPP. Such insert expansions routinely address issues that
stand as preconditions to the doing of a preferred SPP (for instance, locating an
object requested in the FPP).
Post-expansions are highly variable with respect to their complexity. Schegloff
suggests that they can be divided into minimal and non-minimal types. Minimal post-
expansions consist of one turn. “Oh” for instance can occur after a second pair part
and thus minimally expand the sequence as in the following example:
(23) HG:II:25:ST Hyla & Nancy – Heritage 1984: 286–87

25 Nancy: =?hhh Dz he av iz own apa:rt[mint?]
26 Hyla: [?hhhh] Yea:h,=
27 Nancy: → =Oh:,
28 (1.0)
29 Nancy: How didju git iz number,
30 (·)
31 Hyla: I(h) (·) c(h)alled infermation’n San
32 Fr’ncissc(h)[uh!
33 Nancy: → [Oh::::. [Oh::::.
“Okay” may also serve as a minimal post-expansion:
(24) YYZ 2 – Deb and Dick

55 Dick: Are they leaving today?
56 Deb: I don’t know
57 Dick: O(hhh)kay(hh)
Non-minimal post-expansions come in a variety of forms. One kind involves commen-

tary upon the just prior sequence which it thereby expands. Consider again:
Sequence 235
(25) US. 24
1 M: Wuhddiyuh doing wh dat big bow-puh-tank.
2 Nothing?
3 (0.5)
4 V: ((cough))
5 Uh-h-h
6 (1.0)
7 V: I’m not intuh selling it or giving it. That’s it.
8 M: Okay
9 (1.0)
10 M: → Dat wz simple. Khhhh huh-huh-heh=
11 V: → =Yeh.
After the request is blocked, the sequence is expanded by a short assessment sequence
that involves Mike characterizing the interaction as “simple” and Vic agreeing. Clearly
there are many other ways in which a sequence may be expanded after the occurrence
of a second pair part (Schegloff 2007).
7. The power of sequential analysis
The discovery of sequential organization in the mid-to-late 1960s more or less coincided
with the emergence of conversation analysis as a field. Others, including rhetoricians,
playwrights, novelists, philosophers and linguists, had pointed to the existence of
paired actions such as question and answer, greeting and greeting etc. What Sacks
(1995[1965]) and Schegloff (1968) discovered was the normative order or logic that
undergirds such paired actions. This made it possible to examine not just paired
actions per se but conversational sequencing and conditional relevance. Since the early
studies of Sacks and Schegloff a large number of sequence-types and underlying prin-
ciples or preferences have been described across a range of ordinary and institutional
settings. Moreover, empirical research on conversation has provided a window into
the organization of human interaction more generally and recent work has pointed
to the implications for human development (Wootton 1997) and human adaptation
(Levinson 2006).
Earlier, I drew a distinction between “practices of speaking” and the actions which
those practices are used to accomplish within some particular bit of talk. The dis-
tinction plays an important role in conversation analysis and distinguishes it from
many other approaches to language use in the human sciences which, for the most
part, emphasize either practices of speaking (such as repetition, or clause structure)
or actions (promises and such). Conversation analysis, on the other hand, exam-
ines the relation between practices of speaking and actions-in-talk within sequences.
236 Jack Sidnell
Nowhere is the power of such an analysis displayed more clearly than in Schegloff ’s
study of “confirming allusions.” The phenomenon Schegloff examines in that paper
appears, at first glance, perfectly simple: repetition. Here are two examples:
(39) Berkeley II: 103–114; simplified (from Schegloff 1996: 174)
1 Evelyn: =Hi: Rita
2 Rita: Hi: Evelyn:. How [are y’
3 Evelyn: [I hadda come in another room.
4 Rita: Oh:. Uh huh.=
5 Evelyn: =I fee:l a bi:ssel verschickert.
6 (0.2)
7 Rita: W- why’s ‘a:t,
8 (0.4)
9 Rita: → uh you’ve had sump’n t’drink.=
10 Evelyn: → =I had sump’n t’dri:nk.
11 Rita: Uh huh.
(40) MDE, Stolen (from Schegloff 1996: 185)

1 Marsha: =He’s flying.
2 (0.2)
3 En Ilene is going to meet im:.Becuz the to:p
4 wz ripped off ’v iz car which is tih ssay
5 someb’ddy helped th’mselfs.
6 Tony: → Stolen.
7 (0.4)
8 Marsha: → Stolen. Right out in front of my house.
Schegloff shows these repetitions to be instances of a previously undescribed action –

“the practice of agreeing with another by repeating what they have just said is shown to
constitute the action of confirming an allusion – that is, confirming both its ‘content’ and
its prior inexplicit conveyance” (Schegloff 1996: 161). In 39, Evelyn reports that she
“hadda come in another room” and goes on to remark “I fee:l a bi:ssel verschickert.”
Schegloff notes that there is an ambiguity here in the Yiddish word “verschickert”
which may refer either to a state brought on by the consumption of alcohol, “tipsy”,
or one arrived at without – “groggy” might serve as a translation for the latter. After,
Rita offers the understanding check “you’ve had sump’n t’drink”, Evelyn produces the
repetition. “In confirming with a repeat, Evelyn not only verifies the supposition as to
how she came to be in the state she is in, but invokes the presence of that meaning in
her own immediately prior talk” (Schegloff 1996: 186). In 40, Marsha is telling her ex,
Tony, that their son is flying, rather than driving, back to northern California where
Tony is. She accounts for what is a change of plans by saying that the “to:p wz ripped
off ’v iz car” and, apparently catching the possible ambiguity of “ripped off ” between
an idiom for robbery and the act of tearing, extends her turn with “which is tih ssay
Sequence 237
someb’ddy helped th’mselfs.” However, although this is meant to disambiguate, it is

again not to be taken literally. As Schegloff says, “the sense of what has happened has,
then, been conveyed without being given its ‘common name.’ When Tony offers ‘stolen’
as literally what happened, Marsha not only confirms that that is what has happened,
but does so with a form that confirms as well that that is what she was conveying in her
prior talk” (Schegloff 1996: 185).
The action of confirming allusions could not have been noticed without careful
examination of the sequences in which it occurs since without attention to the sequence
of which these repetitions are a part, it would not have been possible to see the “allu-
sive” character of the talk that gets confirmed. Moreover, by bringing together instances
of the practice into a collection, Schegloff is able to specify “environments of relevant
occurrence” – that is, interactional environments in which the action could be done,
even if it is not in that instance. This in turn allows Schegloff to identify a case in which
the speaker appears to specifically avoid repetition in order not to confirm an allusion
– the speaker comes to “the very verge of a possible allusion-confirmation before
veering away” (Schegloff 1996: 194). Schegloff (1996: 199) remarks about such instances
that: ‘‘… it is virtually certain that nothing of interest would have been seen at all were
we not already familiar with the practice of confirming allusions and its environments
of possible occurrence. Here then we may have some of the most distinctive fruits of
inquiry in rendering what would otherwise be invisible visible in its very absence.” This
remarkable analysis is the product of many years of research, patiently tracking the
phenomenon and allowing it to come fully into view as instances were collected and
put side-by-side. It well-illustrates the power of sequential analysis.
One way to begin an analysis of a fragment of talk is by blocking out the sequences
that together compose it. This provides a point of entry – a way of accessing some of
the technical details of talk – even if a more complete analysis, such as Schegloff ’s study
of confirming allusions, requires attention not just to sequence organization but also
to turn-design, turn-taking, repair and other aspects of conversational organization.
References
Davidson, J. (1984). Subsequent Versions of Invitations, Offers, Requests, and Proposals.

——— Dealing with Potential or Actual Rejection. In J.M. Atkinson & J. Heritage (eds.), Structures of
Social Action: 102–128. Cambridge University Press.
Drew, P. (1981). Adults’ Corrections of Children’s Mistakes. In P. French & M. MacLure (eds.),
Adult-Child Conversations: 244–267. Croom Helm.
——— (1984). Speakers’ Reportings in Invitation Sequences. In J.M. Atkinson & J. Heritage (eds.),
Structures of Social Action: 152–164. Cambridge University Press.
Ehrlich, S. & J. Sidnell (2006). ‘I think that’s not an assumption you ought to make’: Challenging
presuppositions in inquiry testimony. Language in Society. 35(5): 655–676.
238 Jack Sidnell
Goodwin, C. & M.H. Goodwin (1987). Concurrent Operations on Talk: Notes on the Interactive
Organization of Assessments. IPrA Papers in Pragmatics 1(1): 1–52.
Goodwin, C. & J. Heritage (1990). Conversation Analysis. Annual Review of Anthropology 19: 283–307.
Heritage, J. (1984). Garfinkel and ethnomethodology. Polity Press.
Jefferson, G. (1988). On the Sequential Organization of Troubles-Talk in Ordinary Conversation.
Social Problems 35(4): 418–441.
Kitzinger, C. (2000). Doing feminist conversation analysis. Feminism and Psychology 10(2): 163–193.
Lerner, G. (2004). Collaborative turn sequences. In G. Lerner (ed.), Conversation analysis: studies
from the first generation: 225–256. John Benjamins.
——— (2006). On the human “interactional engine”. In N.J. Enfield & S.C. Levinson (eds.), Roots of
Human Sociality: Culture, Cognition and Interaction. Berg.
Pomerantz, A. (1978). Compliment Responses: Notes on the Co-operation of Multiple Constraints.
In J. Schenkein (ed.), Studies in the Organization of Conversational Interaction: 79–112.
Academic Press.
——— (1984). Pursuing a Response. In J.M. Atkinson & J. Heritage (eds.), Structures of Social
Action: 152–164. Cambridge University Press.
——— (1986). Extreme Case Formulations: A Way of Legitimizing Claims. Human Studies 9: 219–229.
Quirk, R. (1985). A Comprehensive grammar of the English language. Longman.
Sacks, H. (1975). Everyone Has to Lie. In M. Sanches & B.G. Blount (eds.), Sociocultural Dimensions
of Language Use: 57–80. Academic Press.
——— (1987) [1973]. On the Preferences for Agreement and Contiguity in Sequences in Conversation.
In G. Button & J.R.E. Lee (eds.), Talk and Social Organisation: 54–69. Multilingual Matters.
——— (1995). Lectures on Conversation (2 vols.). Blackwell.
——— (1995) [1971]. Spouse Talk. In G. Jefferson (ed.), Lectures on Conversation, vol. 2: 437–443.
Blackwell.
——— (1995) [1965]. Rules of conversational sequence. In G. Jefferson (ed.), Lectures on Conversation,
vol. 1: 3–11. Blackwell.
——— (1995) [1966]. An introduction sequence. In G. Jefferson (ed.), Lectures on Conversation,
vol. 1: 281–291. Blackwell.
Sacks, H., E.A. Schegloff & G. Jefferson (1974). A Simplest Systematics for the Organization of
Turn-Taking for Conversation. Language 50: 696–735.
Schegloff, E.A. (1968). Sequencing in Conversational Openings. American Anthropologist 70:
1075–1095.
——— (1972). Notes on a Conversational Practice: Formulating Place. In D. Sudnow (ed.), Studies in
Social Interaction: 75–119. Free Press.
——— (1979). Identification and Recognition in Telephone Openings. In G. Psathas (ed.), Everyday
Language: 23–78. Erlbaum.
——— (1984). On Some Questions and Ambiguities in Conversation. In J.M. Atkinson & J. Heritage
(eds.), Structures of Social Action: 28–52. Cambridge University Press.
——— (1986). The Routine as Achievement. Human Studies 9(2/3): 111–151.
——— (1988). Presequences and Indirection: Applying Speech Act Theory to Ordinary Conversation.
Journal of Pragmatics 12(1): 55–62.
——— (1992). Repair after next turn: the last structurally provided for place for the defense of inter-
subjectivity in conversation. American Journal of Sociology 95(5): 1295–1345.
——— (1995). Introduction. In G. Jefferson (ed.), Harvey Sacks: Lectures on Conversation I: ix–lxii.
Blackwell.
Sequence 239
——— (1996). Confirming allusions: Toward an empirical account of action. American Journal of
Sociology 104(1): 161–216.
——— (2002). Reflections on Research on Telephone Conversation: Issues of Cross-Cultural Scope
and Scholarly Exchange, Interactional Import and Consequences. In K.K. Luke & T.-S. Pavlidou
(eds.), Telephone Calls: Unity and Diversity in Conversational Structure Across Languages and
Cultures: 249–281. John Benjamins.
——— (2007). Sequence Organization in Interaction: A Primer in Conversation Analysis I. Cambridge
University Press.
Schegloff, E.A. & H. Sacks (1973). Opening up closings. Semiotica 8(4): 289–327.
Sidnell, J. (2004). There’s risks in everything: Extreme case formulations and accountability in inquiry
testimony. Discourse and Society 15(6): 745–766.
——— (forthcoming). The design and positioning of questions in Inquiry testimony. In S. Ehrlich &
A. Freed (eds.), “Why Do You Ask?”: The Function of Questions in Institutional Discourse. Oxford
University Press.
Terasaki, A.K. (2004) [1976]. Pre-announcement sequences in conversation. In G. Lerner (ed.),
Conversation Analysis: Studies from the First Generation: 171–223. John Benjamins.
Wootton, A. (1997). Interaction and the development of mind. Cambridge University Press.
Transcription systems for spoken discourse
Daniel C. O’Connell & Sabine Kowal

Georgetown University/Technical University Berlin
1. Transcription: Basic terminology
The vast majority of natural language usage is spoken, and most of it is forever lost in
the very next moment. Human memory records very little of it, and very little of it is
audiorecorded, much less of it transcribed. Verba volant: the spoken flies away; but
fortunately, scripta manent: the transcribed remains.
In the context of this article, transcription will be understood to mean any
graphic representation of selective aspects of speaking and of one or more persons’
behavior and setting concomitant with speaking. This representation always presup-
poses a unique performance, it is not meant as a script for a further performance. It
is always sequential in that real-time is necessarily involved. In this respect, transcrip-
tion is to be distinguished from description of some aspect of behavior. For example,
one might transcribe laughter as ‘ha ha ha’, but describe the same laughter as ‘laugh’,
a designation which eliminates the notions of both sequentiality and numerosity
included in ‘ha ha ha’.
‘Transcription’ must be distinguished from another related concept, that of
inscription. Clifford (1990: 51) uses “inscription” to indicate an interruption of ongoing
discourse on the part of an ethnographic fieldworker who wishes to make a note.
In this article, we will reserve the term ‘transcription’ for the generic concept itself,
and we will distinguish between the following terms: ‘transcribing’ for the activity,
‘transcript’ for the result, and ‘notation’ for the tools of that activity, i.e. the set of signs
used to represent selective aspects of spoken discourse graphically. These three aspects
of transcription provide the framework for any scientific research in this domain; to
date, however, notation systems alone, rather than transcribing and use of transcripts,
has been the focus of all the current transcription systems to be considered in what fol-
lows (but see more recent psycholinguistic research on the production, reproduction
and use of transcripts [O’Connell & Kowal 1999; O’Connell & Kowal 2000; Romero,
O’Connell & Kowal 2002; and Spinos, O’Connell & Kowal 2002]).
From time immemorial, people have produced and used transcripts according
to their own needs. One has only to reflect for a moment on the many modern forms
such transcriptions take, e.g. medical records, court reports, therapy protocols, students’
classnotes, and secretarial dictation. Thus, transcription is neither a modern scientific
invention nor the sole preserve of scientists. For the most part, however, such everyday
transcripts have been limited to the verbal component, to the exclusion of prosodic,
paralinguistic, and extralinguistic components. For example, the phone message: Tell
John, ‘Emily called. Your car is ready’, is restricted to verbal components. The intonation
pattern of the speaker, the tempo, and other characteristics of the spoken message are
disregarded for the purposes of this transcribed message.
For many years, the derivation and use of transcripts of spoken discourse in vari-
ous scientific disciplines went on without generating a great deal of critical interest,
although it is true that the linguists themselves have had the longest tradition of interest
in transcription: “The choice of symbols to represent speech sounds has been a matter
of debate for centuries” (Roach 1992: 200).
In the 1970s, however, a new interest arose among a group of sociologists who
concerned themselves with conversation analysis and have come to be known as eth-
nomethodologists. Their effort was to make transcripts, particularly of conversations,
as inclusive and detailed as possible. By contrast, the psycholinguists of this same era
manifested very little interest in the notational aspects of transcription (e.g. Brown
1973: 53), and as for interest in the transcribing itself, many of them even delegated
such operations — clearly considered by them to be an unimportant step in the scientific
elaboration of data — to secretarial assistants (e.g. Maclay & Osgood 1959: 23). Most
of these transcripts for scientific use are derived from audio or audio-video recordings.
But it is also possible to derive transcripts by means of what Shriberg, Kwiatkowski &
Hoffmann (1984: 456) refer to as “live transcriptions”, i.e. the production of a transcript
directly from the ‘live’ performance. Both the method of derivation of a transcript and
the complexity of the spoken discourse (see Pye, Wilcox & Siren 1988: 33ff) affect the
difficulty of transcribing. For example, a ‘live transcription’ from spoken discourse
during a rock concert would pose monumental problems; the signal-to-noise ratio
would not allow adequate accuracy.
Only in the decade of the 1990s did a critical interest in transcription arise,
not limited to any one scientific discipline, but embracing linguists, sociologists,
anthropologists, ethnologists, and psychologists. At least partly responsible for
this critical interest was the potential for computerization of notations and transcrip-
tional procedures, and for the archivization and sharing of data banks. It should be
noted, however, that the focus of most of this critical interest was the notational
systems themselves.
The crucial role of the transcriber as the user of a notational system in the very
process of transcribing, and the role of the reader who is the consumer of the nota-
tional system have still not been adequately studied. This neglect also entails a lack of
concern about reliability and validity in the use of the notational systems, both on the
part of the transcriber and on the part of the transcript reader. Transcription theory
has been concerned with the selection of adequate notation signs, but not with the
psychological processes involved in using them.
242 Daniel C. O’Connell & Sabine Kowal
Apart from these systems, however, there has been some concern regarding
reliability and validity in the transcribing activity itself. Most of this research has con-
cerned children’s and pathological spoken discourse (e.g. Oller & Eilers 1975; Pye,
Wilcox & Siren 1988; Shriberg, Kwiatkowski & Hoffmann 1984); but O’Connell &
Kowal (1994b) as well as Lindsay & O’Connell (1993) and Kowal & O’Connell (1995)
have found similar problems with the transcribing of normal adult spoken discourse.
2. Speaking: The behavior under consideration
Although there is some research available on transcription from the written (e.g.
Keseling, Wrobel & Rau 1987: 349–365; Matsuhashi 1987: 208ff), from the silently read
(e.g. Rayner & Pollatsek 1989: 116), and from the signed (e.g. Boyes Braem 1990: 29ff;
Prillwitz et al. 1989), our coverage in this article is restricted to transcription from
spoken discourse.
2.1 The verbal component

Spoken discourse always involves an audible sequence of sounds. This sequence con-
stitutes the segmental verbal component in transcription. There are traditionally four
ways of encoding this component: standard orthography, literary transcription, eye
dialect, and phonetic transcription.
The most common way of representing the verbal component of spoken discourse
is by means of standard orthography, i.e. the spelling given to lexical items (words) in
a standard dictionary of the language such as Merriam-Webster’s Collegiate Dictionary.
11th ed. (2003). In any case in which a speaker deviates from standard pronunciation,
the transcription will clearly have a loss of information if that deviation cannot be
recorded in standard orthography.
The literary transcription method is a convenient way of taking into account
deviations of pronunciation. Ehlich (1993: 126) identifies this method as involving
“systematic departures from the standard orthographic rendering of an item but in
a manner that is meaningful to someone familiar with the orthographic system as a
whole”. The example he gives is the use of “ye” for “your [sic]” in the transcription of a
Yorkshire dialect.
Eye dialect is a method of transcribing the verbal component of spoken discourse
that is somewhat less systematic than literary transcription, but at the same time takes
into account a greater amount of deviation. It is, in fact, a sort of pseudo-phonetic
transcription, an effort to encode impressionistically all relevant sound categories.
Edwards (1992b: 368) describes eye dialect as “an impressionistic extension of English
spelling”. She offers the following examples: uv for of, cuz for because, and askedche
for asked you. Eye dialect has its origin in fiction, where it serves the purpose of
verisimilitude. Its scientific use has been primarily in ethnomethodological research,
where the emphasis is on everyday conversation. The criticisms of eye dialect include
“uninformativeness, inconsistency, ambiguity, faulty phonetics, and poor readability”
(Edwards 1992b: 368). Gumperz & Berenz (1993: 96ff) add that “eye dialect tends to
trivialize participants’ utterances by conjuring up pejorative stereotypes”.
Phonetic transcription yields a representation of the phonetic categories sequen-
tially realized in a corpus of spoken discourse. For example, what in eye dialect
might appear as cuz may be phonetically encoded as [kh^z] (Edwards 1993b: 20).
The representation may be either ‘broad’ (minimal) or ‘narrow’ (comprehensive) (see
e.g. Kohler 1977: 145). Hence it is ideally suited for producing detailed transcripts.
But except in research in which phonetic considerations are themselves of importance
(including research on previously undocumented languages), this method of transcrib-
ing spoken discourse is not commonly used. The disadvantages of phonetic transcription
are that special training is needed both to transcribe and to read the notations of the
International Phonetic Alphabet (IPA) and that the notations are not all accessible on
word-processor keyboards, much less on ordinary typewriters. The problem of accessibility
has been addressed through a Machine Readable Phonetic Alphabet (MRPA) that
makes use only of the characters of a standard keyboard (Taylor 1990: 184ff). Note,
however, that phonetic transcription is definitely to be recommended “in the study
of children’s early morphosyntactic development” (Johnson 2000: 199) because of
phonological variability.
Kelly & Local (1988: 198) have pointed out still another disadvantage of IPA: “It
is not well suited to work on the systematics of conversation”, not because it gives
too much detail, but quite the contrary, because it does not provide enough phonetic
detail. Accordingly, they recommend “a notational system that is theoretical and, in
principle, indeterminately large” and results in what they refer to as “impressionistic
transcription” — transcriptions reflecting all possible idiosyncracies of individual
utterances. The singular advantage of Kelly & Local’s impressionistic transcription is
that it pinpoints the rich, unique detail of an individual spoken corpus. Although such
very detailed transcription systems have been used throughout the larger part of the
20th century in dialect studies, nothing approaching such detail has been incorporated
in any of the transcription systems to be considered below.
2.2 The prosodic component

The prosodic component includes the features of pitch, duration, and loudness. The
terms ‘emphasis’ and ‘stress’ both subsume one or more of these features indiscriminately.
The features are typically represented by discrete graphic units interposed among the
verbal segments. An example of such a prosodic transcript of falling pitch (\), lengthening
(=), and loud quality ( <F F> ) according to Du Bois et al. (1993: 56, 59, 68) would be
the following:
(1) <F It’s co=ld in here F>.\
But since pitch, duration, and loudness are essentially both continuously variable
and suprasegmental, they cannot adequately be represented by interposed, discrete
graphic units.
The continuous variability over time in the prosodic component calls for a nota-
tion beyond the discrete graphic units that are primarily intended to encode the ver-
bal. For example, the graphemic transcript of a simple utterance such as by no means
cannot represent any of the following continuant and variable dimensions: a pitch
varying from 150 to 170 Hz, duration varying from 0.3 to 0.6 sec, and loudness
varying from 16 to 20 dB. Although all the current transcription systems include nota-
tions for prosody, none of them can cope with these dimensions adequately. Ehlich’s
(1993: 129) “supralinear notations” for increasing ( > > > > ) and decreasing ( < < < < )
tempo come closest to an iconic representation of continuous and suprasegmental
variation in duration:
<<<< >>>>
(2) It’s cold in here.
2.3 The paralinguistic component

The paralinguistic component includes vocal features that occur during speaking but
are not themselves linguistic in nature. Examples would include laughter, breathing,
aspiration, sighing, and crying. Paralinguistic components may occur as separate seg-
ments (e.g. a speaker who stops speaking for a moment in order to laugh), or they may
occur superimposed upon verbal segments (e.g. a speaker who produces an utterance
laughingly, i.e. the audible quality of one or more sequential syllables is altered by the
laughter). In the latter case the same problems arise as with the prosodic component,
namely the inadequacy of static graphic symbols to represent continuous variability
over time. To date, there exist no conventional notations that incorporate a phonology
of paralinguistic phenomena.
2.4 The extralinguistic component

It is difficult in principle to exclude from the extralinguistic component any concurrent
event or circumstance which is not already included in the prosodic and paralinguistic
components and yet influences the spoken discourse in some way. The extralinguistic
component can vary as widely as a nod from an interlocutor to the light from a
nearby traffic signal or the sound of a burglar alarm. Cook (1990: 1) refers to this broad
spectrum of phenomena as the problematic of “transcribing infinity”. A narrow band of
this spectrum — bodily movements — has been addressed by both Birdwhistell (1960)
and Pittenger, Hockett & Danehy (1960). According to Edwards (1992a: 437), “nonverbal
acts” are among the three types of information that must be included in a transcript.
Still, in the current transcription systems reviewed in the following section, the extra-
linguistic component is added as optional description, or as “researcher’s comment”
(Du Bois 1991: 105), or as “nonlexical phenomena, both vocal and nonvocal, that
interrupt the lexical stretch” (Gumperz & Berenz 1993: 121), or finally as “changeable
headers” and “dependent tiers” (MacWhinney 1991: 122 & 125). Ehlich (1993: 138)
includes extralinguistic notation, but adds it as a sort of ad hoc description of “the body
parts involved in a gesture, in conjunction with l and r, for ‘left’ and ‘right,’ respectively”.
For example, a speaker whose hands (HA) remain folded while she or he is speaking is
represented as “-------------- HA folded--------------” (Ehlich 1993: 139; our translation).
3. Current transcription systems
In addition to the well-established transcription system devised by researchers in the

Survey of English Usage in London and Lund (see Svartvik & Quirk 1992), which is still
widely used especially in northern Europe, there are at least five systems in current
use for transcribing spoken discourse, those of John W. Du Bois (1991), Konrad Ehlich
(1993), John Gumperz & Norine Berenz (1993), Gail Jefferson (1988), and Brian
MacWhinney (1991). A comparison of the Jefferson, Du Bois, and MacWhinney sys-
tems with several German transcription systems is presented in Kowal & O’Connell
(2003). They found that, across these American and German systems, there is general
agreement as to what is to be transcribed, but little agreement as to how to do so. It
would appear that there is to date no single widely accepted system for any or all of the
scientific disciplines that require transcription of spoken discourse.
Part of the variety springs from the historical purposes for which the methods
were originally devised. Although Ehlich’s system dates from 1976 (Ehlich & Rehbein
1976), it was originally intended primarily for German. The other four systems
originate in the American English tradition. Of them, Jefferson’s has the most seniority
and enjoys a more extensive use within the ethnomethodological tradition than any
of the others (see O’Connell & Kowal 1994a). Although the transcription system of
Gumperz & Berenz (1993: 92) is very similar to that of Jefferson, the authors distinguish
it by emphasizing in their approach “the rhetorical impact” of “pausing, timing of
exchanges, rhythm, prosody, and other paralinguistic signs” “in affecting the situated
interpretations on which the conduct and outcome of the exchange depends”. They
refer to their approach as a “basically functional perspective”.
Du Bois, on the other hand, is concerned about the computerization of his
transcription notation. Hence, he, more than any of the other authors, spells out a
set of principles for the design of his system. Finally, MacWhinney adds to this
preoccupation with computerization yet another concern, the desirability of archival
data, particularly those data concerned with first-language acquisition. We proceed
now to a thumbnail sketch of each of these transcription systems.
3.1 Du Bois’ discourse transcription (DT)

Among all the authors of transcription systems, John W. Du Bois is unique insofar as
he builds his discourse transcription (DT) system ab initio and top down and defines
it ‘‘as the process of creating a written representation of a speech event so as to make
it accessible to discourse research” (Du Bois et al. 1993: 45). In keeping with Ochs’
(1979) position, it is important for him that “transcription is theory” (Du Bois 1991: 71).
Accordingly, he proceeds to articulate transcription design principles summarized in a
set of five maxims (1991: 78):
1. Category definition: Define good categories.

2. Accessibility: Make the system accessible.
3. Robustness: Make representations robust.
4. Economy: Make representations economical.
5. Adaptability: Make the system adaptable.
On the one hand, these maxims are so stated that no one could disagree with them
or wish to deliberately violate them. On the other hand, they are articulated at
a level of generality that is not particularly helpful in specifying the details of a
transcription system.
Essentially, what Du Bois has done is to select a set of signs for notation and declare
them good, accessible, robust, economical, and adaptable. He provides no empirical
evidence that his notation system is preferable to any other with respect to these qualities,
nor does he establish that these five qualities are the necessary and sufficient ones for
adequate notation.
Du Bois has presented a standard keyboard-friendly set of notation signs for dis-
course transcription and has plausibly grounded them in his design principles. What
is still needed is research on the use of these signs in accordance with Du Bois et al.’s
(1993: 45) own emphasis on the “process of creating a written representation”.
An example of a transcript in DT is the following (Du Bois 1991: 77):
(3) ((LUNCH))
L: .. But ‘they never ‘figured ôut what he had?
R: … He had ^pneumonia.
… [The ^second ‘week] he had ‘pneu@monia,
M: [Êventually].
For the verbal component, Du Bois uses standard orthography. Most of the additional
signs are prosodic (104 ff): Double and triple periods represent short and medium
pauses; the grave and circumflex accents represent secondary and primary accent; the
question mark, the period, and the comma represent transitional continuities: appeal,
final, and continuing. The square brackets indicate speech overlap, the @ symbol indicates
laughter, and LUNCH in double parentheses indicates a researcher’s comment.
3.2 Ehlich’s heuristic interpretative auditory transcription (HIAT)

Konrad Ehlich’s terminology is really an attempt to retain the acronym based on the
German language name of his system: halb-interpretative Arbeitstranskription, which
we have preferred to translate as “semi-interpretative working transcription” (O’Connell &
Kowal 1994a: 84). According to Ehlich & Rehbein (1976: 23; our translation):
This type of transcription is partially interpretative insofar as the transcriber on

the one hand already structures the spoken corpus in terms of both segmentation
and commentary on the basis of his own reflectively applied everyday knowledge of
language, while on the other hand he imposes no further structure on the corpus by
way of interpretation.
The system has been developed and applied in Europe for more than three decades
now, as the original German publications indicate (Ehlich & Rehbein 1976, 1979, 1981).
Essentially, Ehlich’s system is quite like Du Bois’ in that it is based on a listing and
explication of a set of notational signs and rules for their usage. Also similar to Du Bois
is Ehlich’s (1993: 125) reliance on a set of criteria for an adequate and useful transcrip-
tion system: “(a) simplicity and validity, (b) good readability and correctability, and
(c) minimum of transcriber and user training”. However, this is the sole mention of
validity and no further preoccupation with the process of transcribing is in evidence.
Peculiar to Ehlich’s (1993: 125) notation system is what he refers to as its “score
notation” by analogy to musical scores: “Semiotic events arrayed horizontally on a line
follow each other in time, whereas events on the same vertical axis represent simulta-
neous acoustic events” (1993: 129). Score notations have also been suggested by both
Fishman (1978: 402) and Eckert (1993: 60ff) quite independently of Ehlich. In Ehlich’s
score notation, the verbal component is represented in literary transcription. Periods,
commas, and question marks in their functions as “standard orthographic punctuation”
represent “basic intonation” (1993: 128). Hyphens/dashes indicate intonation, and
slashes are used to indicate that the immediately preceding element is interrupted.
Arrows pointing upwards or downwards, linking different parts of an exchange, are
introduced for what Du Bois et al. (1993: 84) have referred to as “presentation of
transcriptions”: they help the reader to follow score notation at ambiguous places.
3.3 The transcription system of Gumperz & Berenz

John Gumperz and Norine Berenz first published their transcription system in the
mid 1990s (Gumperz & Berenz 1993). It has been developed out of a tradition of
ethnographic fieldwork in urban institutions. Peculiar to their approach is the fol-
lowing goal:
To set down on paper all those perceptual cues that past research and ongoing analyses
show participants rely on in their online processing of conversational management
signs. (1993: 92).
This attempt to adopt the perspective of the participants leads Gumperz & Berenz
(1993: 101) to an “interpretive approach to transcription”. For example, unlike Du Bois
et al. and Ehlich, they do not time speech pauses, but instead assess them perceptually.
But like Du Bois et al. and Ehlich, their transcription system is limited to the presentation
of a set of notation signs, without empirical evidence for their usefulness.
The following example is taken from a consultation between a medical student
and her supervising physician (Gumperz & Berenz 1993: 115):
(4) 14. S: ==um so, i think this may be just post traumatic/
15. ==because it’s- {[f] it’s been about three months/}
16. but the pain’s been redu- {[dc] has been reducing/}
17. ==it’s not- definitely not *increased/
In the example, the verbal component is represented in standard orthography. For

purposes of computer search, all text is required to be in lower case letters, hence
the ‘i’ instead of ‘I’ as the only exception to the rule. Double equal signs indicate that
the turn following the signs is latched to the preceding turn. The comma indicates a
“slight rise as in listing intonation”, the slash a “slight final fall indicating temporary
closure” (1993: 121). The ‘f ’ in square brackets indicates more fortis tone for the entire
segment enclosed in curly brackets, and ‘dc’ indicates deceleration of the segment in
curly brackets. The hyphen indicates truncation, and the asterisk indicates normal
prominence or accent.
3.4 The Jeffersonian tradition

The Jeffersonian transcription system has been adapted in many ways and for many
different research purposes by other researchers in the course of time. In sharp
contrast to Du Bois’ system, Gail Jefferson’s system has been developed over the
past three decades in a bottom-up rather than top-down fashion. In fact, no explicit
theoretical orientation or set of foundational principles has ever been provided for
this system. It remains largely a skeletal set of variant notation signs, most often rel-
egated to appendices of research publications. Hence, Jefferson and her colleagues
are innocent of any of the theoretical abstractions characteristic of other authors.
Zimmerman & West (1975: 128) have emphasized this practical approach to tran-
scription in the Jeffersonian tradition:
The transcript techniques and symbols were devised by Gail Jefferson in the course
of research undertaken with Harvey Sacks. Techniques are revised, symbols added
or dropped as they seem useful to the work. There is no guarantee or suggestion that
the symbols or transcripts alone would permit the doing of any unspecified research
tasks; they are properly used as an adjunct to the tape recorded materials.
As the last sentence of the quotation indicates, researchers in the Jeffersonian tra-
dition also emphasize the fact that transcripts are not the primary data but a derivative
database. For example, Button & Lee (1987: 9) remind us:
Nor should it be thought that transcripts are the data of conversation analysis as such.
The data is naturally occurring conversation as a feature of social life, and the use of
tape-recordings and transcripts is a practical strategy for apprehending it, and making
it available for extended analysis.
The reminder is an important one. The danger of allowing derivative data to take on
a life of their own, independent of the spoken discourse from which they originate, is
very real. The search for ideal transcription systems makes this danger all the more
proximate, since the primary emphasis is thereby given to transcripts rather than to
the actual database. The following example is from Jefferson (1988: 167):
(5) ((telephone))
Mr F: Got them sorted out the: tent’s: the tent’s up and everything,
Jessie: Ye:s,
Mr F: → A:nd uh:m (0.6) uh I’ve just given them a mea:.l so: (.)
they’re gonna be uh it’ll keep them warm for a while,
In the example, the verbal component is represented in literary transcription. Short

underscoring indicates light stress; a colon indicates that the immediately preceding sound
is prolonged; a comma indicates a slight rise in intonation; the number in parentheses
indicates silent pause time in tenths of a second; and the period in parentheses indicates
a “tiny ‘gap’ ” (1988: 194). Finally, the arrow marks “an utterance that the author has
referred to in the text” (Button, Drew & Heritage 1986: 110) and therefore constitutes
another instance of Du Bois et al.’s (1993: 84) presentation of transcriptions.
3.5 MacWhinney’s CHAT system for the CHILDES project

The Codes for the Human Analysis of Transcripts (CHAT) system of Brian MacWhinney
(1991; see also MacWhinney & Snow 1985, 1990) has been specifically devised for use
with corpora of children’s language acquisition (especially CHILDES, Child Language
Data Exchange System). It is also explicitly intended for computerization with a view
to archivization of such corpora. Hence, “ease of data entry” (MacWhinney 1991: 5)

becomes an additional criterion for the notation system.
Edwards (1989: pages unnumbered) has criticized three outdated biases essential
to CHAT:
toward written rather than spoken language; toward a reductionist, compositional

perspective on language and discourse; and toward the view of child language as
defective adult language rather than as systematic in its own right.
Thereafter, this controversy continued in a series of articles. Edwards (1992a: 435)

further specified “principles for archive based language research: Maximum Readability
and Minimum Bias; Consistent Encoding for exhaustive computer search; Systematic
Contrastiveness; and Data Comparability in elicitation, transcription and coding”.
MacWhinney & Snow (1992: 459), in turn, asserted that Edwards used “an overly
rigid set of principles”. Finally, Edwards (1993a: 209) replied that “M&S underrate the
impact of corpus inconsistencies on substantive conclusions”. It would appear that the
controversy regarding CHAT was not thereby resolved.
Another criticism of CHAT involves the conflict between two mutually exclusive
goals, both aimed for in the system: optional, variable selectivity on the one hand,
and on the other, comparability of data (see O’Connell 1991: 279). An example from
MacWhinney’s (1991: 108) own corpus is the following:
(6) *FAT: you guys getting ready for school # hmm?
*MAR: except ["] means no ["].
*FAT: except ["] means no ["]?
*ROS: she [!] knows that.
In the example, the verbal component is in standard orthography. The sign # indicates
a silent pause. The punctuation marks are all designated as “Utterance and Tone Unit
Terminators” (1991: 123). Quotation marks in square brackets set off a “metalinguistic
reference” (1991: 48) to the preceding word, and an exclamation mark in square brackets
indicates that the preceding word is stressed.
4. Conclusion: Basic principles for scientific use of transcription
In O’Connell & Kowal (1994a: 102ff) we have articulated a set of basic principles for
transcription (see also the recommendations for transcribers in O’Connell & Kowal
1995: 98ff). The rationale for all of them is fundamentally the same: their usefulness
for the purpose of producing orderly, simple, and clear transcripts that are readable
and meaningful for the reader. These principles, however, are not intended as a vote
for standardization of transcript notation (see O’Connell & Kowal 1999); quite the
contrary, they encourage notation dictated by the purposes of a given research project.
In paraphrase, they are:
1. Parsimony. Only those components of spoken discourse that are to be analyzed

should be transcribed, and only what makes analyses intelligible should be pre-
sented in transcripts for the reader.
2. Conventionality. Graphemes should be used only for the segmental representation
of lexical items, and punctuation marks should be used only for their conventional
purposes as standardized signs to clarify meaning and separate structural units
(see Merriam-Webster’s Collegiate Dictionary. 11th ed. 2003).
3. Lexical integrity. For the sake of readability, the internal integrity of words should
not be interrupted by any supernumerary signs.
4. Objectivity. Subjective perceptions and/or categorizations of the transcriber should
not be recorded as objective measurements.
5. One-to-one correspondence. Signs used in transcription systems should stand for
only one feature of the spoken discourse, and no feature should be represented by
more than one sign.
6. Description. Nonphonological vocal phenomena (e.g. laughter) should be described
rather than transcribed.
7. Separation. Descriptions, explanations, commentaries, and interpretations should
be clearly distinguishable from the transcription of phonological features of spo-
ken discourse.
A great deal of research into the empirical validity of these principles is still needed.
For the present, however, they are to be considered scientific hypotheses to be further
tested. For example, in all the transcription systems reviewed in Section 3, the notation
of speech pauses is considered important. But whether an ethnomethodological per-
ceptual estimation of their occurrence and duration (Couper-Kuhlen 1990: 1) suffices
or whether, in accordance with our objectivity principle, physical measurement is
also needed (Kowal & O’Connell 1993) can only be determined by further research.
The more recent studies of Kowal & O’Connell (2000) and of Spinos, O’Connell &
Kowal (2002) clearly indicate the need for physical measurement. Scholars do not
know much about the psychology of the transcriber or about the psychology of the
transcript reader. However, O’Connell & Kowal (2000) have shown that complex
transcription systems may be too difficult to reproduce from an original source with-
out unintended changes.
All of the principles are violated in one or another of the transcription systems
reviewed in Section 3. Despite the fact that these systems overemphasize standardization,
produce transcripts that are difficult to read, and fail to physically measure prosodic
characteristics of spoken discourse, they remain popular, especially in sociolinguistic
research. It is interesting to note that Macaulay’s (1991: 287) criticisms of transcription

systems have led him quite independently to a set of principles for transcription which
are remarkably identical with the principles presented here.
References
Birdwhistell, R.L. (1960). Kinesics and communication. In E. Carpenter & M. McLuhan (eds.),
Explorations in communication: 54–64. Beacon Press.
Boyes Braem, P. (1990). Einführung in die Gebärdensprache und ihre Forschung. Signum.
Bright, W. (ed.) (1992). International encyclopedia of linguistics, 4 vols. Oxford Univ. Press.
Brown, R. (1973). A first language. Harvard Univ. Press.
Button, G. & J.R. Lee (eds.) (1987). Talk and social organisation. Multilingual Matters.
Button, G., P. Drew & J. Heritage (1986). Transcription notation. In G. Button, P. Drew & J. Heritage
(eds.), Interaction and language use. Human Studies 9 (Special issue): 109–110.
Clifford, J. (1990). Notes on (field)notes. In R. Sanjek (ed.), Fieldnotes: 47–70. Cornell Univ. Press.
Cook, G. (1990). Transcribing infinity. Journal of Pragmatics 14: 1–24.
Couper-Kuhlen, E. (1990). Discovering rhythm in conversational English. Arbeitspapier 13. Univ.
Konstanz.
Du Bois, J.W. (1991). Transcription design principles for spoken discourse research. Pragmatics
1: 71–106.
Du Bois, J.W., S. Schuetze- Coburn, S. Cumming & D. Paolino (1993). Outline of discourse transcription.
In J.A. Edwards & M.D. Lampert (eds.): 45–87.
Eckert, P. (1993). Cooperative competition in adolescent ‘girl talk’. In D. Tannen (ed.), Gender and
conversational interaction: 32–61. Oxford Univ. Press.
Edwards, J.A. (1989). Transcription and the new functionalism. Berkeley Cognitive Science Report 58.
———(1992a). Computer methods in child language research. Journal of Child Language 19: 435–458.
——— (1992b). Transcription of discourse. In W. Bright (ed.), vol. 1: 367–370.
——— (1993a). Perfecting research techniques in an imperfect world. Journal of Child Language
20: 209–216.
——— (1993b). Principles and contrasting systems of discourse transcription. In J.A. Edwards &
M.D. Lampert (eds.): 3–31.
Edwards, J.A. & M.D. Lampert (eds.) (1993). Talking data. Erlbaum.
Ehlich, K. (1993). HIAT: A transcription system for discourse data. In J.A. Edwards & M.D. Lampert
(eds.): 123–148.
Ehlich, K. & J. Rehbein (1976). Halbinterpretative Arbeitstranskriptionen (HIAT). Linguistische Berichte
45: 21–41.
——— (1979). Erweiterte halbinterpretative Arbeitstranskriptionen (HIAT 2): Intonation. Linguistische
Berichte 59: 51–75.
——— (1981). Zur Notierung nonverbaler Kommunikation fűr diskursanalytische Zwecke (Erweiterte
halbinterpretative Arbeitstranskriptionen [HIAT 2]). In P. Winkler (ed.), Methoden der Analyse
von Face-to-Face-Situationen: 302–329. J.B. Metzler.
Fishman, P.M. (1978). Interaction. Social Problems 25: 397–406.
Gumperz, J.J. & N. Berenz (1993). Transcribing conversational exchange. In J.A. Edwards &
M.D. Lampert (eds.): 91–121.
Jefferson, G. (1988). Preliminary notes on a possible metric which provides for a ‘standard maximum’
silence of approximately one second in conversation. In D. Roger & P. Bull (eds.): 166–196.
Johnson, C.E. (2000). What you see is what you get: The importance of transcription for interpreting
children’s morphosyntactic development. In L. Menn & N.B. Ratner (eds.), Methods for studying
language production: 181–204. Erlbaum.
Kelly, J. & J.K. Local (1988). On the use of general phonetic techniques in handling conversational
material. In D. Roger & P. Bull (eds.): 197–212.
Keseling, G., A. Wrobel & C. Rau (1987). Globale und lokale Planung beim Schreiben. Unterrich-
tswissenschaft 4: 349–365.
Kohler, K.J. (1977). Einführung in die Phonetik des Deutschen. Erich Schmidt.
Kowal, S. & D.C. O’Connell (1993). Zeichen für Zeit in Gesprächstranskriptionen. Paper, 7th
International Convention of the Deutsche Gesellschaft für Semiotik, Tübingen.
Kowal, S. & D.C. O’Connell (1995). Notation und Transkription in der Gesprächsforschung. In
D.C. O’Connell, S. Kowal & R. Posner (eds.), Zeichen für Zeit — Zur Notation und Transkription
von Bewegungsabläufen/Signs for Time: Notation and Transcription of Movements. KODIKAS/
KODE 18: 113–138.
Kowal, S. & D.C. O’Connell (2000). Psycholinguistische Aspekte der Transkription: Zur Notation
von Pausen in Gesprächstranskripten. Linguistische Berichte 183: 355–380.
Kowal, S. & D.C. O’Connell (2003). Datenerhebung und Transkription. In G. Rickheit, T. Herrmann &
W. Deutsch (eds.), Psycholinguistik/Psycholinguistics: Ein Internationales Handbuch/An international
handbook: 92–106. de Gruyter.
Lindsay, J.S. & D.C. O’Connell (1993). Transcribers’ changes. Paper, Eastern Psychological Association,
Arlington.
Macaulay, R.K. (1991). ‘Coz it izny spelt when they say it’: Displaying dialect in writing. American
Speech 66: 280–291.
Maclay, H. & C.E. Osgood (1959). Hesitation phenomena in spontaneous English speech. Word
15: 169–182.
MacWhinney, B. (1991). The CHILDES project. Erlbaum.
MacWhinney, B. & C. Snow (1985). The child language data exchange system. Journal of Child
Language 12: 271–296.
——— (1990). The child language data exchange system. Journal of Child Language 17: 457–472.
——— (1992). The wheat and the chaff: Or four confusions regarding CHILDES. Journal of Child
Language 19: 459–471.
Matsuhashi, A. (1987). Revising the plan and altering the text. In A. Matsuhashi (ed.), Writing in real
time: 197–223. Ablex.
Merriam-Webster’s collegiate dictionary, 11th edition (2003). Merriam-Webster.
Ochs, E. (1979). Transcription as theory. In E. Ochs & B.B. Schieffelin (eds.), Developmental pragmatics:
43–72. Academic Press.
O’Connell, D.C. (1991). The spoken flies away; the written stays put. Georgetown Journal of Languages &
Linguistics 2: 274–283.
O’Connell, D.C. & S. Kowal (1994a). Some current transcription systems for spoken discourse. Pragmatics
4: 81–107.
——— (1994b). The transcriber as language user. In G. Bartelt (ed.), The dynamics of language processes:
119–142. Gunter Narr.
——— (1995). Basic principles of transcription. In J. Smith, R. Harre & L. van Langenhove (eds.),
Rethinking methods in psychology: 93–105. Sage.
——— (1999). Transcription and the issue of standardization. Journal of Psycholinguistic Research
28: 103–120.
——— (2000). Are transcripts reproducible? Pragmatics 10: 247–269
Oller, D.K. & R.E. Eilers (1975). Phonetic expectation and transcription validity. Phonetica 31: 288–304.
Pittenger, R.E., C.F. Hockett & J.J. Danehy (1960). The first five minutes. Paul Martineau.
Prillwitz, S., R. Leven, H. Zienert, T. Hanke & J. Henning (1989). HamNoSys Version 2.0 -Hamburger
Notationssystem für Gebärdensprachen. Signum.
Pye, C., K.A. Wilcox & K.A. Siren (1988). Refining transcriptions. Journal of Child Language 15: 17–37.
Rayner, K. & A. Pollatsek (1989). The psychology of reading. Prentice-Hall.
Roach, P. (1992). Phonetic transcription. In W. Bright (ed.), vol. 3: 200–202.
Roger, D. & P. Bull (eds.) (1988). Conversation. Multilingual Matters.
Romero, C., D.C. O’Connell & S. Kowal (2002). Notation systems for transcription: An empirical
investigation. Journal of Psycholinguistic Research 31: 619–631.
Shriberg, L.D., J. Kwiatkowski & K. Hoffmann (1984). A procedure for phonetic transcription by
consensus. Journal of Speech and Hearing Research 27: 456–465.
Spinos, A.-M.R., D.C. O’Connell & S. Kowal (2002). An empirical investigation of pause notation.
Svartvik, J., & R. Quirk (eds.) (1992). A corpus of English conversation. CWK Gleerup.
Taylor, I. (1990). Psycholinguistics. Prentice Hall.
Zimmerman, D.H. & C. West (1975). Sex roles, interruptions and silences in conversation. In
B. Thorne & N. Henley (eds.), Language and sex: 105–129. Newbury House.
Index
A Baynham, M., 199 C

a priori, 68 Beach, W.A., 86, 111, 112, 116, 117 cadences, 29
accompaniment signals, 106 Bean, J.M., 168 Carrell, P.L., 163
accountability, 68 Beattie, G.W., 108 categorization, 7
acknowledgment tokens, 111, 117 Beebe, L.M., 163, 164 Chafe, W., 21, 96, 177, 198
action, 1 Befu, H., 168 change-of-state token, 112
ad hocing practices, 69–70 Bell, A., 22, 111, 141, 166 channel, 134
adjacency pair, 48–49, 178–179, Benford, R., 92 channel conversations, 56–57
220–223 Bennett, M., 104, 105, 109, 110, 119 CHAT system, 249
adjacency pair, sequencing, 2 Benveniste, E., 148 Cherry, R.D., 168
aesthetic semiotic theory, 24 Berenz, N., 243, 245, 248 CHILDES, 249
aizuchi, 115 Bergmann, J., 37, 58 Cicourel, A.V., 5, 6, 69, 75, 76
Allen, G.D., 186 Berk-Seligson, S., 168 Ciliberti, A., 166
Antaki, C., 11, 111 Besnier, N., 196 circular questioning, 60
anthropological linguistics, 97 Billig, M., 8 Clancy, P.M., 104, 107, 114, 115,
aphasia, 42 Bilmes, J., 76, 77 116, 118
approving responses, 109 Birdwhistell, R.L., 106, 245 Clark, H.H., 104, 120, 195, 200
Argyle, M., 109 Bliss, L.S., 165 Clayman, S., 43, 59
Arminen, I., 2, 7, 9, 10, 60 Blommaert, J., 6 Clayman, S.E., 76, 201
Aronsson, K., 93, 94, 168 Blum-Kulka, S., 120, 163, 164, 165, Clift, R., 1, 13, 40, 42, 44,
articulation, 31 166, 167, 168 141, 200
Atkinson, J.M., 5, 7, 42, 43, Blumstein, P., 119 close ordering, 179
59, 233 Boas, F., 150 Coates, L., 104, 105
attention signals, 106 Boddy, J., 142 cognition, 75
audible back-channel Boden, D., 76 Cohen, A., 119, 163
responses, 109–110 Bodman, J., 163, 166 Cohen, D., 190
auditory effects, 174 Bolden, G., 195 collaborative finish, 114
Auer, P., 1, 4, 11, 14, 23, 99, 100, Bolinger, D., 175, 177 collusion, 87
175, 176, 179, 182, 184 Bourdieu, P., 24 commonsense reasoning, 73–75
Austin, J.L., 1, 56, 207, 209 Boxer, D., 166 communicative genres, 37, 62
auto-discussion, 147 Boyes, 242 communicative styles, 11, 20
Braem, P., 242 complaining, 181
B Brazil, D., 177 computer mediated
back channel, 114, 117 broadcast language, 57 communication, 9, 57
back-channel communication, 107 Brown, P., 92, 93, 157, 158, 159, conditional relevance, 48–49
back channel functions, 116 161, 165 Condon, S.L., 112
Bales, R.F., 105, 119 Brown, R., 160, 162, 163, 166 confirming allusions, 236
Banerjee, J., 163 Brunner, L.J., 106 Conger, J.C., 109
Banfield, A., 190, 194 Bryan, A., 168 Conley, J.M., 8
Bangerter, A., 104, 120 Bucholtz, M., 6, 12 constructed dialogue, 198
Bar-Hillel, Y., 70 Buttny, R., 192, 200 consultations and advice
Bardovi-Harlig, K., 163, 168 Button, G., 68, 75, 87, 249 giving, 63
Basso, E., 200 by-play, 141 contact, 134
Bavelas, J.B., 104, 105 bystanders, 86–87 context, 3–8
256 Index
context and conversation D Egbert, M., 42, 58, 134

analysis, 4 Danehy, J.J., 245 Eglin, P., 7
context and interaction, 4 Danet, B., 167 ego identity, 90
context and intersubjectivity, Danielewicz, J., 96 Ehlich, K., 158, 169, 242, 247,
4, 5 data collection, 7 248, 244, 245
contextual styles, 21–22 Davidson, D., 190 Ehlich’s heuristic interpretative
contextualization, 3, 10 Davy, D., 21 auditory transcription, 247
contextualization cues, 3, 23, 174 de Leon, L., 150 Ehrlich, S., 222
conventions of form and means, deictic phenomena, 58 Eilers, R.E., 242
163–164 deixis, 150 Eisenstein, M., 163, 166
conversation, 1, 40, 55 Deng, X., 13, 118 embedding, 88–89
conversation analysis, 1–8, Di Luzio, A., 1, 4, 23, 37, 176 encounters/face
40–43, 56, 76, 207 Dingwall, 8 engagements, 86, 135
conversation styles, 20 direct/indirect reported Enfield, N.J., 8, 150
conversation types, 55 speech, 192 Engeström, Y., 10
conversation types and disapproving responses, 109 Enkvist, N.E., 20, 21, 22
communicative genres, 62 discourse, 56 epistemic dependency
conversational contract, 161–162 discourse functional marker, 113
conversational implicature, 77 linguistics, 97 equivocal participation, 137
conversational maxim, 158–160 discourse identities, 61 Ervin-Tripp, S., 21, 23
convict code, 72 discourse in media, 168 ethnographic description, 4
Cook, G., 244 discourse perspective, 168 ethnographic fieldwork, 5
Cook, H.M., 161 discursive identities, 7 ethnographical techniques, 6
Cook, M., 109, 120 discursive psychology, 12–13 ethnography, 5–6
Cook-Gumperz, J., 11, 15 dispreferred responses, 49 ethnomethodology, 66, 68,
cooperative principle, 158 Dittmann, A.T., 104, 105, 75–76
Coulmas, F., 165, 193, 194 106, 108 etic/emic style, 23
Coulter, J., 68, 76 Dittmar, N., 24 everyday vs. institutional
Coulthard, M., 56, 60, 177 Ditton, J., 94 conversations, 58–59
Couper-Kuhlen, E., 3, 13, Dixon, R.M.W., 190 expression of politeness,
29, 42, 96, 97, 99, 174, Drew, P., 59, 61, 63, 92, 199, 221, 162–164
175, 176, 177, 178, 179, 228, 233, 235, 249 extralinguistic component,
180, 181, 182, 183, 184, DRS, 195 244–245
195, 200, 251 Drummond, K., 104, 111, 112, 114 eye dialect, 242
Coupland, J., 11, 59, 61, Du Bois, J., 100, 196, 224, 245,
165, 168 246, 247, 248, 249 F
Coupland, N., 165, 168 Du Bois, J.W., 100, 196, 245, 246, face concept, 90
courtroom interaction, 7 247, 248, 249 face-threatening act, 158
Crawford, M., 61 Du Bois’ discourse face-to-face interaction, 3
Crittenden, K.S., 72 transcription, 246–247 Fairclough, N., 55, 56
Cromdal, J., 93 DuFon, M.A., 168, 169 Fanshel, D., 56, 59
cross-cultural Duncan, S., 104, 105, 106, 107, Færch, C., 164
communication, 93 108, 109, 113, 114 Fawcett, S.B., 168
cross-cultural study of listener Duranti, A., 4 feedback, 109, 110
response, 114–118 dyadic vs. multi-person Ferrara, K., 168, 195
cross-play, 87, 141 conversations, 57–58 fieldwork, 6
Cruttenden, A., 183, 186 Dzameshie, A.K., 168 Fiksdal, S., 168
Crystal, D., 21, 25, 174, 177, 186 finnish listening behaviours, 117
Cukor-Avila, P., 195 E first in, last out phenomenon, 186
Culpeper, J., 20, 21 Eckert, P., 11, 247 Firth, A., 1, 13, 60, 62, 66, 68,
culture and society, 79 Edelsky, C., 104 70, 72, 74, 76, 78
current transcription systems, Edwards, D., 2, 12, 242, 245, 250 Firth, J.R., 174, 175
245–246 Edwards, J.A., 243 Fisher, L.E., 126, 149
Cutrone, P., 104, 115 Eerdmans, S., 1, 3 Fishman, P., 104, 119
Index 257
Fishman, P.M., 120, 247 130, 131, 132, 133, 134, 135, Hayashi, R., 115, 116, 181
Fiske, D., 104, 105, 106, 107, 136, 137, 139, 140, 141, 147, Hayashi, T., 115, 116
108, 109, 114 148, 149, 150, 151, 152, 153, Heath, C., 6, 7, 8, 9, 226
Fitzgerald, R., 7, 57 157, 158, 160, 161, 191, 192, Held, G., 82, 117, 164, 176
floor, 114 201, 207 Hemphill, L., 24
focused interaction, 86 Goffman, Erving, 79 Herbert, R.K., 163
folk notion, 157–158 Goffman’s attention, Heritage, J., 2, 4, 5, 7, 13, 41, 42,
footing, 88, 139–149 involvement and focused 43, 47, 56, 57, 59, 60, 69, 70,
footing, elaborations and encounters, 133–139 71, 72, 76, 92, 104, 111, 112,
critique of, 149–153 Goffman’s footing, 139–149 181, 182, 187, 217, 220, 222,
footing shift, 146 Goffman’s influence and 223, 224, 225, 227, 234, 249
Ford, C.E., 11, 42, 44, 45, 61, 92, significance, 89–94 Hero, R., 168
97, 99, 100, 107, 158, 168, Goffman’s working Herring, S., 9, 57
178, 187, 202, 224, 240 framework, 85–87 Hester, S., 7, 57, 210
formal talk, 55 Golato, A., 197 heuristic interpretative auditory
Fox, B.A., 97, 99, 178, 181 Goldberg, J.A., 179 transcription, 246–247
frame, 86, 88 Goodwin, C., 2, 3, 4, 9, 10, 12, Hickey, L., 22, 23
frame attunement, 92 28, 42, 58, 60, 96, 98, 111, high considerateness style, 24
Frank, J., 116, 168 112, 128, 130, 134, 135, 152, high-involvement style, 24
Fraser, B., 158, 159, 161 188, 191, 220, 225, 226 Hilbert, 75, 77
Freed, A., 158, 164, 181, 192 Goodwin, M., 8, 24, 141, 148, 152 Hill, R.J., 70, 72
Freese, J., 182 Goodwin, M.H., 43, 61, 153, 201 Hindmarsh, J., 6, 7
French, P., 116, 117, 158, 180, 200 Greatbatch, D., 57, 59 Hinds, J., 115
Friedson, E., 90 Green, G., 160 Hirokawa, K., 115, 116
Fries, C.G., 104, 105 Grice, H.P., 1, 2, 157, 158, 159 Hirschman, L., 119
Fukushima, S., 163 Grimshaw, D., 89, 92 Hockett, C.F., 129, 245
functional gestures, 135 Gu, Y., 158, 160, 164 Hoffmann, K., 241, 242
functions of language, 127 Gülich, E., 37 Holmes, J., 119, 166
Gumperz, J.J., 1, 2, 3, 11, 23, 26, Holt, E, 1, 3, 41, 141, 183,
G 56, 97, 105, 176, 186, 243, 190, 192, 195, 196, 197, 198,
Gallois, C., 168 245, 248 199, 200, 202, 217, 226,
Gardner, R., 42, 57, 104, 111, 112, Günthner, S., 24, 37, 100, 116, 193, 194
113, 120 118, 176, 181, 192, 194 Hopper, P., 97, J
Garfinkel, H., 1, 40, 56, 66, 67, Hopper, R., 57, 111, 112, 114
68, 69, 70, 71, 72, 73, 74, 75, H Horn, R., 82
76, 112, 176, 206, 207, 210, Haakana, M., 62, 194 Housley, W., 7
211, 213 habitus, 24 Houtkoop-Steenstra, H., 57, 111
gaze and body orientation, Hagge, J., 168 Hudson, R., 195
130, 132 Haiman. J., 194 Huiskes, M., 42, 100
gender and ethnicity, 68 Hakulinen, A., 7, 55, 56, 58, 60, Hutchby, I., 1, 5, 7, 9, 13, 28, 40,
gender-differentiated use of 62, 97, 99 42, 44, 46, 48, 50, 57
listener response ., 119–120 Hall, K., 12, 40, 56, 66, 75, 77, 90, Hutchins, E., 10
genres, 62 100, 101, 111, 139, 175, 176, Hymes, D., 1, 23, 37
Gerrig, R.J., 195, 200 194, 195, 200, 209
Gerson, R., 167 Halliday, M.A.K., 56, 90, 176, I
gestalt, 32 194, 195 Ide, S., 158, 159, 161, 169
Gilbert, G.N., 76, 119 Hanke, T., 144 Ikoma, T., 167
Giles, H., 61, 165 Hanks, W.F., 150, 151 imperatives, 150
Gilman, A., 166 Harness, 61 implicature, 2
Glock, N., 198 Hartford, B.S., 163, 168 indexical meanings, 3
go-ahead response, 233 Haspel, K., 12 indexicality, 11–12, 70–71
Goffman, E., 1, 2, 3, 8, 13, 40, 58, Have, P., 6 indirectness and politeness, 167
79, 80, 81, 82, 83, 84, 85, 86, Haviland, J.B., 150 inferential order, 7–8
87, 89, 90, 91, 92, 93, 94, 125, Hayashi, M., 42, 114 informal talk, 55
258 Index
inherent polite speech acts, Jönsson, L., 59 Leet-Pellegrini, H.M., 104,

162–163 119, 120
inscription, 240 K Lehrer, A., 198
institutional discourse, 168 Kalin, M., 62 Lehtonen, J., 116, 117
institutional talk, 42, 58–59 Kallmeyer, W., 24 Leinonen, M., 9
institutional talk, types of, 59–60 Kangasharju, H., 62 Leiter, K., 66, 72
institutionalized multi-party Kasher, A., 160 Lerner, G., 41, 43, 148
conversations, 60 Kasper, G., 1, 2, 13, 120, Lerner, G.H., 42
interaction, 2, 12, 79, 80 157, 158, 160, 162, 164, Levinson, S.C., 1, 3, 49, 55, 58,
interaction analysis, 12 166, 168, 169 93, 92, 141, 148–151, 157, 158,
interactional linguistics, 96 Keane, S.P., 109 159, 160, 161, 162, 163, 165,
interactional linguistics, Kelly, J., 42, 100, 175, 181, 243 167, 229, 231, 232, 236
possibilities and challenges Kendon, A., 8, 92, 93, 104, 105, Lewis, S.H., 104, 109, 110
for, 100–101 106, 135 lexical repetition, 181
interactional organization, 2 Keseling, G., 242 lexico-semantics, 31
interactional sociolinguistics, Kim, K., 42, 161 lexico-syntactic forms, 181
1–3, 56, 97 Kitao, K., 167 lexis, 99
interactional stylistic analysis, Kitayama, S., 159 Li, C., 110
methodology of, 26–27 Kitzinger, C., 42, 228 Liberman, M., 175
interjection, 34 Klewitz, G., 195 Linde, C., 168
international phonetic Klippi, A., 62 Lindsay, J.S., 242
alphabet, 243 Knapp-Potthoff, A., 168 Lindström, A.K., 99, 116
internet relay chat, 9 Kohler, K.J., 243 Lindström, J., 8, 42, 96, 99, 100
interpersonal discourse, 168 Kollock, P., 119 Linell, P., 25, 59, 61, 62, 97, 100
interpersonal meaning, 70 Koshik, I., 41 linguistic communication, 66, 70
interpretive frames, 36 Kostelnick, C., 168 linguistic/non-linguistic cues, 24
intersubjectivity, 70, 71 Kotthoff, H., 24 linguistic pragmatics, 1
interviews, 57 Kowal, S., 3, 13, 240, 242, 244, linguistic signs, 3
intonation, 31, 175, 177 245, 246, 247, 248, 250 listener response, 104
intonational lexicon, 175 Krauss, R.M., 104, 109, 110 listener response, classification
involvement obligations, 136 Kraut, R.E., 104, 109, 110 of, 113–114
irony, 48 Kwarciak, B.J., 165 listener response, structural
Irvine, J., 11, 141, 142 Kwiatkowski, J., 241, 242 properties of, 106–108
Irvine, J.T., 151, 152, 190 listener response tokens, 111
Iwata, Y., 163 L listener responses, roles and
Labov, W., 21, 56, 59, 198 functions of, 108–111
J Lakoff, R.T., 159 literary transcription, 242
Jakobson, R., 21, 126, 127, 128, Lalljee, M., 22, 157, 168, 169 live transcriptions, 241
129, 133, 136, 142, 143 Lambek, M., 142 Local, J., 42, 100, 175, 181
Janney, R.W., 162 Lampi, M., 168 Local, J.K., 243
Jarvis, J., 104, 105, 109, 110 Lange, D., 195 Local, K., 180
Jasperson, R., 99, 181 language and Locastro, V., 115, 116
Jayyusi, L., 7 communication, 79 loci of order, 44
Jefferson, G., 41, 42, 245 language games, 71 Longacre, R., 195
Jeffersonian transcription language ideology, 12 Lord, A.B., 176
system, 248–249 language use, 79 Lörscher, W., 168
Johnson, C.E., 243 Larson, M.L., 198 loudness, 30, 31
Johnson, D.M., 168 Laury, R., 100 Luckmann, T., 37, 61
Johnson, M.D., 168 Lave, J., 10, 11, 157 Lucy, J.A., 190
Johnson, T., 104, 106 Lebra, T.S., 115, 116, 160 Luff, P., 9
Johnstone, B., 12, 168 Lechner, E., 104 lumping approach, 105–111
Jones, K., 168 Lee, J.R., 104, 116, 207, 249 Luong, H.V., 148
Jones, R.H., 10 Leech, G., 157, 159, 162, 165 Lynch, M., 76
Index 259
M Mondada, L., 6, 7, 8, 9, 10 P
Macaulay, R.K., 198, 252 Morand, D.A., 165, 168 paralinguistic component, 244
MacFarlane, S., 179 Morosawa, A., 167 Park, Y-Y., 42
machine readable phonetic Mulkay, M., 76 Parsons, T., 69, 70, 119
alphabet, 243 Müller, F., 175, 179, 182, 184 Partee, B.H., 190
machinery, 2 multi-party conversations, 58 participant observation, 5
Maclay, H., 241 multimodality and participation, 125
MacWhinney, B., 245, 246, mediation, 8–11 participation framework, 58, 88
249, 250 Myers, G., 160, 161, 168, 200 participation roles, 141
MacWhinney’s CHAT participation status, 88
system, 249–250 N passive recipiency, 111
Maeshiba, N., 169 native-non-native pause-bounded phrasal unit, 115
Makri-Tsilipakou, M., 119, 120 conversation, 62 pausing/rhythmic delay, 182
Malinowski, B., 125, 126, 127, natural attitude, 73–75 Pavlidou, T., 24
128, 129, 133 natural language, 71 pay-off, 46
Manzo, J.F., 2 negative and positive face, 160 Pearson, B., 168
Mao, L.R., 159, 162, 164 negative politeness Peirce, C.S., 2, 3, 11, 70
Marche, T.A., 108, 114, 119 strategies, 158, 167 Peräkylä, A., 60
Markus, H.R., 159 negotiations, 62 Perrin, L., 200
Mathis, T., 190, 195 Nesshoever, W., 119, 120 Peterson, C., 108, 114, 119
Matoesian, G., 2 Nevile, M., 9 phatic communion, 125–133
Matoesian, G.M., 201 next turn, understanding Philips, S.U., 190, 196
Matsuhashi, 242 in, 46–48 phonemic clauses, 106
Matsumoto, Y., 159, 161 next turn proof procedure, 224 phonetic cues, 33
Matthews, G.H., 150 non-aligning responses, 230 phonetic transcription, 243
maxims of politeness, 159 non-verbal communication, 133 phonological variables, 22
Mayes, P., 190, 195, 198, 199 noncoercive cultural Pike, K.L., 175
Maynard, D., 2, 42 orientation, 118 Pillet-Shore, D., 111
Maynard, D.W., 5, 6, 7, 182 nonequivalence, 49 Pittenger, R.E., 245
Maynard, S.K. Nordenstam, K., 119, 120 politeness, 157, 158
Mazeland, H., 42, 100 normality, 74 politeness and tact, 162
McCarthy, M., 104, 120 Normann Jörgensen, J., 13 politeness investment,
McDermott, 68 norms and rules, 69–70 variables in, 165–167
McDermott, R., 2 Norris, S., 8, 9, 10 politeness principle, 159
McGregor, W., 190, 195 notation system, 250 politeness studies, 92
McHale, B., 194 Noy, C., 201 Pollatsek, A., 242
McHoul, A., 7 Pollner, M., 68, 73, 77
McHoul, A.W., 60, 76 O polyvocality, 24
McIlvenny, P., 10, 57 O’Barr, W., 8 Pomerantz, A., 2, 7, 13, 41,
Mehan, H., 43 occasion, 86 42, 49, 96, 98, 181, 221,
Mehan, R.P.H., 66, 72, 73, 75 Ochs, E., 37, 42, 97, 246 225, 230
membership categorisations, 210 O’Connell, D.C., 3, 240, 242, positive politeness
Merritt, M., 112 244, 245, 246, 247, 248, 250, strategies, 158
Mesch, J., 57 251, 252 positive reactions, 105
metalinguistic reference, 250 O’driscoll, J., 1, 2, 13, 79, 80, 82, Potter, J., 2, 7, 12, 13, 56
Middleton, D., 10 84, 86, 88, 90, 92, 93, 94 practices of participation, 125–133
Miller, L.C., 104 Ogden, R., 42, 100 pragmastylistics, 23
minimal responses, 110 Oller, D.K., 242 pragmatic approaches in
misplacement marker, 184 Olshtain, E., 163, 166, 167 stylistics, 22
Mizutani, N., 115 Ono, T., 42 preference, 227–229
Moerman, M., 5, 6, 7, 10, 42, Osgood, C.E., 241 preference organization,
43, 68 Östman, J.-O., 99 structural consequences
Molder, H., 13 overlap, 45–46 of, 229–230
260 Index
preferred responses, 49 reported speech in Selting, M., 3, 11, 13, 20, 22, 23,
presentation of storytelling, 198–199 24, 25, 26, 28, 29, 30, 32, 34,
transcriptions, 247 reporting clause, 190, 194–195 36, 42, 96, 97, 99, 100, 176,
presequences, 232 resumptive opener, 114 178, 182
presuppositions, 3 rhythm, 30, 31 Semino, E., 20, 21
Prevignano, C., 1 Rickford, J.R., 11 Semino, E.M., 192, 194, 199
Prillwitz, S., 242 Rintell, E., 166 semiotic structures, 9
production practices, 67–68 ritual and sacred self, 83–85 sequence organization, 2, 3, 44,
production roles, 149 Roach, P., 241 230–235
prosodic component, 243–244 Roberts, C., 92 sequences, 2, 215
prosodic cueing of stance and Roger, D., 119, 120 sequencing of turns, 43
affect, 182 Roger, D.B., 119, 120, 221, 224 sequencing prosody, 178–179
prosodic phenomena, 175–176 Romaine, S., 195 sequential analysis, 235–237
prosodic routines for action, 181 Rosenberger, N.R., 159 sequential order, 7
prosodic signals, 176 Rosenfeld, H.M., 104, 105, 109 Shapley, M., 175
prosody, 31, 100, 174 Ross, S., 168 Sharrock, W.W., 68
prosody and spoken Roth, A.L., 57, 181 Shimamura, K., 167
discourse, 176–177 Rundström, B., 168 Short, M., 192, 194, 197, 199
prosody in talk-in-interaction, Ryle, G., 129, 206, 212 Shriberg, L.D., 241, 242
178–182 side-play, 141
Pye, C., 241, 242 S Sidnell, J., 1, 2, 3, 8, 10
Sacks, H., 1, 2, 5, 7, 12, 13, 26, 28, sign languages, 57
Q 29, 40, 41, 42, 44, 45, 46, 49, Silverstein, M., 1, 3, 4, 12, 176
Quasthoff, U., 37 56, 57, 58, 67, 68, 70, 71, 96, Sinclair, J., 56, 60
Quine, W.V.O., 190 98, 107, 111, 112, 115, 129, 144, Siren, K.A., 241, 242
Quirk, R., 216, 245 148, 178, 179, 182, 184, 206, situation, 85–86
Quis, P., 13 207, 208, 209, 210, 211, 212, situational identities, 7
quotations, 195 213, 220, 224, 227, 229, 233, Slembrouck, S., 199
235, 249 Slugoski, B.R., 166
R Sag, I., 175 Smith, D.E., 68, 107, 115, 201,
Rampton, B., 10, 11, 12, 13 Sajavaara, K., 116, 117 208, 231
Rapley, M., 7, 111 Sandig, B., 20, 22, 23, 24, 25 Snow, C., 249, 250
ratified participants, 86, 87 Sarangi, S., 92 Snow, D., 92
rationality, 73 Saussure, F.D., 133, 135, 143 social action, 10–11, 69
Rau, C., 242 Scheerhorn, D.R., 168 social distance and
Raymond, G., 7, 13, 41, 42 Schegloff, E.A., 235, 236, 237 politeness, 166
Rayner, K., 242 Schenkein, J., 207 social knowledge, 69
reactive expression, 114 Schieffelin, B.B., 97, 150 social marking, 160–161
reactive tokens, 118 Schiffrin, D., 99 social order, 67
recognitions, 109 Schlobinski, P., 24 social organization, 1
redress to face-threat, 160 Schuetze- Coburn, S., 175, 177 social politeness, 162
referential/metalingual, 129 Schulze, R., 168 social power, 165
reflexivity, 71–73 Schumacher, A., 119, 120 social psychology, 55
RehbeinJ., 245, 247 Schütz, A., 70, 73, 74, social situation, 85, 135
repetition, 114 75, 207 social variables, 166
reported/nonreported Schwartz, P., 119 sociolinguistic stylistics, 25
discourse, 191 Scollon, R., 10 sociolinguistics, 1
reported speech, 190, 198 Scollon, S., 158, 168 Sorjonen, M.-L., 42, 57,
reported speech, authenticity Scollon S.W., 10 59, 99
of, 195–198 score notation, 227, 247 Sornig, K. Sornig, K. 163
reported speech, forms Searle, J.A., 1, 56 speech act theory, 2
of, 192–194 self-contextualization, 116 speech circuit, 134, 135
reported speech in self-deception, 81 speech delivery, 43
interaction, 199–201 self-repair, 181 speech functions, 127
Index 261
speech styles, 24 Tao, H., 16, 18, 167 verbal feedback in Maori and
speech styles, intuitive analysis Taylor, I., 243 Pakeha speakers, 118
of, 29 technological mediation, 9 verbal reinforcers, 109
Spinos, A.-M.R., 240, 251 Tedlock, D., 176 verbal response tokens, 57
splitting approach, 111–113 temporal production of talk, 43 Vincent, D., 200
spoken discourse, 3, 242 Terasaki, A., 41, 231 visible back-channel
standard orthography, 242 text linguistic stylistics, 25 responses, 109–110
Steensig, J., 99 textlinguistics, 22 visual and vocal listener
Stivers, T., 2, 8, 13, 41, 43 Thibault, P., 1 responses, 106
story telling, 28 Thompson, G., 194 visuospatial modality, 8
story telling, recipient of, 31–32 Thompson, S.A., 42, vocal-aural modality, 9
Straight, H.S., 163 97, 99, 100, 107, Volosinov, V.N., 24, 29, 141,
Streeck, J., 135 116, 118, 178 191, 193
structural organization of Tomasello, M., 129 voluntaristic theory of
talk, 41–42 Tottie, G., 114, 116, 117, 119 action, 69, 70
Stubbe, M., 116, 118 Tracy, K., 4, 12
style, 11–12, 20–21, 23, 35 Trager, G.L., 107, 175 W
style-constitutive cues, 29–30 transcription, 240 Wachs, I., 24
style in conversational transcription principles, 250 Wagner, J., 62
interaction, 26 transcription system of Wagner, A.R., 141
style-markers, 21 Gumperz & Berenz, 248 Walker, G., 42
styles, functional analysis of, 32, transcription theory, 241 Walker, M.B., 108
33–34 transformational grammar, 22 Ward, G., 82, 104, 115
styles, structural analysis of, transition-relevance places, warranting stylistic
29–31, 33 45, 179 analyses, 32–33, 34–37
styles of conversation, 23, 24 transitional continuity, 177 Watson, R., 1, 7, 8, 13, 206, 208,
styles of speaking, 23, 24, 35 transitional overlap, 180 210, 212
stylistic analysis, 21 Travers, M., 2, 8 Watts, R.J, 93, 158, 162, 169
stylistic profiling, 21 Trevarthan, C., 129 Weber, E.G., 160, 175
stylistic variation, 20 Trimboli, C., 108 Weinbach, L., 163, 166, 167
stylistics, 20 troubles talk, 62 Weinheimer, S., 104
stylistics, criticism of, 25–26 Tsukahara, W., 104, 115 Weizman, E., 167
subordinate communication, 87 turn construction, 178 Wells, B., 179
Suchman, L., 9, 10, 76 turn-constructional units, Wertz, M., 68
Sudnow, D., 68, 76 45, 178 Wetherell, M., 8, 12, 56
Svartvik, J., 245 turn system, 107 wh-question, 181
Swezey, L.W., 104, 109, 110 turn-taking, 2, 41, White, S., 115, 116
symmetry and asymmetry 44–45, 107 Widdicombe, S., 11
conversations, 60–62 Turner, R., 67 Wieder, D.L., 72, 75, 76
syntax, 31, 99 type constructs, 73 Wieland, M., 116, 117
typology of styles, 22 Wiemann, J.M., 108
T Wierzbicka, A., 164, 165, 190,
tact, 162 194, 198
U
Tainio, L., 100 Wilcox, K.A., 241, 242
Uhmann, S., 182
Takagi,, T., 42 Wilkinson, S., 42
unfocused gathering, 86
Takahashi, S., 169 Williams, R., 91, 92,
Urban, G., 142, 144
Takahashi, T., 163, 164, 166 192, 200
utterances, 71, 83
talk-in-interaction, 5, 9, 10, 40, Wilson, T.P., 7, 75, 108
43–44 Wittgenstein, L., 70, 71
Tanaka, H., 42, 99 V Wolfson, N., 166
Tannen, D., 11, 24, 61, Valentine, T., 119 Wood, H., 66, 72, 73, 75,
92, 96, 104, 105, 190, Varenne, H., 2 115, 224
192, 195, 198 Vehviläinen, S., 59, 62 Woodbury, A.C., 176
Tannen, E., 24 verbal component, 242–243 Wooffitt, R., 28, 196, 201
262 Index
Wootton, A.J., 7, 35, 63, 235 Y Z

workplace communication, 168 Ye, L., 164, 242, 249 Zammuner, V.L., 168
Wortham, S., 2, 201 Yngve, V.H., 104, 106, 107 zero-quotatives, 195
written discourse, 168 Yoshinaga, N., 169 Zimmerman, D.H., 7, 42,
Wrobel, A., 242 Young, R.F., 27, 42, 104, 116, 142 43, 69, 73, 75, 108,
Wu, R-J. R., 42 Yule, G., 175, 177, 190, 195 112, 249
In the series Handbook of Pragmatics Highlights the following titles have been published thus far or
are scheduled for publication:
10 Sbisà, Marina, Jan-Ola Östman and Jef Verschueren (eds.): Philosophical Perspectives. ca. 250 pp.
Forthcoming
9 Östman, Jan-Ola and Jef Verschueren (eds.): Pragmatics in Practice. ca. 250 pp. Forthcoming
8 Ledin, Per, Jan-Ola Östman and Jef Verschueren (eds.): Discursive Pragmatics. ca. 250 pp. Forthcoming
7 Jaspers, Jürgen, Jan-Ola Östman and Jef Verschueren (eds.): Society and Language Use. ca. 250 pp.
Forthcoming
6 Fried, Mirjam, Jan-Ola Östman and Jef Verschueren (eds.): Variation and Change. Pragmatic perspectives.
ca. 250 pp. Forthcoming
5 Brisard, Frank, Jan-Ola Östman and Jef Verschueren (eds.): Grammar, Meaning and Pragmatics. 2009.
xiii, 308 pp.
4 D’hondt, Sigurd, Jan-Ola Östman and Jef Verschueren (eds.): The Pragmatics of Interaction. 2009.
xiii, 262 pp.
3 Sandra, Dominiek, Jan-Ola Östman and Jef Verschueren (eds.): Cognition and Pragmatics. ca. 250 pp.
Expected December 2009
2 Senft, Gunter, Jan-Ola Östman and Jef Verschueren (eds.): Culture and Language Use. 2009. xiii, 280 pp.
1 Verschueren, Jef and Jan-Ola Östman (eds.): Key Notions for Pragmatics. 2009. xiii, 253 pp.

The Pragmatics of Interaction

Uploaded by

Copyright:

Available Formats

The Pragmatics of Interaction

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Pragmatics of Interaction

Uploaded by

Copyright:

Available Formats

The Pragmatics of Interaction

Handbook of Pragmatics Highlights (HoPH)

John Benjamins Publishing Company

Cover design: Françoise Berserik

Library of Congress Cataloging-in-Publication Data

© 2009 – John Benjamins B.V.

Preface to the series xi

The pragmatics of interaction: A survey 1

3. Data, transcription and analysis 43

Listener response 104

2.6 Politeness and politic behavior 162

Reported speech 190

Harvey Sacks 206

Transcription systems for spoken discourse 240

–– Key notions for pragmatics

Jan-Ola Östman (University of Helsinki) &

Jan-Ola Östman (University of Helsinki) &

1. Layers of interactional organization

2. Context, ethnography and categorization

3. Multimodality and mediation

understandings of talk-in-interaction that focus exclusively on the vocal-aural modality

4. Style and indexicality

5. This volume: Interaction as a topic

While systematically investigating the organizational patterning of activity constitutes

to earlier). In CA, the interactional patterning of participants’ displays of cognitive

Auer 2007)) or overviews of specific distributions of stylistic practices in particular

Antaki, C. & S. Widdicombe (eds) (1998). Identities in Talk. Sage.

Heritage, J. (1990/1). Intention, meaning and strategy: Observations on constraints on interaction

Moerman, M. (1988). Talking culture: Ethnography and conversation analysis. University of

1. Definition, delimitation, basic concepts

‘Style’ refers to the meaningful deployment of language variation in written and

features. The goal of stylistic analysis is to decompose/deconstruct holistic styles and

2. A few landmark reference works

In my sample analysis in Section 4, I will illustrate styles of speaking. A good

4. Sample data and methodology of an interactional stylistic analysis

The following analysis focusses on styles of speaking in conversational interaction. On

1. Identification of styles as relevant resources in interaction. Start by (intuitively) identi-

all linguistic levels, i.e. phonetic-phonological, prosodic, morphological, syntactic,

1. Transcription symbols

Segmental and sequential phenomena:

Pitch movements on and after accented syllables:

4.2 Structural analysis: Decomposition/deconstruction

1 Mia: =↑ ‘FRIEDrich sachte übrigens zu mir dann. (.) ähm-

Friedrich’s proposal is presented as follows:

4.3 Functional analysis

14 (.) m: dat dat ↓ ‘EINzige was mir dazu ↓ ‘EINgefalln is?

17 /↑‘ANGST daß ich meine semi / /1.0/

4.5 Structural analysis: Decomposition/deconstruction

4.6 Functional analysis

(4–2) Reconstruction of suggested interpretive frames, presupposed stylistic expec-

5. Perspectives for future research

The construction and alter(n)ation of communicative style(s) in everyday conversa-

Labov, W. (1972). Sociolinguistic Patterns. University of Pennsylvania Press.

Rebecca Clift, Paul Drew & Ian Hutchby

2. Origins and overview

Conversation analysis emerged in the pioneering researches of Harvey Sacks into

This structural investigation of talk has imported a rigorously empirical methodology

3. Data, transcription and analysis

Conversation analysis focuses its attention on recordings of actual spates of talk-in-

Preface to the series xi

The pragmatics of interaction: A survey 1

Listener response 104

Reported speech 190

Harvey Sacks 206

Transcription systems for spoken discourse 240