Talk and Social Organisation 9781800418226

This book contains a collection of original studies in conversation analysis arranged and presented both to introduce th

164 27 22MB

English Pages 296 [344] Year 1987

Report DMCA / Copyright


Table of contents :
Transcript Symbols
1 Prologue: Talking Organisation
2 On the Preferences for Agreement and Contiguity in Sequences in Conversation
3 Recycled Turn Beginnings: A Precise Repair Mechanism in Conversation's Turn-taking Organisation
4 On Exposed and Embedded Correction in Conversation
5 Moving out of Closings
6 Notes on Laughter in the Pursuit of Intimacy
7 Unilateral Departure
8 'You want to find out if anybody really does care'
9 Descriptions in Legal Settings
10 Work Flow in a Paediatric Clinic
11 Interdisciplinary Considerations in the Analysis of Pro-terms
12 Epilogue: The Definition of Alternatives: Some Sources of Confusion in Interdisciplinary Discussion
Recommend Papers

Talk and Social Organisation

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Talk and Social Organisation

INTERCOMMUNICATION SERIES Series Editors Howard Giles, Department of Psychology, University of Bristol, BristolBS8 1HH, U.K. Cheris Kramarae, Department of Speech Communication, University of Illinois, Urbana, IL 61801, U.S.A. Editorial Advisory Board William M. O'Barr, Department of Anthropology, Duke University, Durham, NC 27706, U.S.A. Suzanne Romaine, Merton College, Oxford University, Oxford, U.K. Rod Watson, Department of Sociology, University of Manchester, Manchester, U.K. Forthcoming Books in the Series Communication and Crosscultural Adaptation YOUNG Y.KIM Conversation: An Interdisciplinary Approach DEREK ROGER and PETER BULL (eds.) Communication and Simulation DAVID CROOKALL and DANNY SAUNDERS (eds.) Please contact us for the latest information on all our book and journal publications: Multilingual Matters Ltd, Bank House, 8a Hill Road, Clevedon, Avon BS21 7HH, England.

INTERCOMMUNICATION 1 Series Editors: Howard Giles & Cheris Kramarae


Graham Button and John R. E. Lee

MULTILINGUAL MATTERS LTD Clevedon • Philadelphia

Library of Congress Cataloging in Publication Data Talk and social organisation. (Intercommunication series; v. 1) Bibliography: p. Includes indexes. 1. Conversation. 2. Social interaction. I. Button, Graham. II. Lee, John (John R.). P95.45.T35 1987 401'.9 87-7649 ISBN 0-905028-75-9 ISBN 0-905028-74-0 (soft)

III. Series.

British Library Cataloguing in Publication Data Talk and social organisation.— (Intercommunication series; 1) 1. Interpersonal communication 2. Conversation I. Button, Graham II. Lee, John 302.2'242 HM132

III. Series

ISBN 0-905028-75-9 ISBN 0-905028-74-0 Pbk

Multilingual Matters Ltd. Bank House, 8a Hill Road, Clevedon, Avon BS21 7HH, England. Copyright ©1987 G. Button, J. R. E. Lee, H. Sacks, E. A. Schegloff, G. Jefferson, C. Goodwin, A. Pomerantz, W. Sharrock, R. J. Anderson and D. R. Watson. All rights reserved. No part of this work may be reproduced in any form or by any means without permission in writing from the publisher. Cover design by Jussi Nurmi. Typeset by Wayside Books, Clevedon, Avon. Printed and bound in Great Britain by Short Run Press Ltd, Exeter EX2 7LW.

Dedicated to some very important relatives and friends who have all helped us in different ways Cyril, Diane, Elaine, Madge and Wes




Transcript Symbols


1 Prologue: Talking Organisation J. R. E.Lee


2 On the Preferences for Agreement and Contiguity in Sequences in Conversation Harvey Sacks


3 Recycled Turn Beginnings: A Precise Repair Mechanism in Conversation's Turn-taking Organisation Emanuel A. Schegloff


4 O n Exposed and E m b e d d e d Correction in Conversation Gail Jefferson


5 Moving out of Closings Graham Button


6 Notes on Laughter in the Pursuit of Intimacy Gail Jefferson, Harvey Sacks and Emanuel Schegloff


7 Unilateral Departure Charles Goodwin


8 'You want to find out if anybody really does care' Harvey Sacks


9 Descriptions in Legal Settings A nita Pomerantz


10 Work Flow in a Paediatric Clinic Wes Sharrock and Bob A nderson


11 Interdisciplinary Considerations in the Analysis of Pro-terms D. R. Watson


12 Epilogue: The Definition of Alternatives: Some Sources of Confusion in Interdisciplinary Discussion Wes Sharrock and Bob Anderson









As the chapters in this volume are sandwiched between a prologue and an epilogue, both of which have a distinctly theoretical flavour, it is not our intention to use this preface to add much to the arguments which they provide. We would, instead, like to take the opportunity of inviting our readers to enjoy a very special collection of sociological chapters. Each of the contributions to the volume are studies in conversation analysis (hereafter C.A.) and address their concerns to the explication of different features of the social organisation of language. In this respect, the volume is not new as it takes its place in a series of such collections: Sudnow (1972), Schenkein (1978), Atkinson & Drew (1979), Psathas (1979), Goodwin (1981), Atkinson & Heritage (1984), Button, Drew & Heritage (1986), and Psathas (forthcoming). However, these collections taken together with the numerous journal articles which conversation analysts have produced do represent something special in the history of sociology. They represent the development of a sustained, detailed and rigorous style of sociological analysis which takes the social organisation of talk as its topic of enquiry. Now we are, of course, aware that we have just made a controversial statement, and that controversy has continuously attended the development of C.A. and of ethnomethodology of which C.A. is a part. Sociologists are puzzled as to our status. We claim to be sociologists and to be studying social organisation and the ordering of social activities, both of which concepts have long been familiar to, and part of, the standard sociological repertoire. And yet many sociologists have difficulty recognising our work as being sociology, the methods are unfamiliar, the topics unusual and the transcripts of natural conversation which we use as data, besides being 'trivial', are neither collected nor anlysed in the manner approved by the standard textbooks of methodology. Do we represent a rival school proposing alter1



native theories to solve traditional problems? We say we do not. Alternatively, may we be considered as some form of linguist, safe but essentially irrelevant to matters of sociological concern? In recent years the controversy has extended to linguists who are also unsure of the status of this neW discipline. The work is concerned with language and certainly looks like linguistics and its proponents, so it is thought, might therefore properly be asked: what answers does C. A. provide to traditional linguistic problems or what might the sparkling new methodology contribute to the solution of those problems? More recently, the question has arisen as to whether the new discipline can be incorporated into, or can involve itself in, multilingual studies, perhaps to the construction of a new over-arching approach to the study of linguistic matters. We will not speak in any depth to these controversies but in view of their currency, we have added both a prologue and an epilogue which discuss the matter of the origins, objectives and the wider implications of C.A. for those who study related concerns. However, so as not to be denied the chance to enter the fray in some capacity, we wish to make it clear to those who have missed the point, that C.A. is concerned to study the social organisation of natural language-in-use. The concern is with the study of the activities or doings of conversationalists and with the means whereby they achieve order and organisation where they may be observed to have so achieved it. Sociological criticism which suggests that C.A. has omitted or neglected traditional, and therefore presumably sovereign concerns has missed the whole point. It has failed to see how ethnomethodology adopts a quite revolutionary stance within sociology by making topics of the methods which people use to do social life in the everyday world. In this respect, talk is no more, and no less a topic for investigation than any other socially constructed or socially constituted phenomenon. The generic interest is with: '. . . observable practices which make up the incarnate production of ordinary social facts' whatever these may be 'for example, order of service in a queue, sequential order in conversation (or) the order of skilfully embodied improvised conduct.' (Lynch et al., 1983, our emphasis.) This illustrates what it means to subscribe to an ethnomethodologists understanding of the nature of social doings. It is premised on the view demonstrated by Garfinkel many times (see Garfinkel, 1967) that ordered interaction is best understood as the ongoing accomplishment of those who produce it in accordance with methods, roles and structures which are employed and improvised in the very course of achieving it.



C . A . has developed in the light of such an understanding and is the result therefore of a concern to fashion a new style, or styles, of sociology which rejects the positivistic and functionalist assumptions which have marred the development of a rigorous discipline. Instead, it seeks to find ways not only of revealing society and social organisation as an achievement but of observing that order and achievement from inside. That is, instead of using data as a resource to test theories as to the nature of social organisation, it examines the social organisation of materials in an attempt to describe and understand that nature. It seeks, therefore, to describe social organisation of natural interactions asking what their order looks like and how it is put together by those who, in fact, produce it. The research then differs from most contemporary sociology and linguistics in that some theory constructed from without has not been brought to, or used to select, the materials in such a way that they become auxiliaries or means of testing hypotheses or of supporting so-called 'scientific' models. The concern is not to try and fulfill some theoretical version of what constitutes a proper 'science' of society but is instead to determine the orderly ways in which conversationalists shape their talk with one another in mind thereby constituting it as a socially organised phenomenon. This research ideal is so different from that of most current forms of linguistics that it is not surprising that some linguists have failed to understand it. That they have misunderstood it is shown by certain criticisms and by the way in which some have professed a desire to adopt C.A.'s procedures to test their hypotheses. However, as both Lee, and Sharrock & Anderson suggest (this volume, Chapter 1 and Chapter 12), the establishment of a relationship between C . A . and linguists is highly problematic. Linguists do not concern themselves with social activities, social organisation or natural data. They have their own distinctive problems which are not shared by conversation analysts. A prospective marriage is made even more remote by the fact that much of contemporary linguistics employs the same 'constructive' type of methodology that characterises traditional sociology and from which ethnomethodologists broke away in the first place. Of course, we are in no way suggesting that all of the varied concerns and interests of different schools of linguistics are illegitimate or misconceived. Nor are we suggesting that C.A. can have no interest for scholars of other disciplines including those who study language. Apart from anything else, an understanding of C.A. and its style of analysis will provide ways of re-thinking and re-describing the phenomena of language. The emphasis which C . A . gives to language-in-use as social conduct will provide an illus-



trative lesson as to the way in which some of the problems which language raises, or seems to raise, look altogether different when viewed from this perspective. Neither are we naive enough to suggest that other students of talk should abandon their interests and become conversation analysts. To make such a suggestion would be to make the mistake of some of our critics that is of not recognising that there are a variety of ways of conceptualising and therefore presumably of researching into talk. We do not take the view that our set of interests are intrinsically superior to the interests of others. However, this does not mean that we can unproblematically sit down with others and participate in a multilingual or over-arching approach to the study of talk. The results which C.A. produces are the consequences of employing a distinctive methodology and are bound up with a distinctive analytic organisation. If a multi-disciplinary approach is to be desired then it can only be achieved following a thorough examination and subsequent re-organisation of the analytic objectives and methodological styles of those who currently study language. Whilst we do not expect to have laid all the controversies to rest, the editors do hope that students from other disciplines might be spurred to examine C.A. and might perhaps as a consequence find ways of re-formulating their problems so as to embrace the study of social activities. We hope that in this way and in others C.A. might be exemplary. In recent years conversation analysts have developed interests in other than conversational forms of speech exchange. (Hence our title, Talk and Social Organisation, as opposed to Conversation and Social Organisation.) In doing so these sociologists are attempting to lay down new foundations for the analysis of issues more familiar to sociologists. In turning to, for example, talk, in law courts, in classrooms, in doctors surgeries, between children etc. conversation analysts have indicated that there is a high potential for coming to understand those organisational settings through the analysis of the talk which goes on there. Though such work is still exploratory it stands as an example of the possible gains to be had from looking at familiar objects from the viewpoint of a different perspective. With all the foregoing in mind we hope the reader will gain as much pleasure from the articles in this volume as we have gained. We suggest that there are a range of different avenues of interest and we venture to suggest what some of these might be. Firstly, and foremostly, this volume is a collection of work in conversation analysis which contributes to our knowledge of the way in which talk is socially organised.



Secondly, it demonstrates the way in which conversation analysts investigate the social order of talk focussing upon the way in which talk is a sequentially organised production. The first seven chapters particularly take up both of these themes in one way or another. Lee, in Chapter 1, in introducing the volume and C.A. to the neophyte, is concerned to reveal the ways in which traditional sociological problems, particularly problems of social order and organisation are posed and attended to by conversation analysts. It is aimed at those in sociology and in other disciplines whose acquaintanceship with C.A. has been restricted to isolated studies. It places C.A. within the context of the soicological troubles f r o m which it emerged. In Chapter 2, Sacks displays the nature of sequential analysis and reveals some of the ways in which talk between two persons is sequentially constructed. In particular he focusses upon the ways in which the placement of utterances with respect to prior utterances displays a preference for agreement and contiguity. This preference for contiguity is shown to account at least partially for the fact that talk flows on a continuous basis f r o m one person to another. In his treatment of the phenomenon Sacks displays the principles of sequential analysis, the analytic dynamo of C.A. The turn-taking system for conversation is elaborated upon by Schegloff in Chapter 3. He shows how the repetition of components used to begin a turn can be seen as organised by, and understood by reference to the turn-taking system. What might otherwise be seen as redundant (and perhaps meaningless features of talk) can now be seen as a strategic interactional move. Recycling turn beginnings is part of the way by which interactants claim a turn to talk, hold it throughout overlap and get that turn understood as an utterance in relation to prior talk. In Chapter 4, Jefferson shows how 'correction' may be interactionally accomplished over the course of a sequence of turns at talk and how different organisations of correction are produced in different sequential formats. These different formats relate to the doing of correction as either embedded within, or as exposed on the surface of the ongoing interaction and it will be shown that the different formats have different consequences for the ongoing course of the interaction. The next two chapters, Button in Chapter 5, and Jefferson, Schegloff & Sacks in Chapter 6 display the complex and detailed interactional concerns that the fine-grained sequential analysis of talk reveals. Button considers the range of activities that can occur once a closing section for conversation has been initiated, and he considers how these may be used to implicate conversationalists in the conversation continuation or to continue in a sequence



that moves on towards closing whilst at the same time 'getting other things done'. Jefferson et al. explicitly attend to what other analysts of talk treat as an epiphenomenon and often fail to transcribe or simply transcribe as 'laughter'. Instead Jefferson et al. reveals how laughter can be a sequentially organised activity and they thus show the value of attending to sound particles with the greatest attentions to detail. It is shown that laughter is a delicately and collaboratively organised activity through which interactional states such as intimacy can be achieved. In Chapter 7, Goodwin continues to demonstrate how the production of sequentially embedded utterances are related to their interactional consequences. H e considers however, the way in which an utterance seemingly unrelated to the ongoing talk, seemingly sequentially disembodied, in fact achieves the interactional task of 'breaking away' from the group. A third avenue of interest that this volume has for us has already been indicated. It is concerned with the way the included essays indicate possible convergences and divergences between conversation analysis and other disciplines. In a range of ways it exhibits the problems that would have to be surmounted by those who hold store by a multi-disciplinary treatment of talk. Chapter 2 by Sacks and Chapter 3 by Schegloff are based upon presentations made to the 1973 Linguistic Institute at the University of Michigan, and in these presentations they draw preliminary comparisons between the units of study in linguistics and C.A. This comparison is developed in depth by Watson in Chapter 11. He contrasts the ways in which the linguistic theory proposed by Pike and the interactional approach of C . A . considers the use of pronouns. Watson is concerned to display the weakness of the theoretical and 'idealised' versions of language which Pike's theory represents. By contrast he displays the gains to be had from the analysis of pro-terms as interactional objects. In the epilogue Sharrock & Anderson consider the gains and losses that would be consequent upon, and the problems that would have to be overcome in order to produce a multidisciplinary approach. They carry out an in-depth analysis and comparison of two approaches: Conversation analysis and Discourse analysis and make it clear that whilst on the surface the notion of a multi-disciplined approach appears to be a progressive step it cannot be achieved by legislation. It cannot be achieved by welding together essentially different analytic objectives which require different analytic schemes of relevance. A fourth avenue of interest is with the traditional question of social order and organisation. As Lee explains in Chapter 1 C.A. has its origins in the desire to provide a rigorous ethnography of social life. Whilst the order and organisation of conversation itself represents one such form of ordered



activity there are many others and they too might be explored through the analysis of talk. Thus in Chapter 8 Sacks analyses the 'explanation' offered in a conversation: 'You want to find out if anybody really does care.' In this, an adaption of his famous Irvine lectures Sacks, explores the organisation of culture that might be invoked in a suicide bid that is directed to reveal 'if anybody does care'. In a similar tradition Pomerantz in Chapter 9 examines the culture of a court room pointing to the fact that important distinctions between fact and opinion are both signified and evaluated on the basis of the types of description and characterisation which may be used to register events and activities. On the one hand there are official or definitive descriptions that are used to produce official or established versions of things. On the other hand there are contextual describing mechanisms that are relational and judgemental. Though in courts and in ordinary life we may think we have a preference for 'the facts' both descriptive styles have an interactional part to play and have their own logics in use. Anderson & Sharrock, in Chapter 10 show how sequential analysis can be used to explore traditional sociological topics. Thus they show how the talk between patient and doctor organises and schedules the interaction itself. The work of getting the consultation underway is accomplished by reference to a made-visible orientation to the serial organisation of a set of work tasks. Thus the interactants are shown to serially organise and sequence a set of tasks on the basis of the sequential organisation of the talk which constitutes the medical encounter. Over the course of compiling this volume we have incurred a number of debts that we wish to acknowledge. Our authors have endured a journey that has had many twists and turns in it, however they have responded with alacrity and professionalism to all the demands that have been made of them. We thank them most sincerely. In particular we wish to thank Emmanuel A. Schegloff for providing and allowing us to publish Chapter 2 by the late Harvey Sacks which he with the assistance of Jennifer Mandlebaum compiled from a tape-recording of a presentation made to the Linguistics Institute, Michigan 1973. Gail Jefferson kindly produced and compiled a version of two Sacks lectures from 'Fall 64' and 'Spring 65' and this is presented as Chapter 8. Mike Grover of Multilingual Matters and our series editor Howard Giles have shown a nice combination of encouragement and patience. Rod Watson too provided support and help at a vital time. Wes Sharrock solved a large problem for us and an infinite variety of small ones. Karen Futtock provided an invaluable service with the word processor which Mai Lowe made available. We are most grateful also to Mrs Ashton



and the other members of the Faculty of Economics typing pool at the University and wish to pay tribute to their wonderful skills in reproducing such complex data. Elaine Bowdler and Diane Lee are better at spelling and have a finer sense of English than do either of the editors. Last but not least to Jim Schenkein for the work that he did at an earlier stage in the volume's history.

Graham Button John R . E . Lee

Transcript symbols

Before marking out the use of the symbols used in conversation analytic transcripts a few remarks on their history may be in order. It must be borne in mind that conversation analysis addresses naturally occurring activity. The predominant focus has been upon conversation as a form of social activity quite simply because it is the most prolific form of speech exchange, hence the term conversation analysis, though much attention has also been paid to other forms of speech exchange. The interest in naturally occurring activity stresses the analytic interest in conversation as a socially organised phenomena, as social conduct. The term naturally occurring activity is stressed in order to mark out that conversation analysis is interested in what persons actually do, as opposed to what analysts might have thought they did, or as opposed to interpretations or reconstructions of what they did. In order to achieve this goal, conversation analysis makes extensive use of tape-recordings and transcripts made of them in its analyses. This is again quite simple because, if the interest is in what actually happens, taperecording is a more reliable device than reconstructed notes or memory. It allows details that may otherwise escape attention to be apprehended, and it allows analysts to return and return again, to the same phenomena. Thus there is no mystique about the use of tapes and transcripts, as has sometimes been implied. 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, and in the case of transcripts, they are a convenient way of presenting the material that was analysed in research reports. The transcript symbols that are used have evolved to display the sorts of interests that conversation analysts have in talk. They mark out and display 9



features of talk that conversationalists orient to in their production of that talk. Thus, as a matter of principal they do not display pauses of one hundredth of a second as transcripts used by other types of analyst sometimes do, nor do they, as a matter of principle present elaborate representations of intonation, again as some other forms of transcript have. If it could be found that such phenomena were oriented to as such in the production of the talk, then conversation analytic transcripts would reflect them (see Lee, Chapter 1). Thus, the symbols which are explained below are those which capture features of talk that persons have been found to display an orientation to. Gail Jefferson has been largely responsible for developing the transcript notation used by conversation analysis; it is an evolving strategic system, and the symbols presented below are ones that are pertinent to the chapters contained in this volume, and are not definitive of all symbols that have, or will, figure in conversation analytic transcripts.

Sequencing symbols Simultaneous utterances Where utterances begin simultaneously, double left-handed brackets are used to link them together. [[

Ida: Jenny:

r,Bye foh no:w, Look fohw'd tih see in you Buh bye,

Overlapping utterances When one speaker's utterance begins whilst another speaker's utterance is still in progress, a single left-hand bracket joins the two utterances together at the place in the utterance in progress where the next utterance begins. [

Jenny: Mahrks'n s rpencihs shelves w'fcleu : uh. Vera: Well they wouldn't stay for a meal.

Alternatively the place at which overlap begins may be marked by two oblique lines. Thus the next utterance in the transcript begins at the marked place in the current utterance. //

Vic: Mike:

En//then, 's not a lie,



Should an utterance contain more than one such symbol then, in serial order, next utterances begin at the places marked in the current utterance. Vic: Frank: Joe:

Nah kih— no kidding. She//once poisoned//me you see, Yeah. Yuh know?

The end of overlapping talk is marked by a single right-hand bracket. ]

Jenny: Vera:

So e-theh wz^not a thing. I : didn 'knowj wuh : ah wz jst so y'll be busy t'morrow

Continuous utterances When one utterance runs on from a prior utterance without an interval, equal signs link the two utterances together. =

Jenny: Ann:

Eh - wahtuh plea^se, = = Areyou-yousu:uh,

Should two speakers begin their utterances at the same time and run those utterances on from the prior utterance without an interval, an equal sign precedes the double left-hand bracket that marks the simultaneous start. = [[ Vic: Joe: Mike:

We'll get intuh dat Joe (we've gotta get the) = rrI getta kick outa dat story. No no no no.

Should the overlap of two utterances coincide with the end of both utterances, and the next utterance runs on without interval, an equals sign follows the single right-hand bracket that marks the end of overlap. ] = Tom: Bob: Ann:

I used to smoke ^a lot j _ I see = So did I

When, within the same utterance, a speaker runs together parts of the utterance without interval, an equals sign is placed at the place in the utterance when the run-together occurs. Lil:

Ah jus keep ferge:tting=an' I'm sure glad

Equals signs are also used to link together parts of single and continuous utterances which due to the format of transcript presentation has been broken up.



Jenny: t ahn they - they (•) look funny you see when ^yiv been u s e = Ann: we: 11 Jenny: = tih the (t) I _like the. . . .

Intervals of no speech Timed T i m e d intervals within an utterance or between utterances are indicated by placing numbers in parentheses to mark the seconds and tenths of seconds of the interval. Thus within an utterance: (1.1)


0hnoitsalovelyoneitsu-itsaclosedi:n(0.3)eh:m (0.2) onit.

a n d b e t w e e n utterances: Mathew: Jenny:

Okay see yuh. h "(1.2) Wu! (•) won'tchu be bahck home e'tall now.

W h e n there is an interval that is less than a tenth of a second but one which is still discernable, a dot is placed in parenthesis and is described as a 'micro-pause'. Again this indicates a micro-pause within an utterance and b e t w e e n utterances. Thus within an utterance: (•)


they're all very well. Mathew (•)'v. of all things. . . .

and between utterances: Ida:

B't no shoe:s on. hh I wzonly in my: ti:ghts:. (•)



Untimed intervals W h e n an interval within or between utterances is untimed, a description is placed in double parentheses. ((pause))

Rob: Bill:

So I'll see you ((pause)) Tuesday ((pause)) No Wednesday



Characteristics of speech delivery Intonation A full stop indicates a stopping fall in intonation which can occur both within the course of an utterance, or, though not necessarily, at its completion. Thus the full stop does not mark the grammatical end of a sentence. Vera: Vera:

Okay. Right. Bye bye luv,= Okay. She's alright is she.

A comma indicates a continuing intonation. This may, but may not, occur between clauses of sentences and again is not a grammatical form. ,


Uh no she's, gone (up) t'town, h

A question mark indicates a rising intonation; it does not mark a question, though a question may involve rising intonation. ?

Susan: Hello? Nowhere five foh sevin ni:ne?

An exclamation mark indicates an animated tone, but again does not necessarily coincide with a grammatical exclamation. !



Upward and downward arrows indicate marked rising and falling shifts in intonation. U

Jenny: Altright

Sound When a sound is extended or prolonged, colons are used to mark relative durations, the more colons the longer the extension. : Ida: Ye:h, that's eh: :m (0.3) t'morrow night ah thiink, A dash indicates a cutoff sound like a gutteral stop. -


eh u - t h e : u the people- (.) yih know the: wh jewler people.

Degree signs indicate that the part of the utterance they enclose is discernably spoken more quietly than the rest of the utterance, or should they enclose an entire utterance it is quieter than the surrounding utterances.


14 00

Ida: ' E w'so : close t' °Math'w° wa:sn't'e= Ida: = ' E wa.s very :: Jenny: °well this's it chu see :: Ida: An'nowhe'stgone.

Talk that is spoken louder than the surrounding talk is marked by capital letters. CAPITALS


. hhh H A V ' N EV'N washed my f(h)ace yet!

Discernable aspiration is marked by hhh. The longer the aspiration the more hs. Should aspiration occur within a component in an utterance the h(s) are placed in parentheses. h

Lila: Wilbur: Jenny:


Well i t s - u h - i t s uh its , t e n - i t s ten minutes of eleven, l ehhhhhhhhhhhhhhh heh heh heh Loearly die(hh)d(h) I n(h)ea(h)rly(h)y d(h)ied

Discernable inhalation is marked by an h preceded by a full stop. Sometimes this is placed at the top of the h. The longer the inhalation the more hs. Again, should inhalation occur within a component in an utterance the h(s) are placed in parentheses. •hor.h (h)


Ye:ah w'l okay, okay ( )w'l s o : I - 1 don't (•)


hhhhhhhhhh hhOkay, .hhhhhhhhhhhhhhhhhh So, what he con('hhh)flagration means for us:: isn't determinable as yet.

Sounds that have a hardened quality are marked by a full stop placed under the letter. t


Ooo well wasn't that lucky thatchu went when you did then.

Sections of an utterance that are delivered at a quicker pace to the surrounding talk are enclosed between 'less than' and 'greater than' signs. >
Jean: Milly:

. . . and then they said something about Krushchev has leukemia so I thought oh it's all a big put on. Breshnev. Breshnev has leukemia. So I didn't know what to think.

(5) [GJ:FN] Pat: —» Jo: Pat: Jo: Pat: Jo:

. . . the Black Muslims are certainly more provocative than the Black Muslims ever were. The Black Panthers. The Black Panthers. What'd I You said the Black Muslims twice. Did I really? Yes you di: d, but that's alright I forgive you,

(6) [GTS: 11:2:ST] Ken:

—> Roger: Roger: Ken: Ken: Al: Ken: ( ): Ken:

And they told me how I could stick a th-uh:: Thunderbird motor? (0.5) in my Jeep? And I bought a fifty five rThunderbird motor. Not motor, engine. You speak of ,electric motor and a gasoline engine. l Okay Engine. jOkayInternal combus:tion. Alright, So,lookit, mhhhh I moved this thing in the Jeep, yesterday . . .

(7) [Frankel:HB:14-16:ST] EHie: Marny:

I said jeez I said thank god we didn't take Marney's boojks 'hhh and all your reh-your stuff too, Oh::: yhhheah. Oh I didn't even think of that, yeah,' hhh



Ellie: I'm gla:d that your(a) procrastinator as far as Marny: gYehh-hh-hheh! hhh Wait a minute, —» Marny: We're not p u h - we're relaxed about it = Marny: „ We're not procrastinators ehh-huh. Ellie: ~ That's it relaxed I didn't get it righ(h)t). (8) [Goodwin:DP:32-33:ST] Jan: Beth: Jan: -> Ron: Jan: Ron: Jan: Beth: Ron: Ron: Beth:

I guess they paid two-twenty thousand for the house and two thousand for the ki: 1. Mm::, Technically, (It's a) kil:n. Kil:n, I don't know how to say it, You always say kil. I don't know I thought that's right f t. l Ye f ah. 1 It's likeIs that right? You say kil? Kil:n, I don't know I've heard both . . .

These fragments vary in their particulars and have a range of features which warrant attention, but for the purposes of this report I will make three gross observations. (1) Whatever has been going on prior to the correcting is discontinued. W h e r e prior utterances have been occupied with various ongoing matters, utterances are now occupied by the doing of correcting. That is, 'correcting' is now the interactional business of these interchanges. (2) In the course of the business of correcting we can find such attendant activities as, e.g. 'instructing' (fragment 6, 'You speak of electric motor and a gasoline engine'), 'complaining' (fragment 8, 'You always say kil'), 'admitting' (fragment 7, 'I didn't get it right'), 'forgiving' (fragment 5, 'That's alright, I forgive you'), and in other materials, 'accusing', 'apologising', 'ridiculing', etc. That is, the business of correcting can be a matter of, not merely putting things to rights, as in, say fragment 4, but of specifically addressing lapses in competence and/or conduct. Call this class of activities 'accountings'. (3) Whatever else may be going on in the correcting talk, we find an identical series in each fragment: 1. 2. 3.

A speaker produces some object (X). A subsequent speaker produces an alternative (Y). Prior speaker produces the alternative (Y).



(In fragment 3, we find 'Wednesday', 'Tomorrow', 'Tomorrow'; in fragment 4, 'Krushchev', 'Breshnev', 'Breshnev'; in fragment 5, 'Black Muslims', 'Black Panthers', 'Black Panthers'; in fragment 6, 'motor', 'engine', 'engine'; in fragment 7, 'procrastinator', 'relaxed', 'relaxed'; and in fragm e n t s , 'kil', 'kiln', 'kiln'.) That is, whatever else is going on in the course of the correcting, and however it is done, it is this (X, Y, Y) series which constitutes 'correction of one speaker by another'; specifically, someone who at one point produced one sort of object now produces the alternative to it proffered by a co-participant. We can also find materials in which a correction is offered and rejected, in contrast to the above, in which the correction is accepted by prior speaker. When a correction is offered and rejected, we find, not the (X, Y, Y) series, but (X, Y , X ) . (9) [DN:I:2:18(r):ST] Meg: Loren: Loren:

It came from England Loren, (0.4) Ah-ahx_:,


A h - a h it's stampted on the botto: m. (3.5) I:ndia:.


Madid in India.

Meg: Loren: Meg:

°Ma:de. in India not madid, 0 .hh °in India, 0


(you said) m: madid.


You:: shouldn't say madid.


Madid in En . gland.


D a - u h i-in In:dia. .hh


°°England .hh 00


They have s: :ome, .hh they have them sort of like it in Indee (.) in: England, .hh but it's no:t exactly like it,

(4.0) (0.9)

(1.2) (1.0) (0T5) (0.4) (0.4)" (1.4)


TALK AND SOCIAL ORGANISATION (10) [GTS:IV:45] Roger: Al: —» —» Roger: AI: —» Roger: Al: Roger: Al: Roger: Al: Roger: Al: Roger: Al:

Did you have oil in it? Yeah. I-I mean I changed the oil, put new oil filters, r completely redid the oil system, had to put new gaskets on the oil pan to strop-stop the leak, and then I put- and thenThat was a gas leak. It was an oil leak buddy. It's a gas leak. It's an oil rleak. On the number one jug. It's an oil leak! Outta where, the pan? Yeah. Oh you put a new gasket on it stopped leaking Uh huh, °°(No you .didn't have to) 00 Then I-then I had full oil, and I was going up to Lafayette, at about thirty or forty miles an hour . . .

That is, instead of, say, 'madid', 'made', 'made', in fragment 9 we get 'madid', 'made', 'madid'. And in fragment 10 we find, not 'oil', 'gas', 'gas', but 'oil', 'gas', 'oil'. These, then, are the three gross features: (1) Correcting as the current interactional business, with discontinuation of the ongoing activity, with utterances now occupied by the doing of correcting, (2) the possibility for attendant activities — Accountings — which address lapses in competence and/or conduct, and (3) the presence of the (X, Y, Y) series which constitutes 'correction of one speaker by another' (and its alternative, the (X, Y, X) series via which a proffered correction is rejected). These features are found in materials collected as instances of repair, in which one speaker is correcting another. Now let me turn to an altogether different sort of collection, having to do with procedures for consecutive reference to same objects. One common procedure for consecutive reference to a same object is the following. An object is named, and subsequent reference is done with proterms. (11) [Schenkein:II:61 :ST] Kitty:

Don't forget to watch Born Free tonight

ON EXPOSED AND EMBEDDED CORRECTION Gloria: Kitty: Gloria: Miles: Kitty: Miles: Gloria: Kitty: Miles: Gloria: Kitty: Kitty: Stan: Kitty: Gloria: Gloria: Kitty: Stan: Gloria:


Oh yeah I know Oh 11:loved it. Yeah. It was- And we have never seen it. Haven't you = = rrOh you haven't? Oh it's great. No::. Oh:: I wish you kids ( come to our house) and watch it in colour. It's a great picture. ( ) in colour? O h : : : yeah. _ r rYeah I really liked it. We saw Midnight Cowboy yesterday- or suh- Friday. Oh? Did you s - You saw that, = _ „It's really good. No I haven't seen it Jo saw it and she said she f - depressed her terribly Oh it's rterribly depressing Oh it's depressing.

This is a particularly 'pure' instance. There are two series in which a movie is named (first, Born Free and then Midnight Cowboy), and each subsequent reference, over a long string of references, across multiple speakers, is done with a proterm (Series 1: 'Don't forget to watch Born Free . . . ' , ' O h I loved i t ' , ' A n d we have never seen i t ' , ' O h it's great','. . .watch it in colour', 'It's a great picture', 'Yeah I really liked it'. Series 2: 'We saw Midnight Cowboy. . .', 'You saw that', 'It's really good', 'No I haven't seen it, Jo saw i t . . .', 'Oh it's terribly depressing', 'Oh it's depressing'). Another procedure is the following. An initially introduced term is repeated.

(12) [GTS:II; 73] Roger: Jim: Roger: Jim: Al: Ken:

This is an abnormal session see = Yeah, We're not together without the broad. Yeah. See we gotta have the broad here cause she- she unites us. [ heh


92 (13) [NB:IV:13:14:ST] Emma: Lottie: Lottie:

—> —> Lottie: Emma:

So I mean we did a lot of walking arou:nd you know, there's ,not much to getting a turkey dinner, Yuh. (0-4) rrNo. Damn turkey'd put it in the hhh thing in the oven and basting it and everything, and, IT WAS PRETTY GOO:d. Y?ry very good turkey. Where'd you get the turkey. Up at the Ferndale Market,

That, is, simply enough, in fragment 12, following the utterance 'We're not together without the broad', we find, not, for example, 'See we gotta have her here . . .', but 'See we gotta have the broad here . . .', a repeat of the prior term. Likewise, in fragment 13, following 'Very very good turkey', we find, not 'Where'd you get it?', but 'Where'd you get the turkey?', a repeat. And a third sort of procedure is the following. An item is introduced and a next speaker uses an alternate from the same syntactic class, or, using the same item, an alternate pronunciation. (14) [GTS:II:60:ST] —> Ken: Roger: —>

Well- if you're gonna race, the police have said this to us. That makes it even better. The challenge of running from the cops!

(15) [GJ:FN] ((hardware store: customer trying to match a pipefitting)) —» Customer: Mm, the wales are wider apart than that. —> Salesman: Okay, let me see if I can find one with wider threads. (16) [TC:II(a): 14:21 :ST] Griff: —» J.R.: —*

Well I - uh I didn't know anyone: that knew anything about kilns except you:. Whhhhuhhhuh .hh Actually most've my experience's been in gas kils though really

In fragment 14 we find 'police' followed by 'cops'. In fragment 15, the customer uses 'wales' to refer to the ridges on the pipe, the salesman uses



' t h r e a d s ' , referring to the complex of ridge and space. And in fragment 16, we get 'kilns' followed by 'kils'. It appears, then, that something more than sheer consecutive reference is occurring; i.e. that different, and characterisable work may be done when a p r o t e r m is used, or a repeat, or an alternate. For the purposes of this r e p o r t , I will be focussing on the latter procedure. And it turns out that as the latter three fragments unfold, something interesting happens. When prior speaker talks again, he now produces, not the item (word or pronunciation) he, himself initially used, but that which was used by his co-participant. (14a) [GTS:II:60:ST] Ken: Roger: Ken:

Well- if you're gonna race, the police have said this to us. That makes it even better. The challenge of running from the cops! The cops say if you wanna race, uh go out at four or five in the morning on the freeway . . .

(15a) [GJ:FN] Customer: Mm, the wales are wider apart than that. Salesman: Okay, let me see if I can find one with wider threads ((Looks through stock)) Salesman: How's this. Customer: Nope, the threads are even wider than that. (16a) [TC:II(a):14:21 :ST] Griff: J.R.: Griff:

Well I - uh I didn't know anyone: that knew anything about kilns except you:. Whhhhuhhhuh .hh Actually most've my experience's been in gas kils though really I know it! That's what I keep telling myself. Why the hell do you fool with an electric ki(h)l when you can get a gajs kil.

T h a t is, when prior speaker talks again, we find that co-participant's alternative has been consequential in a specific way: it has become a replacement for the initial item, has been adopted by prior speaker. Over and above sheer consecutive reference, then, it appears that when a next speaker produces, not a proterm or a repeat, but an alternative item, correction may b e underway.



Specifically, we find one of the grossly observable features of the f r a g m e n t s collected as instances of the type of repair in which one speaker corrects another. That feature is the (X, Y, Y) series which constitutes 'correction of one speaker by another', via which prior speaker accepts next s p e a k e r ' s proffered correction. In fragment 14a we find 'police', 'cops', 'cops', in fragment 15 a 'wales', 'threads', 'threads', and in fragment 16 a 'kilns', 'kils', 'kils'. In fragments collected as instances of 'consecutive reference', we can also find instances of the (X, Y, X) series. For example, in the following f r a g m e n t , while the two participants are asserting agreement with each o t h e r , a candidate correction of a peculiar phrasing is offered and rejected in t h e series ('eve', 'night', 'eve'). (17) [SBL:3:6:4] Adele: Milly: Adele: Milly: Milly: —* - » Adele: —» Milly: Milly: Adele:

D o you think they might go tomorrow, O h I don't think so, Oh dear. They're (, ) No I don't think until after uh (0.2) after New Years now cause uh, New Y - New Years is tomorrow eve -isn't it. It's tomorrow night uh huh, Yeah tomorrow eve, (1.5) No. .hhh ,Well(I'm just) going to go to the neighbours. . .

A n d in the following fragment, two women are appraising a third. While they are both asserting agreement with each other, and both using positive assessment terms, next speaker's alternate can be seen as a downg r a d e , prior speaker's as an upgrade, next speaker then preserving the d o w n g r a d e term ('pretty', 'beautiful', 'pretty'). (18) [NB:PwT:2:ST] Emma: Penny: Emma: Penny: Emma: —> Penny:

Oh honey that was a lovely luncheon I shoulda ca:: lied you soo: ner but 11: loved it. It was just deli :ghtfu: 1. Well I was gla,d you (came). And your friends are so da:rlijng, = = O h : : it was ThatPa:tisn'tsheado:f:ll? Yeh isn't she pretty, (•)



Emma: Penny: Emma:

Oh: she's a beautiful girl. = = Yeh i think she's a pretty girl,l, And that Henderson::


She SCA: RES me .with eighjt kids . . .

Thus, for both sets of fragments (the 'repair' collection, fragments 3-8, and the 'consecutive reference' collection, in fragments 14a-16a) we find prior speakers changing their terminology following a next speaker's alternative, i.e. we find the (X, Y, Y) series, or we find prior speakers preserving their terminology in the face of a next speaker's alternate (as in 'repair' fragments 9 and 10, and 'consecutive reference' fragments 17 and 18); i.e. we find the (X, Y, X ) series. For this one of the three grossly observable features of 'one speaker correcting another', the two collections are similar. It is on the other two features that the two collections part company, and because of that difference that, although similar in one respect, they look so unlike each other. (1) While the initial collection has the feature that whatever has been going on prior to the offering of a correction is discontinued, in the latter collection the talk in progress continues. The (X, Y, Y) or (X, Y, X) series is embedded into that ongoing talk. That is, the utterances are not occupied by the doing of correcting, but by whatever talk is in progress. Thus, while in the initial collection, correcting has the status of 'the interactional business', in the latter collection, correction occurs, but is not what is being done, interactionally. What we have, then is embedded correction as a by-the-way occurrence in some ongoing course of talk. (2) While the initial collection has the feature that in the course of correcting, as an interactional business, we find attendant activities, 'accountings', which specifically address lapses in competence and/or conduct, embedded correction has no place for such attendant activities. Simply enough, to direct an accusation, apology, etc., to an item in some ongoing talk would necessarily discontinue that ongoing talk, would have utterances now occupied with talk directed to the trouble; i.e. would have that as, now, the interactional business. Thus, the talk which constitutes 'embedded correction' does not permit of 'accountings'. It might be said then, that 'embedded correction' is a means by which correction, and only correction, occurs in contrast to activities recognisable as 'correctings', which permit not only of correction, but of 'accountings'. The distinction between 'embedded correction' and 'correcting' seems to cut across other sorts of classifications in the organisation of 'repair'. For



example, it holds not only for those instances in which one participant corrects another, but also for the types of materials we glanced at to start off with; i.e. those in which a speaker corrects himself, and those in which a next speaker locates the trouble but leaves it to prior speaker to do the correction. To get a sense of that across-the-board working of 'embedded correction; versus 'correcting', we can notice that a variety of 'accountings' show up in the following fragments, which are all instances of 'correcting', but which can be otherwise classified. For example, we find explanations of the error, ridicule, and apology in the following instances of 'self-correction'. (19) [GTS:V:29] Roger:

The mother isn't holding- the father isn't- ah Freudian Slip heh heh mother hah hhehh hhhehh

(20) [Agorio:II:223] Diaz: —* Diaz: Carla: Diaz:

she's the product of a: n incestuous:::: "incestuous0 I'm sorry hih .hhhem::: a u h (2.0) °mm:::° °A,dul (trous)° relationship with another woman,

And, for example, we find apologies and forgiveness in materials in which a next speaker locates the trouble and prior speaker does the correction. (21) [SPC: 10(a):4:ST] Desk:

Mr. O.: Desk: Mr. O.: Desk: Mr. O.: Desk:

. . . but it's a t - on three o'clock and she might j ust be free or between interviews. (1.0) w-What time is it now sir? Three isn't it? (0.7) (We:ll?) I thought it was earlier than tha:t, "(0.3) It's two o'clock I'm sorry. Yeah. I got the hour wrong. But it's just two. hfhh Okay let me call her and then you call her in about fifteen or twenty minutes.



(22) [GTS: 11:2:54] Ken: Al: —> Ken: -> Al:

He likes that waiter over there, Wait-er? Waitress, sorry, That's better,

And in the following fragment, a speaker asks for correction, the correction is supplied by another, and prior speaker then initiates ridicule of the error. (23) [Actors Group:42] Charles: —* —» Lee: —»Charles: —> Lee: -* Martha: Charles:

I mean a - even actors are okay if you pick the ones who are not all hung up in uh you know — wanting — to — well I think you gyrate — not gyrate — is gyrate the right word? Gravitate. Gravitate! heh Gyrate hehh hehh heh ((zig-zag whistling)) gravitate towards the people who are — you know all involved in in ideas and concepts . . .

That is, once 'correcting' has become the business, there is room for 'accounting', regardless of how or by whom the correction is done. 3 With these sorts of observations in hand, we might now re-examine our characterisation of these collections. So far we have asymmetrical formulations: 'correcting' and 'embedded correction'. The former names an activity, the latter names a procedure or device. But we can notice that the former is, equally, a device. Where the latter can be observed to be incorporating the correction into ongoing talk, the former can be observed to be isolating the correction, making of it an interactional business in its own right; i.e. exposing it. And, once exposed, the doing of correction can be invested with a set of activities which would otherwise be unavailable; i.e. the 'accountings'. We have, then, two distinctive forms: 'exposed correction' and 'embedded' correction' as devices for repairing a problematic item in ongoing talk. Having located these distinctive forms we can make a further observation on the materials at hand: Whether he accepts or rejects the correction,


98 prior


does so in the form


by his co-participant.

If next

speaker produces an utterance which discontinues the ongoing talk and is occupied by the doing of correction, then prior speaker does likewise. If next speaker produces an utterance which is continuous with ongoing talk, which happens to have an alternate item, then prior speaker produces continuous talk which happens to repeat the alternate (or which happens to repeat his own initial item). This feature is so recurrent and unproblematic that it appears to be a given, an automatic sequence. It is not. The following fragment demonstrates that while next speaker can initiate correction in one form, this does not guarantee that prior speaker will follow suit. In this case, a next speaker initiates correction, and does so in 'exposed' form. While prior speaker accepts the replacement item, he does so in the 'embedded' form. (24) [GTS: IV: 23-24(r): ST] Jim:

Jim: Roger: Ken: Jim: Ken: Jim: —* Roger: Jim: Ken: Jim: Ken: Jim: —> Ken: Jim: Ken: Ken: Jim:

Like yesterday there was a track meet at Central.Reej_se was there.Isn't that a reform schooj.1, (0.4) Reejse? (•) Ye:s. „Yeah. Buncha niggers and everything? Yeah. (0.3) He went right down on that fie J d and he was just sitting there talking like a nigger and all the guys (mean) all these niggers are a: 11 ^up there in- , You mean Ne gro: don't you. (•) Well and rthey're all-ih-u, _ l AndJi:g, = They f ' r e - they're A:LL up in the sta.ndsyou know all hunh (•) Th: ese guys j ust (are) completely radical T think I think Negroes are cool gu:ys you knoj_w, Some of them yeah. s:Some of them yea.:h but when they get in groups. The others would just as soon = slar:shyourfa:ceasseeyou., forget it you know? hehh



We get the (X, Y, Y) series, 'nigger', 'Negro', 'Negro'. We also get an exposed-form initiation, 'You mean Negro, don't you' and an embeddedform acceptance; i.e. 'I think Negroes are cool guys' is occupied, not with the doing of 'correcting' (in this case the accepting of a proffered correction), but with some ongoing, 'on-topic' talk. The shift into embedded form proposes to exclude the possibility of 'accountings' vis-à-vis the use of the word 'nigger'. It should be noted, however, that while prior speaker rejects the form which provides for 'accountings', he does address the general implications of that form; i.e. that 'nigger' has been seen as a perjorative reference, for which he is being held accountable. And, although I haven't yet captured an instance, people report occasions on which a next speaker initiates embedded correction and prior speaker, while accepting the correction, rejects the form, shifting into exposed form. 4 For example, they say things like 'That's the word I was looking for!' and go on to explain that they knew the word they were using was wrong, but the right word has slipped their mind. That is, they shift into the device which will provide a place for an accounting, and in the accounting they display that an apparent lapse in competence was, say, nothing of the sort, but a matter of problem-solving ingenuity. The possibility for rejecting a form initiated by next speaker leads us to see that the recurrent, unproblematic feature, that the interchanges run off in one form or another, but run off in the form initiated by next speaker, is a collaboratively achieved feature of the phenomenon. The interchanges do not simply run off that way; it is not automatic. Rather, not only is it to be worked out, here and now, step by step, whether a correction will be accepted or rejected (or perhaps reconciled, as is a potential in fragment 8, with a decision that both versions are correct), but it is a matter of collaborative, step by step construction that a correction will be an interactional business in its own right, with attendant activities addressing issues of competence and/or conduct, or that correction will occur in such a way as to provide no room for an accounting.

Notes to Chapter 4 1. T h i s paper first appeared in a limited circulation of Studium Linguistik: 14 pp. 5 8 - 6 8 , Konigsteinlts, 1983.

2. In his unpublished lectures, Harvey Sacks now and again addresses such issues as, 'asking for a name without outright doing so' (Fall, 1964, Tape 1), 'refusing without observably refusing' (Fall, 1964, Tape 5, side 1), producing information in such a way that is is capturable but not respondable-to, in contrast to announcing it



(Winter, 1970, Lecture 2), indicating, versus asserting, a position (Spring, 1971, April 5, pages 7-9), indicating, versus asserting, a relevance (Spring, 1971, April 23, pages 4-5), musing aloud to elicit but not officially request response (Spring, 1971, May 21, pages 5-6), indicating that one knows what is being talked of without naming it (Fall, 1971, Lecture 3, page 4), showing the 'normalness' of an event without asserting that it was normal (Fall, 1971, Lecture 6, pages 5-11). 3. Such a consideration predicts that we will find cases of embedded self-correction whatever that might look like. One form it might take is that, instead of, say, 'X, I mean Y', we get an 'X' which is not correct or acceptable, incorporated into a list, in which context it becomes a type-instance, not an intended specific; i.e. 'X, or Y, or Z'. In fragment 24, although it is an instance of multi-party work, we find the potential for such a procedure. A speaker says 'nigger' and another initiates correction with 'You mean Negro don't you'. A third participant then says 'And Jig'; i.e. proposes that what is going on is a three-party listing of synonyms (nigger, Negro and Jig), rather than a 'correcting'. 4. This phenomenon can provide a further glimpse into the workings of embedded correction. What we have so far is that embedded correction can be a way of doing correction-and-only-correction; of keeping such issues as incompetence and/or impropriety off the conversational surface. In effect, the embedded form provides the opportunity to correct with discretion. That someone rejects the opportunity to correct with discretion can be accounted for in the following way. Initiation of the embedded form is doing something interactionally. Its very discretion constitutes an implicit account of that which it is being discrete about; i.e. to initiate embedded correction is to bestow discretion upon a prior speaker's demonstrated incompetence/misconduct. By accepting the form, prior speaker accepts that implicit account. By rejecting the embedded form and the bestowal of discretion, prior speaker can reject the implicit account carried by that discretion. And by using the form which permits of attendant activities directed to accounting for the item in question, he provides a place in which to offer an alternate to the implicit account.

5 Moving out of closings1

G R A H A M BUTTON Plymouth Polytechnic

Introduction This chapter has a forebear in Schegloff & Sacks' analysis of the sequential organisation of closings for conversation (1973). In order to locate the concerns that will be developed, a brief summary of their findings can be made. Closings have a sectional design that operates to mutually co-ordinate the warranted suspension of a turn's transition relevance. A pervasive way in which a closing is organised spans four turns at talk. Closings may be properly initiated by the production of an item that is bereft of topic continuation or initiation features in a turn subsequent to a topic bounding turn. 2 This item, occupying this turn position 'gives a free turn' to a coparticipant and constitutes a first closing turn. A second closing turn returns a similar item and warrants the production of a first terminal in the next turn. A first terminal is a third closing turn, and under the auspices of the 'adjacency pair mechanism' it provides for the adjacent positioning of a second terminal. This is a fourth closing turn, and it is here that transition relevance is properly suspended. (1) [Erhardt:8:4] Pam: Vicky: *—> Pam:

hh Oh rwell than:ks ,any way , I : ' m so so rry Pa : m (•) Okay, = 101


102 *-» Vicky: *—» Pam: *—> Vickey:

= Okay = =Bye: = = Bye. . . . . end call. . . .

(2) [NB:III:5:10] Guy: Emma: Guy: Emma: Guy: Emma:

I'll be down there, oh en you'll - you'll be aroun' then when I ,(come in) Yeah. Okay. Okay dear, Buh bye, Bye bye, . . . . end c a l l . . . .

(3) [SBL:1:1:1:8] Bea: Dianna: Bea: Dianna: Bea:

And thanks for calling Alright dear, Alrighty Bye Bye . end call. .

The first and second closing turns typically consist of items such as 'okay' and 'alright', 3 and the third and fourth turns of items such as 'bye' and 'goodbye'. Items that occupy the first two turns positions will be respectively referred to as first and second close components, and items that occupy the third and fourth turn positions, as first and second terminal components. When the components are undifferentiated, they will be described as closing components. This Chapter will be mainly concerned with this closing design as the archetype closing,4 The archetype closing organises the termination of conversation over a section of talk rather than, for example, in the course of one turn. This displays an orientation to mutually legitimising conversation's termination. 5 The sectional design of closings is not only sensitive to conversational closure but also to the continuation of conversation. So, although a closing may be properly initiated, and although it may be in the process of being worked through, it is not the case that termination will 'automatically', so to speak, result once a closing has been initiated. In this respect, it is possible to observe that during the course of a closing, other material (in addition to, or instead of) closing components



may be introduced. The next turn may respond to this material and may not be occupied with closing components. (4) [Trio: 2:6-7] Marj: Pris:

Nothing else happen. = =n:Nothin. (0.2)

Marj: *—» Pris: (*)—» Marj: Pris: Marj: Pris:

Yeah. = =Okay. (•) Okay then see yuh-(.) Wednesday. = = Ya:h,= =Yarh (Tha-) No ,not (Wens)j No Thur sday.

(5) [TCI (b) 13:3-4] Jerry: Linda: Jerry: Linda: Linda: Jerry: *—> Linda: *—» Jerry: *—* Linda: Jerry:


Well I c'n leave right now if yih want, =No::,hhh khh-hh Whhi::::hh So:, (0.3) Okay? (0.3) Yah Oj_kayhjOney, Okay. Bye ,b ye (Y'wih) D'You w'n me t'bring home any

lightbulbs or anything? (0.3) —> Jerry: hrhh Linda: Uh:: ::, —>• Jerry: huh-huh-huh-(hr-uh) —» Linda: hhl dhhon't think so. . . . (6) [Closing Tape E: Call 18] A:

B: (*)r-* A: U A:

I'll see youh t'morrow then (•) A:wright. Ketcha in the morning. = =0:kay. =Hey(b)Bill?


104 B:



(•) Hey bring that uh: bone saw (. n) I got - I bought a (p)ork roast.

It is described here, how the introduction of other material into a closing may organise a movement out of closings which is properly reciprocated in the next turn. Attention will be directed to a number of 'sequence types' which are regularly found to occur in closings, and which provide for that moment to be followed by the next speaker. Each 'sequence type' will be individually examined. The positions within closings where a movement out can take place will then be considered, and finally it will be described how particular 'sequence types' are fitted and designed for the positions within a closing that they predominantly occupy.

'Sequence Types'6in Closings The following 'sequence types' are regularly used in, and move out of closings: — Arrangements — Back-references — Topic initial elicitors — In-conversation objects — Solicitudes — Reasons-for-calls — Appreciations Each of the 'sequence types' will be examined in order to observe the particular organisation of its movement.

Arrangements It is frequently found that arrangements are a last topic for conversation (Schegloff & Sacks, 1973). Upon their completion closings may be initiated and run to termination. (7) [JG:II(a):3:8] 7 Heather: Maggie:

Lemme know w't the doctor hastih sa.y. Yeah okay well ah'll call yuh later then.

MOVING OUT OF CLOSINGS *—• *—» *-* *-* *—>

Heather: Maggie: Heather: Maggie: Heather:


TOkay :,y Okay sweetie. Okay Bye bhhye Bye . . . . end c a l l . . . .

The regular achievement of arrangements as having the status of 'last topic' for the conversation involves at least two features displayed by arrangements. First, arrangements orient to conversation-in-a-series (Button, 1985) and arrangements may be used to provide an orderly relationship between 'this' encounter and a 'future' encounter — as opposed to 'next' encounters being by chance, for example. Second, by providing for a 'future' encounter, they may propose that a current encounter could be appropriately concluded and may, thus, also propose that further topics may be 'reserved' for 'the next time', or are, at least, unnecessary now. Arrangements may, consequently, be oriented to as closing implicative by participants, and this can be seen when following their completion one speaker initiates a closing section. However, the closing implicative nature of arrangements, notwithstanding, arrangement items may appear in closings. When they do, they typically display one of two organisational configurations. They either (1) overspill from the topic just prior to closing initiation, the arrangement itself, or (2) they reintroduce an arrangement which, indeed, was not the last topic of the conversation but which had been made sometime prior to a last topic. The overspill of arrangement items into closings may occur where the topic prior to closing initiation was arrangements, or where arrangement items, and arrangement related business, occupied a prior turn to closing initiation. (8) [JG:I(S)X15] —> —> *—* —»

Pete: Marvin: Pete: Marvin:

I'll see you Tuesday. Right. Ok Marv ,in You-you're alright? You can get there?

(9) [TCl(b): 16:90] —> Joan:

the t'wunny sekint Wednesdee night. (0.3)


106 —» Linda: *—* Joan: Linda: —» Joan:

Yeah nOka:f:y l Okay,= = ' hh A : : n ' p::referably my mom's if pohhss(h)ib(h)le

A r r a n g e m e n t overspill effectively continues arrangements as a topic, and s p e a k e r s may take on opportunity, before conversation termination, to introduce additional arrangement related business. W h e n arrangements are reintroduced (as opposed to overspilling into closings) arrangements have typically not been the immediately prior topic, or an item carried in a prior turn to closing initiation. (10) [MC: 11:3:4—5] Lila: Bush: Bush: Lila:

Yes. I told that tuh somebuddy but I-probably (didn't ). A'right, (0.3) rrFine, yeah that's right its is onna right hand. Okay. Ah—I'll be there between two thirty en four?

(11) '[Mc(Coven):53] Lila: Wilbur: Wilbur: Wilbur: Wilbur:

Well it's - u h ' i t ' s uh it's ,ten- it'sten minites of eleven. l Ehhhhhhhhhhhhhh heh heh heh Okay, Well -well maybe we will see each other uh : : : maybe not Thursday but ezsoon ez you c'n get do:wn. =

T h e actual arrangement may be preserved in the reintroduction by the inclusion of the time and/or day etc. the arrangement is made for. One p u r p o s e this may serve is to check out the arrangement made earlier (after all, meeting someone, or contacting them at a particular time involves some resource commitment, at least that another person is 'kept in mind', and not meeting, or not contacting someone when arranged is accountable; to check an a r r a n g e m e n t now may be resource effective) and the reintroduction thus furnishes recipients with items that can be confirmed or disconfirmed in the next turn. (12) [SBL:2:2:3:46] Chloe: Claire: Chloe:

Well, it was fun Cla jire Y e a h . I en,joyed every minute of rit. AndYe l ah.

MOVING OUT OF CLOSINGS (*)-» Claire: Chloe: Claire: Chloe:


Okay, well then we'll see you Saturday. Saturday night. Seven thirty? Yah.

(10) [MC:II:3:4-5] (*)—» Lila: —> Bush:

Okay. Ah—I'll be there between two thirty en four? Very good.

Following overspill, next turns can mark the receipt of continuation and development, or respond to the form of overspill (e.g. answering a question contained in the overspill). So, for examples of overspill: (8) [JG:I(S)X15] Pete: Marvin: Pete:

Ok M a r v i n . You-you're alright? You can get there? Yeah.

(9) [TCI(b): 16:88-90] —» Linda: £ Joan: —> Linda:

Okay. = = • hh A::n p::referably my mom's of pohss(h)ib(b)le h u h - h u h - , huh.-huh h,uh hh, hhhe: hhh = l Ye: ah. Right J

Arrangements in closings, thus, form a 'sequence type', where the arrangement reintroduction or overspill provides for a sequentially relevant next activity, and where upon its production the 'sequence type' is recognisably complete. This provides for, third, that next turns which follow items which are offered as the relevant next activity to the arrangement item, can be occupied by a closing component. That is, closings may be re-entered. However, the components that are regularly used to re-enter closings may not continue the closing in which an arrangement item or items were produced. It is a regular observation that they may be close components that operate as closing initial components. So, the next turns following a relevant response to arrangements may not contain terminal components that return to closings as if the arrangement intrusion were a 'side sequence' (Jefferson, 1972). Rather they operate to start closings all over again; to reinitiate closings.



Pete: Marvin: *-» Pete: —> Marvin: —* Pete: *—» Marvin: Pete: Marvin:

I'll see you Tuesday. Right. OK Mar f vin. You-you're alright? You can get there? Yeah. Ok. Ok. Thank you. You bet bye . . . . end c a l l . . . .

(9) [TCI(b): 16:90] Joan: Linda: *—> Joan: *—» Linda: *r-> Joan: L-» Linda: *—» Joan: Linda: Joan: Linda: Joan: Linda:

The twunny sekint Wednesdee night. (0.3) Yeah. nOka: f :y Okay. = = h h A : : n p : :referably mu mom's if pohhs(h)ib(b)le huh-huh ,huh-huh, -h,uh hh, hhhe : hhh = 1 Ye:ah. J l Right' Oka: :y. ÎO,ka ,. . y. hh Wi'll talk tih you la,ter. Oka^y. Bye bye. Buh bye .h . . . . end c a l l . . . .

In this respect, and in summary, arrangements in closings may be observed to move out of closings, whilst preserving the sequential relevance of closings. They may not be terminal elicitive, that is, they provide for material other than a closing component to be produced in a next turn and a movement out of closing is oriented to by the next speaker who provides sequentially relevant and appropriate returns to the arrangement overspill or reintroduction. However, the closing implicative nature of arrangements constrains the degree to which an arrangement 'sequence type' moves out of closings. This feature of closing implicativeness possessed by arrangements, and which is preserved in overspill and reiteration, provides for the relevancy of reinitiating closings. This closing implicativeness is also oriented to by the next speakers when they make a return to the arrangement overspill or reintroduction with an appropriate sequential response which is minimal in form, e.g. 'Yeah', 'right', etc. Although arrangement intrusion



may move out of closings, closings may be immediately relevant and reentered. This re-entry, however, may not continue the prior closing sequence that was in progress but may reinitiate closing. The movement out of closings made by this 'sequence type' is thus minimal. That is, although the arrangement projects that the next turn responds to the arrangement, the closing implicative nature of arrangements also projects that following the next turn responses to the arrangement, the subsequent turn may be occupied with a reinitiation of closings. In this respect arrangements move out of closings but another closing is immediately relevant following the completion of the 'sequence type', and, hence, the movement out of closings made by the arrangement 'sequence type' is a minimal movement.

Back References8 Material that has been previously topical for 'this' conversation may be back-referenced in closings. (14) [JG:1:10:7] Sam: Marge:

But then Fridee there's no schoo:l. Ye: ah.

(•) °Yeah° hhhhh ( W e i - ' hhh O: Kay honey, (•) Marge: Well y,ea::h, but I certainly feel thheirribly let down tuh think thet you didn't reconize mee. (15) [SBL: 1:1:11:4]

Marge: *—> Sam:




Bea: Tess: Bea: Tess: Bea: Tess:

Well I'll see you at-at Tomorrow^night, At six-At six o'clock Tomorrow ni-tomorrow night at six Yeah, okay Uh huh, And I'm sorry I didn't get Margaret, I really ve been wanting to

(16) [MC:III: 1:2-3] Lila: Jan:

I'll get off the line en order it right away. Yah




Jan: Lila:

Okay ^Lila Listen honey, if I—If you get my letter before you leave don't be ala:rmed hh

Unlike the sort of movement out of closings that is made by arrangements in closings, back-references are not closing implicative. Whilst arrangements were observed to move out of closings, the movement was minimal due to the closing implicative nature of arrangements that occasioned the immediate relevance of reinitiating closings. Back references do not operate in this fashion. They do move out of closings, and this can be appreciated by observing that they also, like arrangement components, occupy a turn that could be used to produce a closing component, and are overwhelmingly followed by a next turn that responds to the back-reference, as opposed to that turn continuing closings with the production of a closing component. However, the relationship between the back-reference turn and a next turn does not project a sequence trajectory that would occasion the relevancy of re-entering closings. This can be accounted for by the fact that a back-reference re-topicalises the material presented, and a next turn can now continue on topic in a manner that does not provide for the reinitiation of closings being relevant. (14) [JG:1:10:7] Sam: Marge: —» Sam: —> Marge: Marge:

O : kay honey, (•) Well yea: :h, but I: certainly fee:l thhe:ribbly let down tuh think thet you didn't recognize m,e. w'l tell you rthet i wz jsthhh you jus' t o o many Tgirl,friends tthat's all j


well the trouble is you don't call me

(15) [SBL:1:1:11:4] Bea: Tess: —» Bea:

Yeah, okay, Uh huh, And I'm sorry I didn't get Margaret, I really jve been wanting to Well I think she must've stayed out of town,

A weak formulation (a stronger one will be given shortly) of the sequential relationship of a back-referencing turn and the next turn, and of their movement out of closings, would be that next turns have the option to



topically develop the material referenced, and that if the option to topically develop is exercised, the relevance of closing reinitiation is undercut. This sort of movement out of closing which does not project the relevancy of reinitiating closings is a drastic movement. Neither the backreference turn, nor the next turn which exercises the option to develop on topic, provide for the reinitiation of closings in the manner described for arrangements. They are not closing implicative, and in contrast to the minimal movement made by arrangements, they drastically move out of closings. In this weak formulation, however, if the option to develop on topic from out of the material back-referenced is not exercised it might be expected that closings could be re-entered as they are for arrangements. This would, then, organise a minimal movement out of closings. On the basis of this weak formulation of the relationship between the two turns, it is the next turn to a back-referencing turn that is decisive for the degree of movement out of closings. However, a stronger formulation of the relationship between a backreference turn and the next turn can be provided when it is noticed (1) that back-references do, overwhelmingly, develop on topic and (2) that even if a next turn does not develop on topic for the material back-referenced, speakers who made the back-reference may, themselves, develop on topic. These two observations can be elaborated upon. Taking the first: the form that the back-reference can take may not provide just an option for development but may actually project that development is sequentially relevant with the inclusion of components that are designed to elicit particular next activities from the next speaker. Thus in example (14) Marge's back-reference to Sam's not having recognised her, is done in the form of a 'mild complaint'. This occasions the relevancy of some sort of apology or excuse, and is designed to receive this sort of activity. Also, in example (15) Tess' back-reference takes the form of an apology, 'And I'm sorry I didn't get Margaret, I really've been wanting to' which can project some sort of 'justificatory marking' in a next turn, and, again, is designed to receive this sort of activity. In this respect it is not just the exercise of an option provided by the back-reference turn, to develop on the material referenced that undercuts the relevancy of closing reiteration. Rather it is the actual backreference turn which projects development, and which, itself initiates a drastic movement out of closings. This means that the next turn may be sequentially constrained to develop on topic, and where not to do so would be to curtail an organised movement out of closings. T h e second observation can be elaborated on by looking to the development of fragment:



(16) [MC:II: 1:2-3] Jan: Lila: -» Jan: —> Lila: —* Jan: Lila:


Okay ,Lila Listen honey, if I-if you get my letter before you leave don't be ala:rmed 'hh Yah. Uh I just tried-y-have you gotten it ye:t, No, Well don't be upset a'tall, hh because I dis tried tuh . . .

Here, Jan minimally responds to Lila's back-reference to prior talk about a letter she has sent Jan. The back-reference does not project development in the ways described for the previous examples. However, the backreference does make the material 'current' again, and even a minimal return does not undercut this 'currency', and the back-reference is available to be further elaborated upon and developed by Lila. So, unlike arrangement reintroduction where a minimal return is recognisably the end of the sequence type, a minimal return to a back-reference may not necessarily close down the back-reference 'sequence type'. The material is still wide open for topical development. Thus, back-references within closings may make a movement out of closings which is reciprocated in a next turn. Even on the weak formulation of the relationship between a back-reference turn and the next turn, a drastic movement out of closings can eventuate. However, on the strong formulation the back-reference turn drastically moves out of closings by projecting the relevance of development, and provides an occasion for development by the back-referencing speaker even if the next speaker minimally responds. Back-references, then, do not project the relevancy of reinitiating closings. Their movement is a drastic movement out of closings. Topic Initial Elicitors9 The first part of a sequence designed to generate a new topic may appear within a closing section. It operates as a topic initial elicitor, and is oriented to eliciting a newsworthy item from a next speaker that has the status of topic initial. (17) [HG:15] Nancy: Hyla:

= You'll come abou:t (,) eight. Right? = =Yea::h,=



= Okay (0-2) Anything else to report

(18) [JG:III: 15:3]


Maggie: Marge: Maggie: Marge:

H'ri ((bruskly)) Okay? Bye ((bruskly)) Okay. Iz there anything else you: u-happen today of any interest?

(19) [JG:I:(S):X15] Pete: Marvin: *—» Pete: —» Marvin:

Yeah ¡'11 be there It's alright huh? Ok Marvin. How are things goin?

(20) [Frankel:QC:2:l:10] Ma: Kiddo: *-» Ma: *-» Kiddo: —> Ma:

Uhri:ght? Yeah. pt. Oka:y = = Uhriight. = = What else,

(21) [Krakowski:D + R: 11-12] David:


Robin: David: Robin: Robin:

Well so, I-I don' hhhhhh ohkay. H H H H H H H (0.3) (rih) Ohkey. Ahright. (0.2) Ahhh (1.0) I don' know what else to say to you David =

Topic initial elicitors signal a continued availability for conversation but they do not present a specific item, or items, for the next speaker to talk to. This feature is particularly sensitive to closings since they can result in a drastic movement out of closings whilst orienting to the relevancy of closings. This can be seen by noticing that topic, initial elicitors make provision for the introduction of further newsworthy items, or possible topic initials and in so doing they signal that speakers are available for further conversation as



a possible new topic. Next turns may be subsequently occupied with the introduction of further newsworthy items. (17) [HG: 15] Nancy: Hyla: Hyla:

Anything else to report (0.3) Uh::::::m:::, (0.4) Getting my hair cut tihmorrow,=

(18) [JG:III: 15:2-3] Marge:

C Maggie:

Okay. Iz there anything else yo: u-happen today of any interest? No (0.5) huh hu hh (throat clogged)) sevente ((clears throat)) seventeen dollar for Ronald's teeth

(21) [Krakowski:D + R: 11-12] Robin: —* David:

Ahh( 1.0)1 don' know what else to say to you David = = Are you goin' over to my mother's house?

However, next turns to topic initial elicitors may be used to decline to present topic initials, and following a decline, closings may be reinitiated. (20) [Frankel:QC:2:l:10] Ma: Kiddo: Ma: Kiddo: Ma:

pt. Oka:y = = Uhri:ght. = = What else, Noth:in', Ok:ay.=

Topic initial elicitors, then, may make a movement out of closings. They occupy a turn that could otherwise be used to produce a closing component and provide for next turns to produce a sequentially relevant item, either the presentation of a newsworthy item or a decline, and thus provide for the next turn to be occupied with material other than closing components. However, they are organised in a way that displays characteristics associated with a drastic movement out of closings, as exemplified by back-references, and characteristics of a minimal movement, as exemplified by arrangements. With respect to the first: in making provision for a topic initial to be produced in next turn, then, if that is done, neither the topic initial eliciting turn nor the topic initial turn occasion the relevancy of



closings, and here a drastic movement out of closings may be organised. H o w e v e r , and with regard to the second characteristic, should a decline to present a topic initial occur, closings are again relevant and a minimal m o v e m e n t takes place. This 'sequence type' can thus be seen to operate in closings with an 'unless' clause. 'Unless' a topic initial is produced, closings remain relevant. H o w e v e r , this 'unless' clause operates within a 'preference system' (Sacks, this volume, Chapter 2) that prefers a topic initial utterance over a decline (Button & Casey, 1984). Although a topic initial elicitor does not provide the next speaker with an item that can be 'talked on' nevertheless, the very occurrence of an enquiry into the possibility of presenting further newsworthy material for the conversation, and the display of availability for talk on that item that this involves (Button & Casey, 1984) further displays that a speaker may be orientated to conversation continuation. Thus in as much as a topic initial elicitor, which 'prefers' the production of a topicaliser in a next turn, occupies a turn position that could also be occupied with a closing comp o n e n t , this 'preference' may be extended and also displays a 'preference' to conversation continuation over conversation closure. Next speakers who follow a topic initial elicitor may display a sensitivity to this 'preference' for continuation by producing a topic initial. It can be n o t e d , however, that topic initials or newsworthy items which follow topic initial elicitors have the status of candidate topic initials and require the next s p e a k e r to upgrade them. This may be done by topicalising the newsworthy event that is reported and, by providing for talk on that item. For example in extract (17) it is found that Nancy who initiated the sequence, 'topicalises' Hyla's possible newsworthy item and it is following this topicalisation that the item is topically developed. (17) [HG:15] Nancy: Nancy: Hyla: Hyla: —> Nancy:

Okay. Anything else to report (0.3) Uhi:::::m:::, (0.4) Getting my hair cut tihmorrow, = = Oh rilly (continues on topic)

In this respect both speakers are thoroughly implicated in the generation of a new topic for a conversation where the initiation of closings has



proposed that the conversation may be concluded, and this displays a sensitivity to a closing environment whilst also demonstrating the 'eliciting force' of topic initial elicitors for conversation continuation. This 'eliciting force' is also displayed by activities that can be produced following a decline to present a topic initial. Here the next speaker, can 'recycle' the decline. (19) [JG:I(S):X15] Pete: Marvin: Pete: —* Marvin:

Ok Marvin. How are things goin? O h - h - h - h nothin, doin. Nothin' doin' huh?

It has been observed that a decline to present a topic initial can occasion the relevancy of reinitiating closings. Recycling the decline in a position in which closings may be reinitiated, however, provides a further opportunity to present a possible topic initial (even if it is that there is no-news) and, thus, may continue a display that the current speaker is oriented to conversation continuation even though they are not, themselves, proposing a new topic. Recycling, can then be used to display that topic initial production was the 'preferred' activity. In view of the eliciting force of topic initial elicitors, the previous description of their movement out of closings can be revised. Since they invoke a 'preference' constraint for next turns, topic initial elicitors may initiate a drastic movement out of closings that does not project that relevancy of reinitiating closings 'unless' next turns are unresponsive to the preference constraint in which case the movement may be curtailed and result in a minimal movement, with the subsequent reinitiation of closings. 'In-Conversation' Objects 'In-conversation' objects may be used in closings and may be used to initiate a movement out of closings. (22) [F: TC: I: i: 22—23] Shirley: Geri: Geri: *-» Shirley: —» Geri:

hhhhhhh ,Good w'l have coffee. loo ( (0.3) °Oka:y Alright? M m - h m?



(23) [TCl(b)13:5] Linda: Gerry: *—> Linda: —> Gerry:

I'mfi:ne? Fi:ne en you rokay, Okay hon, Orkay, 'hh Uh::::m,

These sorts of items may be used in conversation to mark the receipt of prior items, and to provide for the prior speaker to continue. For example, the recipient of a story may use them to signal the receipt of story items, whilst providing for the story-teller to continue the story (Sacks, 'Second Stories', unpublished). When they are used in closings it may be observed that the next speaker rather than continuing with a closing component, actually offers new material. It, thus, appears that a next speaker may orient to these items as signalling that the prior speaker is 'in-conversation', and available for conversation continuation. (22) [F:TC:1:1:22-28] Geri: Shirley: Geri: Shirley: Geri:

°Oka:y Alright? Mm-h,m? D'yih talk tih Dana this week? hhYeh...

(23) [TCl(b)13:5] Linda:

—> Linda:

okay hun, hh o,kay, hh Uh::::m, (0.6) U h : : huh, tryina(c) ( ) clean house . . .

• The display of availability for conversation continuation resembles, in aspects, the display of availability made by topic initial elicitors. That is, although they may result in new material for the conversation, 'inconversation' objects do not themselves, offer any new material. Yet, it is following them that new material is introduced. In this respect, notice for examples (22) and (23) that the production of new material by the recipients of 'in-conversation' objects is specifically done on behalf of the prior speaker. So, for fragment (22) Shirley's enquiry as to whether or not Geri talked to Dana (Geri's boyfriend), nominates a newsworthy event as a possible topic which belongs to Geri. In extract (23), following Jerry's ' U h : m ' , Linda exhibits a 'search for something to say'10 — 'Uh:hh' — and comes up with something that may qualify as a non-newsworthy item which requires Jerry's collaboration to transform it into a newsworthy item.



Thus, the new material introduced by the recipients of 'in-conversation' objects may not be closing implicative in as much as it does not project the relevancy of reinitiating closings but provides for the next speaker to continue some form of topic productional activity. In this respect it is a drastic movement out of closings. However, notice for the two examples that the new material is initiated in response to the prior turn, and thus the movement out of closings can be located as originating in the 'in-conversation' object. Thus, 'in-conversation' objects, like topic initial elicitors, may display an availability for conversation although they do not introduce new material as is done by a back-reference. However, topic initial elicitors have closing implicative features — noted in the operation of the 'unless' clause. 'In-conversation' objects do not, quite simply because they do no more than signal continuation. Thus, in as much as new material for the conversation is offered as a result of a signal for continuation the movement out of closings originates in the 'in-conversation' object, and in as much as this material may nominate a new topic, the movement out of closings may develop as a drastic movement. So, intriguingly, a 'minimal utterance' can initiate a drastic move out of closings, and this exemplifies the highly sensitive character of closings which is referred to later. Solicitudes Solicitudes are another 'sequence type' that frequently occurs in closings. Like arrangement overspill and arrangement réintroduction they are not terminal elicitive, and this is because they may provide for, as a relevant next activity, items other than closing components. (24) [JG:III:10:1] *—» -» —» —>

Pam: Marge: Pam: Marge: Pam:

Thanks a lot. 'N I'll see you soon Okay honey = = Okay Dri:ve ca:reful I: wirllh::

(25) [JG: 1:24:14] Marge: Laura: *—» Marge: Laura: Marge: Laura:

O r et least write a note, Alrighty, Okat Ho f ney Alrighty,honey.. Okay Bye: TGive my love to the: uh-family. I sure will



Thus, solicitudes may be observed to organise a movement out of closings which if followed by the next speaker. Again, similar to arrangements in closings, the turn following a solicitude response can be used to initiate closings. The response is the recognisable end of the sequence following which closings may be returned to. It is observed that, again, as for arrangements, closings are reinitiated rather than continued. This displays an orientation to a movement out of closings having taken place rather than a 'side sequence' having been produced within a closing. (25) [JG:I:24:14] Marge: Laura: Marge: Laura: *—> Marge: *—» *—> Marge: *—> Laura:

Okay horney, Alrighty,honey.. Okay Bye: TGive my love to the: uh-family. I sure will. Okay hon,ey. Okay (sweetheart) B yebye , Bye-Bye, . . . . end c a l l . . . .

In this respect, solicitudes in closings may be a 'sequence type' that makes a minimal movement out of closing. Instances can be found, however, that do not run off in this fashion. For example, solicitudes may be elaborated upon in such a way that the relevancy of reinitiating closings is not preserved, and also closings may be continued as opposed to reinitiated following the solicitude 'sequence type'. Nevertheless, such activities may exhibit an orientation to the appropriate sequence projected by a solicitude. With respect to the first point, it can be seen in the following fragment that Marge elaborates upon her solicitude and in so doing may undercut the appropriateness of a minimal response and the subsequent relevancy of reinitiating closings. (26) [JG:Reel6:10] Marge: Marge: * Maggie: *-» Marge: *-> Maggie: [—> Marge:

Yeah. Oh yeah I will tell you when ( ) 'hhh Okay then I'll see ya later. A'ri. Okay = = „Bye. Well take it easy comin home cuz they'll probably



-» -» -» L-»

aleverybody be trying t'get home. = Now I noticed that down here across the streetuh they 'ave been runnin crazy. = So evidently people are beginning t'already get home

Marge elaborates on the solicitude and uses it to introduce a possible new topic. However, unlike topic initials that follow a prior topic initial eliciting turn which warrants and legitimises their introduction, in the present case, the nominated topic is not provided for by a prior turn. Thus Marge may attend to its improper placement by embedding it within the solicitude 'well take it easy comin home' which it also expands. Although Marge might attempt to move away from the relevancy of reinitiating a closing by nominating as a topic, and attempting to get talk going on 'the traffic', she, nevertheless, orients to the reinitiation of closing as an appropriate sequence projected by a solicitude through attending to a topic's nomination as improperly placed. In the next example the return to the solicitude is not followed by a close component that reinitiates closings, but by a first terminal that continues closings. (24) [JC:III: 10:1] Pam: Marge: Pam: Marge: Pam: Marge: Pam:

Thanks a lot. 'N I'll see you soon Okay honey = = Okay Dri: ve ca: reful I: wi:ll rh:: h::ahaByeby r e Bye bye . . . . end c a l l . . . .

However, the production of the first terminal is oriented to its possibly abrupt and misplaced character. Marge 'softens' its placement by using a 'laughing preface'. That is, although Marge places a first terminal following the return to the solicitude she displays a sensitivity to its abrupt placement by delaying its delivery through laughter, and so 'softens' or 'eases' its production. So, solicitudes within closings appropriately make a minimal movement out of closings and following the completion of the sequence, closings may be reinitiated. An orientation to this projected sequence can be preserved even in cases where the course of the projected trajectory is not followed.



Reason-for-Calls Referencing, or reiterating the 'reason-for-call' is one way in which the relevancy of closing initiation can be established (Schegloff & Sacks, 1973) The 'reason-for-calls' may operate in this way because they may occupy a turn position that could also be used to continue on a topic, or initiate another topic, and similar to arrangements can display that no further new material for the conversation is being introduced. A next turn can then initiate closings. 'Reason-for-call' reiteration is thus closing implicative. (27) [Lerner:SF:II:24] ->1 2 3 r 4 ->5

Mark: Bob: Mark: Bob: Mark:

Ue *—>7 Bob:

Uh . . mhh I didn't feel rebu:ffed, Wul good. .hhh , B u : t u h = Good = having talk't' Jo Ann I did wanna git thee f:full skinny = = h h - h h h h ' h h Okha(h)y,

In this example, Mark's 'reason-for-call' is to reproach Bob for not inviting him to a party. They have gone on to talk about a previous party that was held at Mark's house but which Bob had missed. Mark, however, returns to the present 'reason-for-call' (line 1) and Bob subsequently (line 7) initiates a closing. 'Reason-for-calls' may also be referenced within closings. As with arrangements and solicitudes they may occasion a returned utterance, but also preserve closing implicature; so that following the return, closings can be re-entered. This re-entry can, again, be observed to be organised as a reiteration as opposed to closing continuation, and again this displays an orientation to a movement having been made out of closings. (28) [NB:IV:12:3] Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma:

Well I'll talk tih yuh later dear. „Weill, U IWell I wz just getting dressed. I know yer goin. Yeah. AARIGHT



-» Emma: *—» Lottie: Emma: Lottie:

I JUS T H O U G H T I'D TELL YUH. T H E R E W'Z [ [ ONE KUHOkay. Yea:h, Okay hon, Bye, Bye. . . . . end c a l l . . . .

Similar to arrangements and solicitudes, reason-for-calls in closings are not terminal elicitive in that they may occasion a relevant next activity, material which is not a closing component, but they may retain their closing implicative character and hence their movement out of closing is organised as minimal movement.

Appreciations Appreciations in closings take two forms.'' First, they may be appreciations which locate the call as the object of appreciation. (29) [SF:I:13:C] Mark: —> Joanne:

Ah'll talk tuh yuh later Okay Spark ^thanks for calling

(30) [MC:I(Coven):54] Wilbur: Lila: Wilbur: Wilbur: Lila: Wilbur: —» Lila:

Well then let's do it this week. En all ,we'll haftuh do = Yeah. = is just-have a chat fer a few minutes. „That's all, Oh that's right, but w e - i t isn't gonna be anything ( ). 0:kay. Okay. And I'm just delighted to talk t'you.

(31) [MC:II: 11:38] Wilbur: Lila: (*)—» Wilbur:

Okay. Well up t'this moment, three thirty Friday. Tha:t'll be firne. Okay. Thanks a million fuh calling Lila.

Second, they may be appreciations which refer back to an 'appreciable' — such as a favour — that is locatable somewhere in the prior talk.



(32) [JG:1:10:7] Marge: Sam: Sam: *—» Marge: —» Sam:

'hh A : n ' uh tell uh:ar li'l girlfriend or oTther little girlfrie,nd hello an' everything like that. = l ( ) = Iwilldeah. 0:kay.= = Thank you,

(33) [SBL:1:1:9:4] Avon Lady: Meg: Avon Lady: *—> Meg: —» Avon Lady:

Uh that is a very good value. Uh uuh, SoOkay now. Thank you dear,

(34) [NB:IV: 11:4] Emma:

If you need us? or n'd-uh want a thing y'know we're right he:re ,so, Martha: " Well- thank you very much ,dear en I'll Emma: Aa'right, Martha: probly see you later r on in the da: y Emma: Aa'right. Emma: Aa'right, —» Martha: Thank you. Appreciations can be followed in a next turn by closing components which continue the closing which the appreciation appears in, as opposed to the reinitiation of closings which has been observed to operate for the other sequence types so far examined. That is, appreciations following closing components may, themselves, be followed by the sort of closing component that would occupy that turn position in an archetype closing. So, an appreciation within a first turn can be followed by components such as 'okay', and an appreciation in a second closing turn can be followed by a first terminal. (24) [MC:II: 11:38] Lila: —> Wilbur: p> Lila: L^ Lila:

Tha:t'll be fi,ne. Okay. Thanks a million rfuh calling Lila, Okay Okay.


TALK AND SOCIAL ORGANISATION (35) [Kamunsky:ll:7] Alan: —> Shawn: -» Alan:

'hh Well Xav s'm homework t'do I thought I'd js stop 'n call you. A't,hank you., l Oka:y? J

(32) [JG:1:10:7] Marge: Sam: Sam: Marge: —> Sam: —> Marge: Sam:

hh A : n ' uh tell uh:ar li'l girlfriend ar oTther little girlfrie: f nd hello an' everything like that. = l ( ) =Iwilldeah. Oikay. = =Thankyou, Bye:,bye, B'bye. . . . . end c a l l . . . .

(33) [JG:1:10] Riva: —> Sam: —» Riva: Sam:

O^kay. = :Thank you Bye-bye [ B'bye. . . . . end c a l l . . . .

In these cases appreciations appear to operate as floor-passing components. However, speakers may differentiate appreciations from floor-passing components such as 'okay' in that they may organise a returned appreciation, as located in a prior appreciation. This can be seen in the following example: (36) [JG:I:8:6] Marge: Colin: Marge: Colin: —» Marge: Colin: Marge: Colin:

Al:riight? = =Ofka:y, A: n: d uh, hh (•) Right. 'hhhhh Oka:,y, thankank you Mrister Han ,S'n = Oka:y thanks a lot =Bfye: Kay G'bye:, . . . . end c a l l . . . .



Where a close component is properly used a next speaker may legitimately begin a next turn and return a close component. However, should the shape of a prior turn change with the inclusion of an appreciation, a current turn which began by exercising the option of advancing closings may also change. In the above example, Colin begins his turn slightly overlapped with Marge's close component. But Marge changes the shape of her turn by continuing with an appreciation. Colin, in response, also changes the shape of his turn by returning an appreciation. That is, Colin produces components in his turn which reciprocate each of the components making up Marge's turn, and does so as Marge changes the shape of her turn. Although appreciations may also receive reciprocal turns, nevertheless, the reciprocation can also be followed by a closing component which continues the closings in which the appreciations appear, as opposed to reinitiating a closing. This can be seen in the above, and in the following two fragments. (37) [MG:II:6:9] Alfred: Lila: Alfred: Lila: Alfred: Lila:

ahh ha-ha-ha hh Well, we'll see what happens. Okay, Thank you very much rMissiz Asch, Thanks for calling, G',bye Bye bye, . . . . end c a l l . . . .

(30) [MC:I(Coven):54j Wilbur: Lila: Wilbur: Wilbur: Lila: Wilbur: Lila: Wilbur: Lila: Wilbur:

Well then let's do it this week. En all ,we'll haf tuj do = Yeah. = is just- have a chat fer a few minutes. = That's all, [[Oh that's right, but w e - it isn't gonna be anything ( ) Otkay. Okay. And I'm just delighted to talk t'you. Thank you fer calling. Goodbye::, Goodnight. . . . . end c a l l . . . .

Appreciations can, then, operate as an independent 'sequence type' but still receive terminal utterances. Unlike the other 'sequence types' that implicate closings, appreciations, as a 'sequence type' in closings, may not



occasion the reinitiation of closings, but are followed by closing components that continue a closing in progress. Apprectiations may also form an independent 'sequence type' where the appreciation is not reciprocated, but, nevertheless, responded to by an acknowledgement such as 'alright'. In these cases, again, a first terminal may eventuate. (38) [SBL:3:5:10—11 ] Milly: *

Ginny: -» Milly: —» Ginny: *-» Milly: * Ginny:

An' then Marcie called me about another meeting tuhday but I'll heftuh letche go. hhhh'hh'0::,kay, Okay thanks ^Ginny, Alright, Bye::, Bye bye. . . . . end c a l l . . . .

Although appreciations can operate as an independent 'sequence type' within closings the fragments so far examined suggest that they do not operate to move out of closings since the closing is not reinitiated, but continued, following the completion of the appreciation sequence. Appreciations can operate either as floor-passing close components or generate an independent 'sequence type' which can operate as a closing extension (Schegloff & Sacks, 1973), but without moving out of closing. Appreciations may consequently be seen as either close components or as an optional closing insert. It is observed, however, that appreciations can be associated with movements out of closing. Whilst appreciations can operate as though they were closing components they may have sequential implications for next turns that are not associated with the archetype close components. Unlike components such as 'okay', appreciations are not totally devoid of topicalisable features. As a consequence of this, appreciations may initiate a movement out of closings. This is because, unlike archetype close components, appreciations can be addressed and talked to in next turn. In the following example James specifically addresses the fact of Sue's appreciation having occurred and exploits this to do a back-reference. Following Sue's reiteration of a previous 'excusing marker' closings are reinitiated.



(34) [SB:0:4] James: Sue: —> James: Sue: James: Sue: James: Sue:

0:kay Thank u-thank you Well I'm not so sure taking you a way from v'tea I said that didn't matter, Alright, 'kay By^e Bye . . . . end c a l l . . . .

So, although appreciations may operate as though they were close components, unlike archetype close components they can be specifically referenced in the next turn. This feature of appreciation derives from the fact that they may reintroduce conversation into closings that was relevant prior to closing initiation. In this respect appreciations can make a movement out of closings. However, in as much as it can be observed that appreciations can also operate as though they were close components, this movement does not have such strong sequential implications for next turns that other 'sequence type' movements may have. The movement may, or may not, be picked up in a next turn. The movement is 'concealed', and it takes a next turn to reveal it. For cases, such as the above example, where a next turn to an appreciation reveals the movement made out of closings by appreciations, closings are reinitiated within the sequential environment of the movement-revealing turn. In this respect, appreciations when they are revealed as having moved out of closings seem to be associated with a minimal movement out of closings. The 'sequence types' that have been examined, regularly, and repeatedly, appear in closings. Each one of them has been observed to move out of closings, and the degree of movement made by each 'sequence type' has been examined. It can be concluded that each of the 'sequence types' falls into one of two categories of movement out of closings; either a drastic or minimal movement. T o summarise: a minimal movement is made when the 'sequence type' is not terminal elicitive but is closing implicative. The initial sequence turn projects the shape of the following turn. The next turn properly responds to the movement, and marks the end of the sequence. However, the closing implicative nature of the initial movement is preserved in that turn, and following it, next turns reinitiate closings. 'Sequence types' that fall into this



category of minimal movement out of closings are 'arrangements', 'solicitudes', 'reason-for-calls' and 'appreciations'. 12 That a movement out of closings has taken place, as opposed to the sequence type being a sequence inserted into closings, is oriented to by participants in as much as closings are reinitiated. The closing is not continued where it 'left o f f . Drastic movements, on the other hand, are not closing implicative. They move out of closings without projecting the relevancy of reinitiating closings. 'Sequence types' that produce this movement out of closings are 'back-referencing', 'in-conversation-objects' and 'topic initial elicitors'. 13 The fact that particular 'sequence types' are associated with degrees of movement out of closings can be observed to be sensitive to the particular places within a closing sequence in which the 'sequence type' regularly occurs. The particular 'sequence types' can be found to be fitted for the particular positions within a closing, and the degree to which they move out of closings is directly relevant for the position they occupy. Before turning to examine this feature of 'sequence types' it is necessary to back-track a little to reveal closings as having systematic places where a movement out of closings can be organised.

Opportunity Spaces for Moving Out of Closings It is possible to observe that movements out of closings occupy particular positions within the boundaries of the archetype closing section. Simply, these positions follow closing components. Of particular interest, however, is the observation that can be made that some sequence types are overwhelmingly used in particular positions within closings. Before examining this distribution, however, an array of data can be presented to elaborate upon the actual places within a closing where movements out can take place. 14 Following a First Possible Close Component A movement out of closings can take place following a first possible close component in either the same turn or in the next turn position. The onset of a turn can initiate closings: (40) [Erhardt:10:3] Karen: Vicky:

°(It wz pretty neat r yeah]° l °Good.° (0.2)

—» Vicky:




(41) [JG:III:21:7] Rita: Nell: Rita: Nell: Rita: *—» Rita:

Well they may have gone out of town Yeah they might of Yeah But anywa:y um: (1.0) everything else is oka:y Yeah'hh Well then okay then honey

but may have its shape changed by the inclusion of components that move out of closings. This constrains the shape of the next turn. If only a closing component is produced in a first turn, next turns may either return to a closing component or continue in conversation by organising a movement out of closings. But if a first closing turn has its shape changed by the inclusion of material that moves out of closings, next turns may attend to the movement and not to the closing component. (40) [Erhardt:10:3] Vicky: Vicky: Vicky: Karen: Vicky: Karen: Vicky:

Oka:y well I jis carlled tu:h (0-4) " teh:: (•) ask, = = Thanks f a lo:t , though'v cour se I knew rthe ans,wer would really n g - hah

(41) [JG:III: 21:7] Rita: Nell:

hh Well then okay then honey I'll see you tomorrow evening Alrighty I'll be there ( ).

Where a first closing turn does not have its shape changed by the inclusion of components that move out of closing, but stands as a closing initial, the next turn may be used to move out of the closing. Again, this has consequences for a following turn which may attend to the movement and not continue the closing. (42) [SBL: 1:11:4] Bea: Tess:

^At six At six o'clock. Tomorrow ni- Tomorrow night at six.




Bea: Tess: Bea:

Yeah, okay, Uh huh, And I'm sorry I didn't get Margaret, I really fve been wanting to Well I think she must've stayed out of town,

(23) [TCI(b)13: 5] Jerry: Linda:

En you^? Fijne?

Linda: Jerry: Linda: Jerry:

" (1-0) En you^, (0.8) I'm fijne? Fi: ne en you ,-Okay, Okay, hon, hhh o^kay hh Uh::::m,


Uh::hh, tryina(co) (.) clean house . . .


(0-6) For all movement out of closings in these positions, and in the following positions that will be noted, next turns are constrained to reciprocate under the auspices of the particular 'sequence type' that makes the movement. Following a Second Close Component If a turn following a first component is used to offer a second closing component rather than being used to initiate a move out of closings, then, closings are further advanced and developed. A closing component in this second position provides for the warranted production of a first terminal. However, a second turn begun as a closing turn can have its shape changed by the inclusion of components that move out of closings, and also a third turn that could be used to offer a first terminal may also be used to initiate a movement out of closings. The following are examples of second closing turns that are being used to move out of closing through changes in the turn's shape. It can be noticed that the next turn responds to this movement. (30) [Mc(Coven):54l Lila: Wilbur: ( * ) - » Lila: —» Wilbur:

Oh that's right, but we-it isn't gonna be anything ( )• 0:kay. Okay. And I'm just delighted to talk t'you. Thank you fer calling



(43) [SBL:3:5:10] Milly: *->• Ginny: (*)-» Milly: —> Ginny:

A n ' then Marcia call' me about another meeting tuhday but I'll heftuh letche go. hhhh'hh 0::fkay, Okay thanks ,Ginny, Alright,

The following fragments are examples of third turns being used to move out of closings rather than being used to produce first terminals. Again, the 'sequence type-initial' provides for the next turns to respond to that movement. (24) [JG:III: 10:1 ] *—» *-» —» —»

Pam: Marge: Pam: Marge: Pam:

Thanks a lot. 'n I'll see you soon Okay honey = =Okay Dri:ve ca:reful I: will h::

(20) [Frankel:QC:2:1:8-10] Kiddo: Ma:

Kiddo: Ma: Kiddo: Ma: Kiddo: —> Ma: —» Kiddo:

We'll s,ee. When you and Vicki comes home. When Mark comes ho:me. (1.0) Uhright. (0.2) Uhrhght? Yeah. pt. Oka:y. = =Uhri:ght. = = What else, Noth:in',

Following a First Terminal The production of a first terminal in the third position in closings provides for the next turn to be occupied with a second terminal at which place transition relevance may be suspended. However, a movement out of closings can also be found to be initiated in the next turn following a first terminal.


132 (43) [TCI (b): 13:3-4] Jerry: Linda: Jerry: Linda: Linda: Jerry: Linda: Jerry: Linda: Jerry:


Jerry: Linda: Jerry: Linda:

Well I c'n leave right now ih yih want, = =No::, fhhh khh-hh hhi::::hh So:, (0.3) Okay? Yah. 0 : k a y hroney, Okay. Bye b,ye, (Y'wih) D'you w'n me t' bring home any lightbulbs er anything? (0.3) hrhh l Uh::[::, huh-huh-nuh(h,uh) hh I dhhon't think so. . .

(44) [JG:III: 15:2-3] Marge:

*—» *—> *—* —»


Maggie: Marge: Maggie: Marge: Maggie: Maggie:

hh Oh: well. Okay uh then I'll letchu( ) ah well I mean I won't call you if I'm gonna - if they're gonna drop me off If they're not gonna drop me off then I'll just call up (0.5) Ah'ri ((very brusquely)) Okay? Bye ((very brusquely)) = But go ahead an feed hi ,m ( ) imagine hez starving Yeah Yeah . . .

The data corpus 15 reveals that movement out of closings following a first terminal is not as frequent as movements following first and second close components. Also, no cases were found where a movement out of closings was initiated within a first terminal turn and where the first terminal followed a second close component. All cases where movement out of closings was initiated, following a first terminal, occurred where the first terminal did not occupy its sequential position as third turn in the archetype closing. The following fragment exemplifies this for now, but both of these observations will be returned to later.



(45) [NB:III:2:4—5] *

Jim: Ted: —> Jim: Ted:

We'll see, yuh (later) 0 : : K a y Jim, Bye bye - H o w big'r those waves down there. (0-6) Oh, about thirty foot I guess.

Having examined the places within a closing section where movements out of closings may take place, it is possible to find that closings are designed in such a way that they provide an opportunity for movements out of closings to take place. Closings are designed with 'opportunity spaces' 16 which can be used by particular speakers to initiate 'sequence types' that move out of closings. There are five opportunity spaces in an archetype closing. Two are available to first speaker and three are available to second speaker. All the opportunity spaces are ordered in relation to closing components. The first two opportunity spaces are ordered with respect to a first close component. The first is available to first speaker. A first speaker who produces a close component can change the shape of that turn by including material that moves out of closings. This takes place in that turn's transition relevance space. 17 If that opportunity space is not used, a second opportunity space exists for 'this' closing which is a first opportunity space for second speaker. This is in next turn position to the first close component. Although it will be seen that there are 'good organisational reasons' for moving out of closings in any of the opportunity spaces, the second opportunity space is particularly crucial as this turn represents a first position where second speaker can display whether or not they are collaborating in closure. Should a speaker offer a close component here, a third opportunity space is provided by that turn's transition relevance space. A second speaker who produces a second close component may change the shape of that turn by introducing material that moves out of closings. Should a second closing turn only be occupied by a close component, a further opportunity for first speaker to move out of closings is provided by the next turn position. This is first speaker's second opportunity space, and the fourth opportunity space for the closing. If the first speaker produces a first terminal then the second speaker has a third opportunity space which is the next turn position to move out of closings. This is a fifth opportunity space for the closing. Movements out of closings do not seem to be organised by the first speaker within a first terminal transition relevance space in the archetype closing. This is a



methodic and systematic result of the distribution of opportunity spaces for particular speakers. The first terminal's transition space would be a third opportunity space for first speaker. However, the prior opportunity space is also first speaker's. Thus if a movement out of closings is going to be initiated by the first speaker within the environment of a first terminal turn, the movement may be caught in first speaker 'second opportunity space', and a first terminal would not subsequently result. Cases where a movement out of closings in a first terminal's transition space does occur are associated with the misplacement of a first terminal. In the archetype closing, a first terminal is produced by first speaker. A first terminal may be misplaced when the second speaker produces a first terminal in the position occupied by a second close component in the archetype closing. If the second speaker produces a first terminal in this position, then this turn's transition relevance space is the second speaker's second opportunity space. But since a first terminal could be followed by a second terminal and the suspension of transition relevance, this opportunity space is also second speaker's potenitally last opportunity space to move out of closings. Changing the shape of a first terminal turn in this position can regain an opportunity space that the premature placement of a first terminal may have lost. It can also attend and orient to the first terminal's misplacement. The archetype closing is designed with opportunity spaces in which a movement out of closings can take place in the interests of preserving the mutuality of closings throughout the course of a closing. That is to say, closing components establish a warrant for the production of further closing components, and working towards termination becomes a thoroughly mutual matter. Unlike some sequences in conversation, where a prior turn constrains next turn to produce a particular activity, a next turn to a closing component is not constrained in such a strong implicative fashion. 18 In these terms closings provide opportunities for movement out of closings where a speaker can abandon the mutual progression towards termination. The early placement of a first terminal noticed above introduces the observation that variants on the archetype closing may be produced. These variants affect movements out of closings. In order to display this, two variants on the archetype closing can be examined: foreshortened and extended closing. Foreshortened closings may reduce the number of opportunity spaces for a closing, and may display that one speaker is making a strong attempt to close the conversation, whilst extended closings increase the number of opportunity spaces in which a movement out of closing may be initiated and may display that a speaker is 'looking for' continuation.




In the archetype closing, the relevance of lifting a turn's transition relevance is established over the course of four turns at talk, each turn producing different components that make up the closing. Closings may be constructed, however, by packaging more than one closing component within a closing turn. In so doing, a turn in the archetype closing may be skipped over and a foreshortened closing sequence that advances the imminence of termination may be accomplished. A first closing turn may be used to produce a first close and a first terminal component. (46) [JG:IV:3:314] Ronald: Marge: *—> Ronald:

When d'you wan'us tuh pick you up about - five thirty? Oh, I think so, Okay, guhby Duval,

(47) [NB:IV:5:4] Martha: Emma: *—> Emma:

I'll rsee you. In a minute. Thank you. Alright, bye by,

By using a first closing turn in this way speakers may not only initiate closings but also offer that a second terminal is relevant. This precludes opportunity spaces in which a movement out of closings could have been organised through by-passing two possible turns at talk; a first speaker's second turn, and a second speaker's first turn where a second close component could have been produced in the archetype closing. Foreshortening in a first turn can mark the inappropriateness of moving out of closings by providing for the immediate placement of a second terminal. A next turn can produce the 'looked for' second terminal. (46) [JG:IV:3:3-4] *—> Ronald: Marge:

Okay guhbye DuVal, Alright honey bye

(47) [NB:IV:5:4] *—> Emma: *-» Martha:

Alright, bye by,e Bye



For some cases of foreshortening in a first turn, two further observations may be made. First, as in example (46) a second terminal may be produced, but is preceded by a close component in the turn initial position. This is the component that could have exclusively occupied a second turn but foreshortening had not been offered in a first turn. By constructing the next turn in this manner recipients of a foreshortened first closing can produce the sought after terminal whilst also preserving closings as mutually produced. They respond to eacH of the closing components in the prior turn, and display a sensitivity to the prior speaker's push for terminating whilst displaying that to be mutual. Where a second terminal is not preceded by a close component in turn initial position, termination may be accomplished but that closing may appear particularly brusque. A second observation relates to the brusqueness of foreshortening. In a first turn it can appear to be brusque because first speakers preclude recipients from taking what would have been a legitimate turn in which to reciprocate an orientation to closure. Second speakers may reciprocate in a manner that re-establishes mutuality, but, nevertheless, closings have been brusquely initiated. Having secured a second terminal, first speakers may attempt to soften the brusqueness of closings by producing a further terminal before, in the case of telephone conversations, the connection is broken. Brusqueness is softened through lengthening closings by a further turn, but in a manner that does not provide the opportunity for a movement out of closings. (48) [JG:III:14:1] Maggie: Marge: *—> Maggie:

^'hh Alright. G'bye Okay rbye Bye . . . . end c a l l . . . .

Foreshortening may also be accomplished in a second closing turn. This can involve the early placement of a first terminal that was encountered above. In the examples below, a close, and a terminal component, occur in the second turn. The close component is in the turn initial position, and following turns produces second terminals. (49) [SBL:3:2:5] Chloe: Claire:

Well listen, we-we'll see when we get home. Okay,



Alright. Bye f bye Bye . . . . end c a l l . . . .

(50) [TCl(c): 12:23] Linda: Joan: Linda: * Joan: *—> Linda:

Have a nice weeke: nd = = Yeh you too, Okay, = = Okay buh bye, = = Buh Bye, . . . . end c a l l . . . .

(51) [TG:28] Bee: Ava: *—> Bee: *-» Ava:

"tch I'll t—I'll talk tihyou then t'mor.row. 0:kay,: = Okay buh f bye, Bye bye. . . . . end c a l l . . . .

The relevance of the second terminal production is proposed by a second speaker with the production of the first terminal. The participant who produces the first terminal has the status of a first speaker in the archetype closing. Second speaker can, however, produce the first terminal by foreshortening closings in a second turn. Again, foreshortening can forestall an opportunity space in which a movement out of closings might take place and both advance and attempt to secure termination, by displaying the inappropriateness of moving out of closings. Foreshortening in either a first or second turn can be used to both advance and attempt to secure termination. Opportunity spaces in which a movement out of closings can take place, are precluded and hence sequence types that move out of closings, and will later be seen to be fitted for the opportunity spaces they predominantly occupy, may not be introduced. Foreshortening may then be a strong method for providing for termination in that it may not provide the opportunity to move out of closings in the manner described for the archetype closing. Extended Closings In contrast to foreshortened closings, closings may be extended by the production of a further close component in a third turn that displaces a first terminal component.



(52) [Krakowski: D + R: 15 ] David: *—> Robin: *—> David: - * Robin:

Okay good. Good. (1.4) Awright then. (0.2) Okay. A'right

(53) [FD:I:66:R] Jack: *-> Terry: *—> Jack: —» Terry:

See y' later, Okay. Okay, Okay, hhh

(25) [JG:I:24:14] Marge: (*)—> Laura: * Marge: Laura:

or et least write a note, Alrighty, Okayho,ney, r , Alrighty honey

(54) [SBL:2:2:3:48] Claire: Chloe: Claire: Chloe:

I'll write it down but I might d-ah be late an' dash out the door without' em. Yeah, okay. Okay honey. Alright,

T h e extension turn may displace a first terminal but it can still preserve the relevance of first terminal production. In this case the extension delays its occurrence. In the following example the extension is followed by a first terminal in the same turn. (52) [Krakowski: D + R-.15] Robin: David: —> Robin: David:

Awright then (0.2) Okay. A'right bye bye. Bye bye. . . . . end call. . . .



Unlike other items that have been seen to be positioned in a third turn, extensions do not exploit an opportunity to move out of closings, nor do they project the relevance of closing reinitiation. They are both closing implicative and terminal elicitive. But by delaying the production of a first terminal they stand in sharp contrast to the brusqueness of foreshortened closings. Whereas foreshortened closings advance termination, closing extensions can delay termination. An extension of closings in the third position displays further characteristics. Although extension components do not, themselves, exploit an opportunity to move out of closings they can achieve a sequential environment that is needed for such an occurrence. First, they can produce an additional opportunity space in which a movement out of closings can take place and simply increase the 'chances' that a movement will take place. Second, they may display a 'reluctance' to terminate. That is, an extension can be used to both provide an opportunity, and to signal the appropriateness of a next speaker organising a movement out of closings. Both being done without the speaker actually making any movement out of closings. Thus, in the following, Terry's extension in third turn is followed by Jack who initiates a movement out of closing, with an arrangement offer. (53) [FD:I:66:R] Terry: Jack: Terry: —> Jack:

Okay. Okay, Okay, hhrh See yuh coffeetime,

An extension does not elicit a topic initial but it does provide for continued talk without 'contributing' to that talk. Recipients of a closing extension may also 'go along with' a prior speaker's 'reluctance' to terminate but without moving out of closings. This can be done by reciprocating an extension in a third turn with another extension. Thus: (25) [JG: 1:24:14] Laura: Marge: Laura: —» Marge:

Alrighty, Okay horney, Alrighty ,honey,j Okay


140 (54) [SBL:2:3:48] Chloe: Claire: Chloe: —> Claire:

Yeah, Okay. Okay honey. Alright, A'right,

It is possible to conceive of an indeterminate string of closing extensions turns, however, the data corpus does not disclose such an occurrence. This can be found to be a systematic consequence of closing extensions produced by both speakers. In the last two examples it would seem that each speaker having delayed termination, but without actually introducing further inconversation items, may find that no further items are being introduced and that termination may be appropriate. In the first example the terminal is produced within the same turn, and in the second it is produced simultaneously with a terminal in next turn. (57) [JG:I:24:14] Laura: Marge: Laura: —> Marge:

Alrighty, Okay ho,ney, Alrighty ,honey, Okay Bye: . . .

(58) [SBL:2:3:48] Chloe: Claire: Chloe: Claire: Chloe:

Yeah, okay. Okay, honey, Alright, A'right, b,ye. Bye bye. . . . . end c a l l . . . .

The fact that participants may find that no further items are being introduced is a sequential matter. That is to say, both participants in consecutive turns display that they are not going to make any movement out of closings. It is not just the absence of 'something to say', for simply, a movement out of closings could be produced, as in example (56) by just offering an arrangement item. In contrast to foreshortened closings, extended closings, may be produced that create further opportunity spaces in which a movement out of closings can take place. 'Sequence types' may be used in these spaces to move out of closings, and this variant on an archetype closing increases the 'chances' that for 'this' conversation a movement out of closings will take place. Thus foreshortening and extended closings can be variants on arche-



type closings that change the distribution of opportunity spaces and can have attendant consequences upon the placement of 'sequence types' in closings. For foreshortened closings, opportunity spaces are restricted and 'sequence types' that could be placed in an archetype closing may not be produced, and for extended closings, further opportunity spaces are provided for 'sequence types' to be produced that move out of closings and enhance the 'chances' that movement will occur.

Initiating a 'Sequence Type' in an Opportunity Space In the second section it was seen that a number of 'sequence types' may be used in closings that organise a movement out of closings, and in the third section, the archetype closings was examined in order to discover the places within a closing where a movement out could be systematically organised. In this section the 'sequence types' will be seen to be commonly used in particular opportunity spaces provided by a closing section. If a corpus of a particular 'sequence type' is examined it is possible to observe that the 'sequence type' is used in a variety of opportunity spaces, by either first or second speaker. However, each corpus of 'sequence types' reveals that a particular 'sequence type' is overwhelmingly used in a particular opportunity space. For example, out of 45 cases of arrangements in closings, 31 were found to occur in the first opportunity space, five in the second, four in the third and five in the fourth opportunity space for a closing. 19 Each 'sequence type' displays a particular distribution in closings, and the 'sequence types' can now be re-examined in order to see how their production is designed for the particular opportunity space they are predominantly used in.

Back-references Back-references mainly occur in the second opportunity space. This is the next turn to a first close component and is second speaker's first opportunity to move out of closings. (15) [SBL: 1:11:4] *

Tess: -> Bea: Tess:

Tomorrow ni-tomorrow night at six Yeah, Okay Uh huh, and I'm sorry I didn't get Margaret, I really ve been wanting to



Back-references make a drastic movement out of closings by retopicalising prior material. The second turn is second speaker's first opportunity to collaborate or not in closings, and is thus a crucial opportunity space for declining to enter into the collaborative production of closings. This turn can then be used to continue conversation, and in as much as back-referencing makes a drastic movement out of closings and is oriented to conversation continuation it is fitted for use in a second speaker's first opportunity space. Out of the three 'sequence types' described here as associated with a drastic movement out of closings, back-references are the most prevalent and would seem to be the prototypical way of declining to enter closings. The other 'sequence types' initial turns both indicate availability without offering material for talk to. A back-reference may, then, be a strong attempt to drastically move out of closing; it specifically introduces material for next turns to topically develop. 'In-conversation' Objects 'In-conversation' objects also cluster in the first opportunity space for recipients of closing initials, a closings second opportunity space. (23) [TCl(b)13:5] Linda: Jerry: *—> Linda: —> Jerry:

I' Fi: ne en you ,Okay, Okay hon, hh 0 , k a y ' h h Uh::::m,

This opportunity space is, as was noted above, a first place where the second speaker may decline to collaborate in closings. 'In-conversation' objects have been observed to initiate a drastic movement out of closings and display an orientation to continued conversation, and are not oriented to closings. Thus similar to back-references they are fitted for a second opportunity space where a second speaker can produce them and display that closings are not being collaborated in. Topic Initial Elicitors Topic initial elicitors occur most frequently in the closing section's third turn. This is the fourth opportunity space, and first speaker's second opportunity for moving out of closings. (20) [Frankel:QC:2:l:10] Ma: Kiddo:

Uhright? Yeah,

MOVING OUT OF CLOSINGS *—> Ma: Kiddo: —»• Ma:


pt. Okay. = =Uri:ght. = = What else,

It has been seen that in order to make a drastic movement of closings, topic initial elicitors require recipients to produce a topic initial or newsworthy item. It is in this respect that they are fitted for a fourth opportunity space. This can be seen by noting that a fourth opportunity space is the third turn in the archetype closing that could be used to produce a first terminal. There have been previous opportunity spaces for both speakers to move out of closings but in not moving, both speakers have displayed an orientation to closure. Topic initial elicitors display a sensitivity to the offered and accepted possibility of now offering a first terminal. Although they can occasion a drastic movement out of closing, they require the co-operation of both participants (both of whom have displayed a previous orientation to closings) in order for a drastic movement to be accomplished. This contrasts with a back-reference, which makes a drastic movement out of closings and is used by the second speaker who has not yet displayed an orientation to closure. Thus topic initial elicitors are fitted for a third turn: both speakers have displayed an orientation to closure and, as an organisational feature of topic initial elicitors, it now requires both speakers to collaborate with one another in generating a topic in order for a drastic movement out of closings to be made. Arrangements Arrangements massively cluster in first turn. They exploit the first opportunity space post a first close component in that turn's transition space and are used by first speaker. (12) [SBL:2:2:3:46] Chloe: Well, it was fun C l a i r e Claire: Yeah. I e n j o y e d every minute of rit Chloe: And yeah. (*)—> Claire: Okay, well then we'll see you Saturday. Arrangements have been seen to be closing implicative but not terminal elicitive. That is, they move out of closings but this movement is minimal and provides for the reinitiation of closings following a minimal return to the arrangement. This feature of arrangements in closings is sensitive to their use in a first opportunity space. Arrangements can, in effect, be used to project a longer-than-archetype trajectory for closings (Button, forthcoming). This can have two interactional advantages for first speaker.



First, a first closing turn is first speaker's entry into closings, but second speaker may move out of closings in a second turn and not orient to closure. Thus, first speaker does not know, in advance of its accomplishment, that termination will ensue. Using a longer-than-archetype closing may take more turns to produce termination than an archetype closing but it does frequently produce termination. By entering into closings, but immediately moving out, though using closing implicative material, first speaker can secure a closing implicative turn from second speaker by providing for the second speaker to produce a minimal response to the arrangement introduction. This response has been seen to be also closing implicative. Thus, a first closing turn in the archetype closing has no way of providing for the next turn to be oriented to closure over continuation, but by minimally moving out of closings with arrangement material a first speaker can provide for a response from next speaker which although it follows a minimal movement out, is indeed, closing implicative. Thus, first speakers may be able to reinitiate closings following this turn, which places second speakers in a position whereby they have implicated themselves in closings. This is a position a second speaker would not be placed in by the archetype closing. Thus, projecting a longer-than-archetype trajectory, in the first opportunity space, may take more turns to reach termination than the archetype closing, but it can be a way of eliciting an orientation from second speaker to closings, and to attempt to provide for actual termination. Arrangement intrusion in closings is consequently designed for a closings first opportunity space. Data reveals that arrangements are by far the most predominant 'sequence type' to appear in closings, and it would seem that they are a prototypical way to actually initiate a closing sequence. Solicitudes Solicitudes also cluster in the first opportunity space. (54) [TCl(b):7:3] Lily: Cora: (*)-» Lily:

'hh Sa:m says how much he li:kes you:, Yeah = =So:, 0 : i k a y , ' h h well gee you go back tihbed'n take good care a'yourse: If.

Like arrangements, solicitudes are also closing implicative. They can be used by first speaker in much the same way as arrangements, in order to elicit from second speaker a displayed commitment to closings. However, they are not as predominantly used as arrangements. They tend to be used



when it is not possible for 'this' conversation to use arrangement items. They can project a longer-than-archetype closing, but, as was noted when they were examined, although this projection may be preserved, solicitudes may receive turns that move away from the projected sequence. Closings in which solicitudes are placed do not, as frequently as arrangements, result in a conversations termination. Reason-for-calls Reason-for-calls are reiterated and referenced predominantly in the first opportunity space. (60) [Erhardt:8:3]


Vicky: Pam: Pam:

I know that's lousy for you, = = n N o that ken you sa:y. Well, okay.IJis wunduh kno:w. I-(yeh) I wz j s calling you en finding out.

To expand on this observation it can be noted that a first closing turn is not a freely occurring position in conversation but is organised with respect to a closing implicative environment. However, using these environments to offer a 'free turn' and thus initiate closings, is a locally determined activity. In the following example, Nancy, having confirmed an arrangement continues in conversation by 'extending' her turn as opposed to allowing Hyla to begin a turn in a position that, due to the closing implicative character of arrangements, might be 'ripe' for initiating closings. (61) [HG:13] Nancy: Hyla: Nancy: —> Nancy: Hyla:

Oh: wo:: ow, so it starts et eight thirty? hh Yeah. So, u f j - k - p i c k you up li:ke by eight uh 'clo:,ck, Yeah = = She said -(0.2) well cuz I'm so clojse too = = Yeah

So, though a closing could be initiated following a closing implicative turn, actual initiation is locally decided. Initiating closings, but immediately moving out of closings by reiterating the reason-for-call, can be a way that first speaker who inititates closings can 'soften' an initial move to enter closings, and 'account' for closings occurring 'now'. But because 'reason-



for-calls' are closing implicative, the first speaker is able to preserve the relevancy of closings. Thus, 'reason-for-call' reiteration can be used in a first opportunity space to 'soften' and 'account' for an initial entry into closings and preserve the relevancy of termination. Appreciations Appreciations display two distribution patterns. Appreciations which locate the call as the object of appreciation occur in the first opportunity space, and appreciations which locate a particular appreciable in the conversation occur in the first two opportunity spaces. (29) [SF:1:13:C] Mick: (*)—• Joanne:

Ah'll talk tuh yuh later Okay Spark thanks for calling

(62) [Kamunsky:ll:7] Alan: (*)—» Shawn:

'hh well I av s'm homework t'do I thought I'd js stop 'n call you, A'right th ank you.

(32) [JG:1:10:7ff] Marge: Sam: Sam: *—Marge: —> Sam:

hh A : n ' uh tell uh:ar li'l gi:rlfriend ar oTther little girlfrie: rn hello an' everything like that. = =Iwilldeah. 0:kay.= =Thankyou,

( )

The first form of appreciations operate in a similar way to 'reason-forcalls'. That is, appreciations that appreciate the call can be used to 'soften' an initial entry into closings. Thus, they are used by first speaker in the first opportunity space. The second form of appreciations which can appreciate some 'favour'; for example, that is locatable in the conversation may be placed with respect to which speaker initiates closings, and which speaker can produce an appreciation. That is closings are initiated independently of activities such as 'favours', so that the speaker who initiates closings may be either a 'favour' recipient, or a 'favour' doer. If, then, appreciations are to be done in a closing, the appreciator could either be first or second speaker, and hence this form of appreciation may occur in first or second speaker's first opportunity spaces. Further, appreciations are placed in the first opportunity space available to a person who does them. This is because apprecia-



tions can o p e r a t e as though they were close components, and may thus also o p e r a t e as closing extensions. But, as was noted when closing extensions w e r e e x a m i n e d , this can provide further opportunities to move out of closings. Appreciations, in as much as they can operate as though they were close c o m p o n e n t s may then extend closings if used in other positions to a first and second opportunity space, and their use, whilst actually appreciating m a y increase the possibility that a movement out of closings may take place. T h e i r use in the first available opportunity spaces to accomplish appreciation may then accomplish the activity of appreciation without also making a closing m o r e vulnerable to movements out of closings. So, 'sequence types' can be found to be designed for the opportunity spaces they predominantly occupy. In this respect it is possible to draw some conclusions concerning the organisation of movements out of closings in general. First, drastic movements are mainly made by second speaker, and minimal m o v e m e n t s by first speaker. A minimal movement may be used by t h e first speaker in the interests of attempting to elicit closing implicative next turns f r o m a second speaker, or for 'softening' and 'accounting' for closing inititation. Drastic movements may be used by a second speaker and d o not display an orientation to closings. Thus the degree to which a m o v e m e n t out of closing is made may be related to an issue of who amongst first and second speakers, makes the movement. Second, some opportunity spaces are exploited to move out of closings m o r e than others. Opportunity spaces which follow first and second close c o m p o n e n t s are overwhelmingly used to make a movement out of closings, whilst an opportunity space following a first terminal is used with less regularity. This can be accounted for in a number of ways. First, there are m o r e opportunity spaces for moving out of closings following close comp o n e n t s than following terminal components. Second, opportunity spaces following close components precede a terminal component and its opportunity space, and hence, any movement out of closings may be 'caught' b e f o r e a terminal component is reached. Third, the sequence types have b e e n observed to be used, and fitted for, particular opportunity spaces, and all of these are spaces following close components. Cases of movements out of closings following a first terminal are the result of a 'mis-placed' terminal, or are cases w h e r e the 'sequence type' design is less sensitive to its position t h a n the places it predominantly occurs in. This can give a flavour to some m o v e m e n t s out of closings as being 'forced' by one speaker. Fourth, it is with r e g a r d to first and second close components that a warrant for termination is p r e d o m i n a n t l y organised, since the constraint placed upon next turns by



prior close components is not as strong as the constraint that a first terminal places on a next turn. A first terminal provides specifically for the production of a second terminal. This stronger relationship between first and second terminal can account for second terminal following first terminal with greater regularity than second terminal being used to move out of closings.

Conclusion It can be concluded: (1) there are various 'sequence types' that may be used in an archetype closing. (2) When they are used a movement out of closings is made. (3) There are two categories of movement, a drastic and minimal movement. (4) Closings are designed with opportunity spaces in which a movement out of closing can be made in the interests of preserving the mutuality of closings. (5) Particular 'sequence types' are fitted for the opportunity spaces they predominantly occupy. (6) 'Sequence types' that make a drastic movement out of closings tend to be used by a second speaker, and 'sequence types' that make a minimal movement by first speaker. Movements out of closings predominantly follow close, rather than terminal components. Given the above, closings appear to be a particularly delicate section of conversation and are very sensitive to conversationalists' orientations to conversation continuation or conversation closure. They are constituted from out of speakers' 'negotiations' with one another for continuation or closure. They, thus, appear to be used not only to organise termination but also to determine how speakers will 'leave' one another. In this respect they may be used to characterise the relational states speakers may have achieved in their talk. Further, in as much as conversations can be in a series, how speakers leave one another may be a resource for their further interactions and may bear upon their initiation, or the immediacy of a 'next contact'. These are matters that require further elaboration, and having, in this chapter, established how movements out of closings may be organised around certain 'sequence types', the use of these 'sequence types', singularly or in various combinations, will be described in order to display the variety of forms that closings can take that elaborate on the archetype form that has been examined here, and the variety of relational states speakers may propose in closings (See Button, forthcoming).



Notes to Chapter 5 1. This Chapter develops and revises some aspects of 'No-close closings' presented at the International Conference on Practical Reasoning and Discourse Processes, St. Hughes College, Oxford, 1979.1 am in debt to Gail Jefferson for providing a wealth of data and for a detailed reading of an initial version of the analysis. John Lee made valuable recommendations for which I am very grateful. 2. There are a number of closing implicative environments in conversation. Topic bounding (where a topic is 'shut down') is one of them. Others are: prior turn repetition, prior material reiteration, arrangements, aphoristic conclusions that draw a point, and activities such as 'reason-for-call' reiteration which may propose closing initiation. It is beyond the scope of this chapter to undertake a detailed consideration of these sequential environments. For a more thorough examination see Button, 1985. Consequently, when fragments of conversation are introduced in the text, an account will not be given of the description: 'a closing initial turn'. All relevant fragments can, however, be accounted for as being initiated by a closing initial turn in terms of their occurrence and operation in the above forms of closing implicative environments. 3. It is also found that a first closing turn often uses 'address terms'. See, Jefferson, 1973. 4. There are other forms of closings (see Button, forthcoming). This chapter will also briefly introduce 'foreshortened' and 'extended' closings. 5. Cases where termination is attempted in one turn can be found. However, this can be oriented to by conversationalists as illegitimate because a warrant for a termination has not been mutually established. Thus, in the following example, Maggie's first terminal has not been mutually and collaboratively provided for in the manner described above, and the next speaker orients to its unwarranted production. (63) [JG:I:9] Ronald: Maggie: Ronald:

hhhhhh What-waddiyou want fer dinner I won'-I::, jist I'll take care of myself you do the s a m e . G.oodbye Whaddiyou mea:n

6. Sacks (this volume, Chapter 2) describes how pairs of utterances go together and form a sequence of conversation. The 'sequence types' that will be described here seem to be less powerful than the adjacency pair format Sacks refers to. However, it will be found that upon the occurrence of the types of utterances described, a next turn is constrained in more or less projected ways to be occupied with some particular material. This may seem strange for some of the items where should they occur in other places in conversation to closings, they may not necessarily project relevant next activities. However, it will be seen that in closings they can initiate sequences of activities and this may, in part, be due to the delicate nature of closings with respect to the close monitoring of participants' utterances that can be observed. Nevertheless, in recognition of the fact that some of those 'sequence types' may not form sequences of conversation elsewhere, the term is placed in inverted commas. 7. This example includes a closing extension turn, the operation of which will be elaborated on later. 8. Back-references might be seen to include 'arrangement reiteration' and two



other 'sequence types' to be examined, 'reasons-for-calls' and 'appreciations'. However, these 'sequence types' are always oriented to the same particular objects, whereas, the objects described here as back-references can implicate various material. Also, as will be seen the other 'sequence types' project a different degree of movement out of closings to the degree of movement projected by back-references. Thus the term 'back-reference' is being used to describe the re-topicalisation of previous topical material which is not closing implicative in character. 9. See Button & Casey, 1984, for a detailed description of the operation of topic initial elicitors. The remarks here formed the basis for that consideration and, hence, although they are not as elaborated as they subsequently became, they are retained at some length. 10. Notice the similarity and difference with example (17): (17) [HG:15] Nancy: Hyla: Hyla:

Anything else to report (0.3) Uh:::::m:::, (0.4) Getting my hair cut tihmorrow:

H e r e Hyla displays a 'search for something to say'. But an orientation to Nancy's topic initial elicitor does not result in Hyla introducing something on Nancy's behalf, but on Hyla's own behalf, even though it is a response to prior display of availability which it can be noticed also did not provide material to talk to. 11. This observation does not have any consequences for the operation of appreciation as examined in this section. However, it does become consequential for the positions in closings occupied by the two forms of appreciations which is examined in a later section. 12. It must be remembered that appreciations can be an independent 'sequence type' but that returned appreciations are not strongly projected by a prior turn as are some activities for next turns which are projected by current turns in other 'sequence types'. 13. It can be noted that all minimal movements out of closings can eventuate in drastic movements, and that drastic movements can be curtailed and closings reinitiated. The descriptions that have been provided encompass the ways in which an utterance may project some form of constraint for a next turn and how that turn may display that it has been produced under the auspices of that projected constraint. However, it is humans that are being dealt with, not machines, and the descriptions are not descriptions of programmes that run the machine. The descriptions are intended to touch the oriented-to methods involved in constructing some course of interaction. Whilst these may be universally usable — for example, irrespective of the particular persons talking — they are contextually achieved for a particular occasion. Thus, although one speaker may introduce some item and provide for the reinitiation of closings, nevertheless, that speaker has introduced some material that a next speaker may topicalise and thus attempt to undercut its closing implicative projection. And, one speaker having drastically moved out of closings, a next speaker may curtail that movement and attempt to provide for conversation closure. Interestingly, both of these activities may be followed by further varieties of activities oriented to closure or continuation. These developments cannot be described here (though



they depend upon the methods being detailed here) and will be given full attention elsewhere (Button, forthcoming). 14. It is possible to find that talk can continue past a second terminal. H o w e v e r , a second terminal marks the end of a closing section and provides f o r no further talk. Should further talk be produced its inclusion in 'this' conversation may be problematic — a speaker may be left talking to a dead connection if speaking on the telephone, or have to 'summons-back' a co-participant in face-to-face interaction. T h e organisation of further talk past a second terminal may thus be considered separately f r o m movements out of closings and will not figure in this chapter. 15. In all 401 closings w e r e examined. With respect to the issue of the distribution and frequency of movements out of closing, see note 19 below. 16. See Schegloff, Jefferson & Sacks (1977) f o r a description of opportunity sources with respect to self and other correction. 17. S c h e g l o f f , Jefferson & Sacks (1977) describe transition space in the following way: T h e transition space, roughly, is the environment of a turn's possible completion, at which possible transition to a next speaker becomes relevant. A l t h o u g h the transition space may begin a bit before the possible completion point, and last a bit into the beginning of next turn. . . if may be thought of as the 'beat' that potentially follows the possible completion point of a turn (Fn. 12, p. 366). 18. It was noted in the introduction that terminals fall under the auspices of the adjacency pair mechanism. This is a particularly powerful mechanism for organising the relationship of adjacent utterances. H o w e v e r , a first terminal appears to a c c o m m o d a t e activities other than a projected next activity in the interests of preserving the mutuality of closings. Movements out of closings following a first terminal are though 'scarce'. This is partly the product of the adjacency pair relationship, but their occurrence is the result of a continual mutual coordination of closings. 19. Such statistics are mentioned with uneasiness because it might be taken that these are the findings, and then questions concerning significance and the randomness o f the data would have to be considered. It should be understood that mere statistical occurrence is not the finding nor, in itself, very interesting. Distribution and frequency is merely, here, a device used to become sensitive to the organisational design of the 'sequence types' examined. That is to say, having noticed that some 'sequence types' were often found to occur in the same opportunity spaces it became possible to see that they displayed a sensitivity to those spaces; w e r e designed to g o there. W h e n they were used in other opportunity spaces to those in which they regularly occurred, they displayed a sensitivity to those sources which also preserved a sensitivity to the spaces they predominantly occurred in (see Button, forthcoming, for further elaboration). T h e statistic, then, is a by-product that caught the eye and ear.

6 Notes on laughter in the pursuit of intimacy1 G A I L JEFFERSON University of York H A R V E Y SACKS Late of University of California, Irvine E M A N U E L SCHEGLOFF University of California, Los Angeles

'Laughter does not seem to be a sin, but it leads to sin.' St John Chrysostom

Laughter as a Socially Organised Activity This chapter considers laughter as a systematically produced, socially organised activity. As a first approach, laughter can be treated as a nonspeech sound among others produced by co-participants to a conversation, vocally or in other ways; sounds which may or may not show up in a transcript, and, if they show up, tend to be described rather than transcribed. Many of these can be profitably transcribed; that is, a transcription of non-speech sounds can make available systematic features of the sound's production, and can permit observation of how the talk might accommodate the occurrence of such sounds. So, for example, in a conversation in which one participant is hammering, it appears that the hammering is produced as a pulsed burst, tending to start with a light tap and end with a wallop, and a co-participant appears to be monitoring for completion of a burst of hammering before producing talk directed to an utterance which preceded the burst. 152




1 Joe:

B u t they never called us back.

2 3 Mike: 4

C) ekhheh = = ((buk b a n g b a n g bang bang bang bang bang))



6 Vic: 7 Mike:

Who. = = Mayer.

(2) [Frankel :US: II: 17] 1 Joe: 2 3 4 Joe: 5

E n dih, guy with d u h - e h (0.2) ((buh)) e a : r d r u m : k e m i:n. = = ( ( b a n g b a n g bang B A N G B A N G ) ) =

6 Mike:

= G u y wit d e e e a r d r u m ?

About these fragments it can also be noticed that the hammering appears to be fitted to the talk. For example, in Fragment 1 the hammering starts up after a recipient of hammerer's utterance (line 1) has produced a form of comment (a laugh, line 3), and in Fragment 2 a soft tap accompanies the searched-for and found word 'eardrum', a burst of hammering starts up on completion of hammerer's utterance, and, once recipient's talk is underway the hammering does not start up in its course. In the following fragment, someone starts to cough while another is talking (lines 17-24). The cough occurs in three pulsed bursts, the first two perhaps marked as components of an uncompleted cough with an inbreath (' hhh', lines 18 and 20), which can be heard as preparatory to a next burst, the third perhaps marked as completed with the final particle 'hn' (line 22). Speaker appears to be monitoring, correctly, for completion of the serially produced cough. (3) [ F r i e d e l l : A l t : 9 ] 1 Hank: 2 3 Sheila: 4 5 6 7 8 9 10 11 12 13

Sheila: Sheila: Hank: Hank:

Hank: Sheila:

e - H e went t o : one mixer, et some, ' (1.3) A r e rmxers et girl'schools better th(h)'n mixers e t (•) yihknow. Ri:ce'n, (.) o h : , g r a : d mixer, ::,s a ( h ) n : , l 'Definitely. Definitel y. = = Mixers et girl'schools usually have a b o u t : :t hh en ev'n ratio. (0.7) ( ) c . T h e r e ' r of.ten m o r e girls'n boys. l R e a lly? 1


154 14 15 16 17 18 19 20

Sheila: Hank: Sheila: Sheila:

21 22 Sheila: 23 24 Hank: 25

(0.9) W h a t a s a : d state fer th'gi: rls. (2.0) E n d he went to one et some, jCatholic girl' jschool. u k h h - h n h ekhhh hhh K e h h - h u h kuhh! hhh (0.3) E K K H H - h u u h , ikhh-hn now knows four'r five girls including one eez probly going tuh take out.

As in Fragments 1 and 2, not only does the talk accommodate the nonspeech sounds, but the cough seems attentive to the talk. For example, the cough starts at a point in a current utterance (line 17) at which, in a prior version of that utterance there had been a pause (line 1 and 2). And, for example, the cough may be produced to be as unobtrusive as possible, each burst increasing in force when a prior, less obtrusive cough was ineffective. Finally, it appears that cougher is attentive to speaker's attention to the cough series' uncompleted status. That is, the second burst follows the first after a silence of less than two-tenths of a second (indicated by the period in parentheses, line 19). Given that the second burst occurs in the clear, cougher may take it that speaker has understood the inbreath as preparatory to a next burst and has relinquished the floor. Thereafter cougher takes (and speaker permits) a longer pause before the next, projected-by-an-inbreath burst (lines 20-22). And in the following fragment, while a speaker and a laugher appear to be pursuing their own activities independently of, and perhaps competitively with, each other's, each pursues his activity with an orientation to the other's (lines 13-24). (4) 1 2 3 4 5 6 7 8 9 10 11




A n d most a'the people who write about it are m o r e - are y'know straight- sorts'v people w h o - a r n ' t - t o o : , (0.3) y'know,

Nora: Ed: Ed:

(0.4) hh W e : 11 w h o - a r e (0.7) not used to, (0.2)

LAUGHTER IN PURSUIT OF INTIMACY 12 13 14 15 16 17 18 19 20 21


Ed: Nora: Ed: Nora: Ed: Nora: Ed: Nora: Ed: Nora:

= ha ha ha = = N o w't I'm .saying, is .t h e t hhhh heh-hn. .Well .like t h i - , h e h , h e h , h e n heh h e h , = ,Oneri:l, , h h h e : : :h

Nora: Ed: Nora: Ed:

h e h heh = = y ' k n o w " I .discovered love through L S D " = hhheihhhhhhh = en this wz a fsuccessful businessman en,


y ' k n o w things were going alright b ' t . . .

p r o f o u n d things I guess er s o m e thino In a n y



23 24 25 26 27 28


Like the hammering and coughing, the laughter occurs as a pulsed burst. In this case, it appears that speaker is attentive to sub-units of pulses within the bursts, placing his talk with an orientation to those units as possibly completed laugh-bursts. For example, at two points in the transcript speaker starts to talk just as laugher stops (lines 15-16 and 23-24); that is, speaker starts talking in the clear. At lines 15-16, while the total burst may be seen as 'hhmhhhhh hih heh ha, ha ha ha', it may also be subdivided into three discrete units: ' h h m h h h h h ' + ' h i h heh ha', + 'ha ha ha', the onset of the second unit marked by an increase in amplitude and raise in pitch, unitcompletion marked with the intonation indicated by the comma. Speaker may be using features of the second unit to find a completion point in the third; that is, both have three pulses or particles. A similar sort of attention may be operating at lines 19-24. Laugher has produced a two-particle sub-unit after speaker has stopped (line 19, 'heh heh,'), and speaker may be monitoring for completion via two-particle units by reference to which he positions his talk (lines 20 and 24). And at one point in the transcript, laugher appears to be attentive to the talk. That is, having produced what might be a laugh-terminal inbreath in overlap with an utterance-beginning (lines 20-21), 2 finding that speaker has stopped such that a silence is occurring (line 22), laugher produces a next unit of laughter (line 23), and thereafter produces what turns out to be a laugh-terminal inbreath (line 25). The foregoing considerations have treated laughter as one among various sorts of non-speech sounds such as hammering and coughing, which occur during, and might constitute possible disruptions of, ongoing talk. Observation of detailed transcripts suggest that the sounds have systematic productional features which can be used by speakers, who can accommodate



their talk to the occurrence of such sounds in orderly ways. There are also indications that the sounds are produced with some attention to the talk which they may be disrupting. That is, some of the productional features of the non-speech sounds are interactionally based. However, laughter may be distinguished from other non-speech sounds in that it has, for participants, the status of an official conversational activity. It can be a relevant, consequential next action to some prior (see Jefferson, 1979). And it can be named as a response given to a quoted utterance in a report of a prior conversation (as, for example, someone can report 'I said X and he agreed,' someone can report 'I said X and he laughed.'). For example: (5) [Labov:BG:] 1 Alice: 2 3 ( ): 4 5 Alice:

I says I w'like to have some ni: :ce, fresh, pardon the expression horseshit. hhmh! huh ^huh huh huh, hhhj hhOh my goo'ness. (•) ~~ Well they die : d laughin.

(6) [SBL:2:2:3:] 1 Chloe: 2 3 4 5 6

So I said wz wondering did yuh

(•) Chloe: bust w.'n ri-.ght . 'n the (h)no(h).se, eh! hhhhhh . Claire: u7hnhhh' ' hnhh h n h - h n h ] Claire:_ ,,'h n h . Chloe: ~~ En she 1: laughed

Thus, a speaker who is talking in orderly ways by reference to laughter is perhaps best seen as co-ordinating his conversational activities with another relevant conversational activity, rather than accommodating merely disruptive non-speech sounds. And it is empirically observable that in general, the transition from talk to laughter to talk is done in an orderly fashion. The following fragments display a range of orderly transitions, and, as an extra fillip, indicate that not only can speech be co-ordinated with laughter, but that multiple laughers co-ordinate their laughter, producing coherent, monitorable units. In Fragment 7, the transition is utterly clean, the laugh burst consisting of synchronous simultaneous onset particles ('ehh' and 'nkh'), a simultaneously begun and co-terminous pulsed burst, each with stressed initial particles ('heh' and 'hih'), and terminal inspiration (''hhhhh' and l 'nh nh'), immediately after which, a next speaker starts up in the clear.




1 Al: 2 Bev: 3 Bev:

W u h y i h wan'talk a b o u t . . n k h . h i h - h u n h , h u n h h u n h , . nh n h . " ehh'heh-heh-huh-ehh-huh.'hhhhh1 = H o w wz yer day,

In Fragment 8 there are simultaneous onset particles ('e' and 'ih') and coterminous pulsed bursts. In this case, one party produces a terminal inbreath, and almost simultaneously, both start to speak; i.e. in this case each attempts to occupy the projected post-laugh position with speech. (8) [ F r a n k e l : G S : J P : 11 :r| 1 2 3 4 5

Sara: H e rilly d i d n ' h a v e too much'v a p r o : d u c ' t u h s h o : :w. = Sara: _ . , e - h u h h u h h e h heh huh huh, Mary: i h h h u h h h h e h h a h h a : h a h 1 hihhh = Sara: =C|'ny'see-, Mary: W e hed a guy in the airplane the o t h e r d a y . . .

In Fragment 9 the initial start on speech (line 6) may be produced by reference to the two-part sub-units of the ongoing laughter (line 3 'hhhm h h h m ' , line 4 'mih-hih', line 5 'haa-haa' + 'haa-haa', line 6 'eh-heh!'), and after dropping out by reference to the fact that two others are still laughing, a next start on speech (line 7) occurs at a next two-part sub-unit completion (line 4 'hee-hee' and line 5 'a-ha::'). Further, it is perhaps by reference to the initial attempt to speak that it turns out that the two laughers have stopped, the next start on speech occurring in the clear. (9) 1 2 3 4


Delia: Bart: Sandra: Bart:

5 Jill: 6 Delia: 7 Delia:

H e r e ' e c o m e s , h e r e ' e comes be .quiet, H e r e ' e comes, .be q u i e t " j hhhm hhhm = W u h d j i h say jab(h) o u t | t T i i h - h i h - j h e e - h e e | .haa-haa .haa-haa . a - h a a : : , : : . = l eh-heh! 1 Yu'll- 1 = Yu'll hear l a t e r . . .

And in fragment 10 a cohesive single burst is constructed with one laugher producing continuous pulses (Jill, lines 3 and 6) as another drops out (Sandra, line 2) and still another joins in (Art, lines 4-5). While continuing laugher produces an unbroken series of pulses, the laughter appears to be attentive to the juncture; specifically to the entry of a new laugher. That is, the two sub-units of the continuous burst are quite distinctive (line 3 'haa' vs. line 6 ' ah'), and the second sub-unit starts up just after new laugher has



joined in (lines 4-6). Such a phenomenon suggests that ongoing laughter can, for a variety of reasons be renewed and thus the occasion of laughing together can be extended. (10) [Goodwin: AD(b): 8: r) 1 2 3 4 5 6 7

Lenny: ((mock fag)) Now cut tha:tou:,tBa:::jrt, Sandra: ,m-hm-hm:eh-hahhh hh , u h h . _ Jill: a-ha-ha-ha-haa-haa- .h a a . Cal: e-heh Cal: _ ,,hehh e-heh e-huh e-huh e-huh ha, Jill: ah ah ah ah ah ah ah ah ah . 1 Lenny: Y'(de::vi:l,)

Not only is laughter produced in an orderly fashion, but it appears that an occasion of laughing together is an activity in its own right, an achievement of various methodic procedures. For example, in the following fragment, the telling of a joke is recessed for an extended laughing-together. Some details of the constructedness of the event can be observed. At lines 3 - 4 , an initial burst may be terminating; that is, one laugher has produced an intonation contour captured by a period (possible terminal intonation), the other has produced a terminal inbreath. Each thereafter produces at least one next laugh pulse (lines 5 and 6); 3 i.e. each, at a possible termination, provides for extension. A third participant simultaneously provides a different device whereby the laughing-together may be extended, a lexical reference to the prior joke segment (line 7 'Ooops!'), which, reinvoking the joke, provides official impetus for more laughter by reference to the joke itself. T h e r e a f t e r , all three laugh (lines 8-11). (11) (11) [Goodwin:AD:56:r] 1 2 3 4 5 6 7 8 9 10 11

Cal: Bart: Cal: Bart: Bart: Cal: Len: Bart: Cal: Len: Bart:

Lits heah i,t one tahm fe.r Li:::::::za! mmh, e-huh-rillyya.a::::::::y! . he:uhhe:uh hu,:ehhu:eh. , eh-hih hhh hh = jhihh jheh-he:hjhe| :h, ,

Oo::oo::jpSjSS, | he :uh|hejuh-hji J he,ha : :.ah ha-ha-ha-ha-ha-ha-a-ha. hhh.he.he,: ah!ah! ah!ah!ah!ah!ah!ah! ah'ah! Jah!a! a! âh hh

At the end of lines 9-11, only one participant is laughing (line 9). As he starts a next pulse; i.e. provides for continuation, another produces a second lexical reference to the joke (line 13), which, again, serves as impetus for



another round of laughing together. It can be noted about lines 1-16 that although participants are entering and exiting the event, there is a continuous flow of laugh pulses. 12 13 14 15 16

Cal: Bart: Len: Bart: Len:

= h e , . h h e - e h h , e - h e - , h e - , ' e - h h e e . h h l . e : : : :a: :yee: ,°ee: , [ [ Oo::^ps,' n-helu-huh eh! uh!ah!ah!ah! lo ehhhh hh° lo uhhhhh° =

At the end of lines 12, 15 and 16 there is a marked diminution of volume (indicated by the degree signs). Once again, the laughing-together is potentially terminating. At this point, joketeller provides a lexical token indicating preparation to return the joke's telling (line 17 'So:.'). Teller does not move directly into the telling's continuation. There is a brief silence (line 18) followed almost simultaneously by teller's moving into a next joke segment (line 20 'They-;) and a third reference to the prior segment (line 19 'Ooops'). Not only does teller defer to the lexical reference to the prior segment and drop out, but he joins in the next round of laughter for which the reference has recognisably served as impetus (line 22). 17 Cal:

18 19 20 21 22 23 24

= So:.

(.) Len: Cal: Bart: Cal: Len: Bart:

O o ::p(h)s = 'îhey= neghh-.heh heh. 'huh ha- 1 ha: ha: ha:.ha a h ah! ah! l ~ ~ e!e! e ! e ! a ! l hgheh-heh'

It is after this third extension of the laughing-together that the joke's telling is resumed (data not shown). The foregoing considerations suggest that laughter can be an achieved product of methodic, co-ordinated processes, with occasion of laughing together oriented to, produced, extended, as an event in itself. However, it appears that in some characterisable situations laughing together occurs as an accessory activity, performed as a way to arrive at some specifiable outcome. It thus becomes not only a relevant, consequential response to a prior utterance, but has a significant bearing on a, or some next action(s). The next section focuses on laughter in such a type of situation, considering its systematic, socially organised aspects by locating it as a component of a particular type of sequence: the expanded affiliative sequence.



Laughter as a Component of a Sequence In the course of ongoing talk someone may say something which breaches conventional standards of courtesty, propriety, tact, ethics, commonality, etc. etc., the breach in conventional standards at least potentially being offensive to other parties to the interaction. While there are various ways to breach conventional standards, the major focus here will be on such obvious breaches as rudeness and obscenity. The introduction of 'improper' talk 4 may have an interactional basis. That is, it is a convention about interaction that frankness, rudeness, crudeness, profanity, obscenity, etc., are indices of relaxed, unguarded, spontaneous; i.e. intimate interaction. That convention may be utilised by participants. That is, the introduction of such talk can be seen as a display that speaker takes it that the current interaction is one in which he may produce such talk; i.e. is informal/intimate. Further, the introduction of such talk may be, not only a display of a perception by one party of the status of the interaction, but a consequential, programmatic action. By introducing such talk, a speaker may be initiating a move into intimate interaction from a status he perceives as non-intimate so far. Speaker may be offering an invitation to his co-participants to produce talk together whereby they can see themselves as intimate; together they will be constructing intimacy. If that is so, then recipient treatment of such talk may be produced by reference to its invitational properties; a recipient, then, not merely deciding whether the object itself is attractive or repugnant (in general or for this recipient in particular), but may be seeing an invitation which is to be accepted, rejected, or otherwise managed. Such a characterisation of improper talk provides a framework which leads to examination of talk following the impropriety; to an investigation of recipient treatment of an invitation to intimacy. 5 A collection of actual responses to improper talk can be made, and instances of such response can be arranged on a hypothetical continuum ranging from rejection to enthusiastic acceptance, from disaffiliation to escalation. In the following fragment, recipient of an impropriety (line 2) disaffiliates (line 6). (12) [Cole:I:5:r] 1 Jo: 2 3

U h : m , (0.2) make su:re, evrything i:s, (0.4) organic and orgasmic. (0.4)



°hhhhhhhh-° (0.4) °hh-hh° ih(hh)y(h)ou said it I didn't,

In the following, recipient declines to respond (line 2). Response is pursued by offerer and again declined (lines 3-5). The problem is resolved with a shift in topic, initiated by recipient (line 6), accepted by offerer (line 8). (13) [GTS:III:8:r] 1 Ken: 2 3 Ken: 4 5 6 Louise:

She's gotta jacket thet's diarrhea brhhhohhhwhnh, (0.8) hhhhihh!(h)N(h)o j(h)oke. u - hhhihhh! Ri(h)illy. hhh It's ho:rrible. 'hhh Jis tuh think about it. It gits you si(h)ck, (1.0) Yihknow it's almos' twenny e-it's um,

7 8 Ken: 9 Louise:

" (•) W e may not,have a session. .th's morning. (seventeen) after?

In the following, recipient of an impropriety (line 5) disattends it while responding to an innocuous aspect of the carrier-utterance, providing an innocent 'understanding check' (lines 9, 11). In this case it is offerer who abandons the entire carrier-topic (line 14). (14) [NB:X:6] 1 2 3 4 5 6 7 8 9 10 11 12 13 14


Emma: Lottie: Emma: Lottie: Emma: Emma:

Yeah it's jis scaling o . f f , ' n and uh it's jis, evry time I take a bath'n soak why they jus' come o:ff.yihknow en then the ta: r, I don'know what t h e - hhh I haftuh have two tablespl- s - my tub is really beautiful at home you oughta see it. Looks like a niggersss (.) khh (.) O h it'sbla:ck hu.:h, l Yeahhhhh En yih jus soak in tha: t .hu:h, Yeah, (0.5) tlh How'v you been.

In the following two fragments, recipient appreciates the impropriety; in the first fragment with a lexical token (line 3), in the second with laughter. 6


162 (15) [Goldberg:JG:I:3] 1 George: 2 3 Hap:

We came ho: m e n (0.4) screwed aroiKn' ((clears throat)) Literally, "'uhh'hhh.hhhhhh 0 l WelH'tlbeda::rned.

(16) [Labov:BG:5] 1 Joan:

a load of shit.

2 3 (

O e h h h n h h u h huhlhnh!


In the following fragment, a recipient affiliates by replicating the impropriety in his own next utterance, and thus accepts the invitation offered by the impropriety. (17) [Labov:TA:l 1] 1 Daniel: 2 3 Daniel: 4 Arlene:

hh en halfway ho:me all'va sudden I's threw up. dry:. — (1.0) 'n I pulled duh car OjVer'n I wz covered with .sweat. 'khh! Wul(h)wuun throw up wet'n nat new car!

And in the following fragment, a recipient escalates the impropriety. T h a t is, not only does he accept the invitation to intimacy, but himself adopts the position of an offerer, inviting acceptance of an invitation to still deeper intimacy. The fragment is excerpted from a sensitivity training session for prison guards (recorded in 1963) in which the members are being encouraged by the group leader to name things as disgusting to them as a homosexual inmate. There is a gradual progression (not shown) from 'Oh I can't think of a particular thing right now' to 'things . . . that don't uh follow in the social acceptance of society - 1 guess.' to 'Someone who'd take off their shoes and put 'em up on the table-'. Thereafter, the following occurs: (18) [Ward, Kassebaum: 11:25] 1 Donnely: I can give you a good example. I was getting ready to go 2 back to the ship one night, and this one fellow who worked 3 for me, he was a little bit drunker than usual and he kept 4 going on a crying tear, and lying on the fence there, and 5 when we had to go back to the ship, "Ain't you gonna help 6 me back?" and then he would heave, and roll around in it — 7 That ain't exactly disgusting. Sickening maybe8 Arlett: I'll go for that. 9 Donnely: You have all the puke and vomit from him, and he rollin 10 around in it, and keep on crying and needs help back to the 11 ship and all t h a t -

L A U G H T E R IN P U R S U I T O F INTIMACY 12 13 14 15 16 17 18 19

Falker: Arlett: Baines: Benson:



I w o u l d feel a b o u t the same way about that m a n that I would a b o u t a q u e e n . T h a n k s , M r . Donnely. Disgusting. I'd feel the s a m e way about something like that, or how a b o u t a h a n d f u l of shit. O r h o w a b o u t s o m e guy drinking his piss or something like t h a t a f t e r a twenty mile hike or something like that. W e got a couple of q u e e n s over there in my building that disgust m e in that s a m e way. N o w the rest of t h e m , no trouble whatsoever.

T h e instance of a disgusting thing is arrived at via a story, and the transcript shows a pause thereafter (the double dashes, end line 6); i.e. no recipient has o f f e r e d a response. Perhaps by reference to an observable declination to r e s p o n d , offerer starts to disaffiliate from his own position (line 7), but is overlapped by recipient's assertion (but not demonstration) of affiliation (line 8; in this transcript a dash at the end of an utterance indicates it is overlapped by a next utterance). Thereafter, and perhaps by reference to the asserted affiliation, offerer reasserts the improper component (lines 9 - 1 0 ) and another recipient asserts (but does not demonstrate) affiliation (line 12). T h e initial affiliator (Arlett) now provides another assertion of affiliation (line 14) and moves to an escalation (line 15), to which still a n o t h e r recipient affiliates, now by demonstration (line 16). The series is t e r m i n a t e d with yet another member of the group asserting affiliation (lines 18-19) while shifting focus from a search for apt comparisons to the troublesomeness of particular queens. While actual instances of the various response-types were arranged on a hypothetical, perhaps arbitrarily ordered continuum, Fragment 18 indicates that at least a portion of that continuum, may reflect an actual type of progression: Affiliation followed by escalation (e.g. lines 14-15). It turns out that various segments of the continuum do occur in sequence. A regularly occurring progression is: Disattention followed by appreciation followed by affiliation. T h r e e extended fragments are shown and the sequences within t h e m are minimally sketched. 7 (19) [ G o l d b e r g : I I : 1:2:1) 1 Gene: A r e yih avoiding m e like the plague, = 2 Maggie: = n N o : : : of wahl yih nah you know better th'n.t


3 Gene: I'm not syphletic, = 4 Maggie: = h h h N o .1 know yer no .t, h h h h . h h, 5 Gene: ( ) 'heh, h e - h e h - h e h - h e h - h e h - . h e h - . h e h

6 Maggie:

h h heh heh huh =

7 Maggie: = h h h h h h . I k e e p running t e : s t s o n y u h I know yer not. =

8 Gene: 9 Gene:

10 Maggie:




= e h h he-,heh-heh-heh-heh-heh.°hn° =


TALK AND SOCIAL ORGANISATION 11 M a g g i e : 12 13 14 15 G e n e : (20)

Y e : h . What's happ'ning.


1 Ted:

2 3 Guy: 4 Ted: 5 Guy: 6 Ted:

8 Guy: 9 Ted: 10 11 G u y :

N o we c'm in fr'm the beach'n then we c'm in'n to take a n a : : p yih k n o : :w, it's .really jwe really get—j _ ~ Ye::h y o u : : , _,'hhhh screwin a r o u , : n there uh:,. Y e : : : uh: 'n then we t - k (0.9) W ' n . a ' t h o s e k , i d s - .come.s in'n dih ((f)) d a : d d y . m u m m y : lh e v a 'bee::l::r, 1

12 G u y : 13 T e d : 14 G u y : 15 T e d : 16 17

= U H : : h h h h h h h h N o G e n e I'vejust hh hhhhhh been in en been out'n sometimes hhh y'know the p a : t h s 'hhh c r o : s s , bu:t uh th'ti:me is b a : d , (0.7)

(•) ( ( f ) ) daddy = = hhhhe , h - h e h - h e h - # h - = ) ) = h : : : : : : . eYe:::::hhhah-huh. h

eh-hah-he::h u-,hu:h, h e : h uh'hhh.hhh hu-uh,

(•) Guy:

18 T e d : 19

Guy: Ted: 21 Ted: 22 Guy: 23 Ted: 24 25 Guy: 26 Ted: 27


°huh-e-hufh.° nn (ff) G e t outta the: re. = = h e : h heh ,heh heh h e h h u : h h u - u h ' h h h h a h h h h h . _ 'huh huh hnuh heh-huh- uh-uh-°uh° =hh.hhh A a n Wj'l (that's th'way it goes), N o : : : no hankypanky.


" N o hanky panjky. 0 j N o : : :: hanky panky. (0.3) 28 Guy: Well have a good time. (21) [ N B : H T : 2 : r ] 1 Emma:

W'l M a r t h a ? N o , w I'd love tuh have you join us,

2 Martha: 3 4 5

6 7





(•) Emma:

If y o u : : feel ez though you'd like t'come over,


u-Well thank you dear I don't think so I had my little hh he:n, .a:nd uh, loMmhm,°1= ' h h I ' m looking forward to just uh: h h h a v i n g a : ( ' h h ) a little:: (0.4) ti: me tuh myself, = =°A::r,ight,° , T v e lo ok'forward t(h)o(h)i(h)t s(h).o l(h)o- hh '((f)) O h : : : 1


Emma: 9 Martha: 10 11 Emma: 12 Martha: 13 Emma:


L A U G H T E R IN P U R S U I T OF INTIMACY 14 Emma: l*t I^MlllUt._c, ,,::::::: 15 M a r t h a T

16 17 18 19 20 21 22

,, *oUli.., ohh::::

Yihknow like- Garbo,

Emma: Martha: Martha: ..heh-heh, Emma: Ah wagh nt t'be alo. Martha: Emma: Martha: = An'Guygotdo:wn.=

23 Emma:


hkhh.hh y.Yehh.hheh ehhh

= ' h h h h Y e : s he w z h e : : r e after I: c a m e h o : m e . . .

In Fragment 18, a same speaker (Arlet, line 14) might be characterised as preparing the way for escalation with an affiliation. Specifically, he is demonstrating that his obscenity is occasioned by a prior and occurs as part of an acceptance of the prior's invitation to intimacy. In Fragments 19, 20 and 21, a similar characterisation might be applied to two participants. That is, in these cases the participants are collaboratively preparing the way for affiliation, providing a display of its sequential occasioning and its character as an acceptance of an invitation. Thus, the sequences can be treated as arrivals at affiliation which, when it does not occur in a recipient's next u t t e r a n c e , as it regularly can (see e.g. Fragment 17 lines 1 and 4 and F r a g m e n t s 18 lines 6 and 8 , 9 and 12), then it can be achieved over a series of internal expansions of the base sequence (Jefferson & Schenkein, 1977). T h e expansions in the three fragments under consideration run off in three discrete steps, as follows: Impropriety

followed by disattention

In each fragment, although an acknowledgement is done, there is no explicit u p t a k e of the impropriety. (19) 3 Gene:

I'm not syphletic, =

4 Maggie:

= h h h h N o I know yer not,


3-5 Guy: 4-6 Ted: 7 9 Ted:

Y e : :h you : :, .screwin a r o u . : n there u h : , , hhhh

Y e : : :

u h . ' n then we t - k -

(0.9) hevabee:::r,

(21) 9 Martha: 10

hh I ' m looking forward to just u h : hh having a: ( hh) a little:: (0.4) ti : m e tuh myself, =

11 Emma:

=°A:: right,0



Disattention followed by laugh-appreciation Recipient's laughter is itself arrived at over a series of moves: (a) offerer issues an invitation to laugh, and (b) recipient accepts that invitation. In Fragment 19 the invitation to laugh is itself laughter (see Jefferson, 1979). (19) 4 Maggie: 5 Gene: 6 Maggie:

h h h h N o I know yer no,t, h h h h h j"h h. heh, he-heh-heh - h e h - heh-.heh-,heh. h h heh heh huh

In Fragment 20, the invitation to laugh is a comedic, falsetto-voiced enactment (lines 8-11, ((f)) indicates falsetto), followed by a laughed invitation to laugh (line 12). (20) 8 Guy:


11 Guy: 12 Guy: 13 Ted:

W'n a'those kids-comes in'n dih ((f)) da:ddy.mummy:


((f)) daddy = = hhhhe,h-heh-heh~#h-#h::::::. 'eYe:::::h hhah-huh, h 1 e h - h a h - h e : : h u - h u : h

And in Fragment 21, it is a recasting of the carrier-utterance, now with laugh-particles inserted (line 12, cf. lines 9-10). The sound which receives it (line 13) may be sympathetic, but it observably is not laughter, and thus declines the current invitation, which is to laugh. Thereupon, the invitation is cut off (end line 12), and a next is produced, in the form of a comedic comparative reference (line 15). (21) 11 12 13 14 15 16

Emma: Martha: Emma: Emma: Martha: Emma:

A : :r,ight,° I've lo ok'forward t(h)o(h)i(h)t s(h) o l(h)o- hh ((f)) o h m 1 ::::::: *oh:: Yihknow like- 'Garbo, hkhh hhh y:Yehh hheh

Laughter followed, by affiliation In Fragment 19 the affiliation proposes independent tracking by recipient of offerer's status as possibly syphletic.



(19) 3 Gene: 7 Maggie:

I'm not syphletic, ((laughter)) hhhhhh I keep running te:stsonyuh I know yer not.

In Fragment 20, affiliation is done by producing a next activity in the falsetto-voiced enactment. (20) 8 Guy: 9 10 Guy: 18 Ted:

W'na'those kids-comes in'ndih(ff)) da:ddy.mummy: (•) ((f)) daddy ((laughter)) ((f)) Get outta the:re.

A n d in Fragment 21, the affiliation is done with the Garbo signature. (21) 15 Martha: 19 Emma:

Yihknow like- Garbo, ((laughter)) Ah waghnt t'be alo::::ne.

In each case, the affiliation not only follows laughter, but laughter by both parties (F.19 lines 5-6, F.20 lines 12-17, and F.21 lines 16-17); i.e. the affiliation occurs within an occasion of laughing together, an event which can be constructed and expanded in its own right (see p. 158ff). And a standard way in which such an event is expanded is by one of the participants producing lexical reference to the talk out of which the laughter was initially generated (see pp. 158-59), where contributions to the occasion can be and are made by others than the initial speaker, as in Fragment 11 and the following, in which there is alternation of two speakers as next contributor (lines 15,20 and 24). (22) [GTS:II:2:90:r] 1 2 3 4 5 6 7 8 9 10 11 12

Roger: Ken: Roger: Al: Ken: Ken: Al: (Jim): (Ken): Roger: (Al):

Hey ah'll bring in a gi:rl hhhe:h ehh heh- hnh = = Yah.Less alluring innagirj Wewon'getmucha,cco:m,plish\hih , E Y. 1 = 'Hey 1 = Let's all bring inna. g i r 1 next week.. Tell'er it's a da:te. (.) hhih-hhi.h 'ek.hnhh [ h e h - H A HA hhehh. hehh. _ hnhh


168 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Roger: (Jim): Al:

= It's nchi(h)pe(h)h'hheneh = = hhhh 'hehh = = W'rgoin Du:tch.

Roger: Ken: (Jim): Roger: Al: Roger: Al: Al: Roger: (Jim):



= hhhh We're.GOING G R O U : : P , =

( )

= hheh hih-e Hheh-ih°nhh' 5 hhhh = = Goin group'n the group's Du:tch y'know.(w't ^w 11I mean mean?) : '1 = e h - heh-heh = mhhhhh hmh-hmh

The occasion of laughing together can serve as an environment in which recipient, to contribute to its extended occurrence, might properly produce a lexical reference to the source of the laughter; i.e. to the impropriety. And, as shown in Fragment 22, any next contribution may be tightly related to its prior, may work off it, play with it in a range of ways, and may thus demonstrate an understanding of the impropriety itself (rather than the at best equivocal tokens which occur in the disattention component of the sequence). By producing such an object, recipient thus becomes implicated in the sort of mentality which produces such talk; i.e. affiliates to the impropriety. The occasion of laughing together also provides a restricted field. That is, while the affiliation implicates recipient in the mentality which produced the impropriety, it has as its specific, local, sequentially appropriate job, the extension of the occasion of laughing together (in contrast, perhaps, to such direct affiliation as that in Fragment 17). In the fragments under consideration, the laughing-together may be characterised as pre-affiliative. It provides an environment which simultaneously urges for and restricts the domain of an activity (affiliation) which might relevantly have occurred earlier but has been withheld pending just such negotiations as would provide that its eventual occurrence is both sought after and restricted in its domain. In this sense, laughter is not merely a particularly apt next event following recipient disattention to an impropriety, but may be specifically relevant, given the occurrence of disattention, as a way to arrive at affiliation. That is, laughter systematically occurs as a mid-component in an expanded sequence of which the base sequence is Impropriety followed by Affiliation. Further, in the fragments under consideration it appears that the occur-



rence of affiliation completes a sequence. In each case it can be observed that soon after affiliation is achieved the laughing-together is terminated. In Fragment 19, offerer produces a next laugh-burst (line 9) and on its completion (a diminished laugh particle, cf. Fragment 11 lines 12, 15, 16) recipient provides an answer (lines 11-13) to the question to which the impropriety was an appendation (line 1 'Are you avoiding me like the plague?', line 3 'I'm not syphletic,'), and offerer takes up the return to business (line 15). That is, while offerer proposes to extend the laughingtogether after affiliation has been achieved, recipient moves to terminate it. In Fragment 20, offerer initates and recipient joins in a next laugh-burst (lines 19 and 20). Thereafter both move to terminate the laughing-together, offerer with a closing assessment 'Aah well that's the way it goes,' (line 22), recipient with a denial of the initial characterisation of his activities (line 23, vis-à-vis lines 3-5). s That is, the laughing-together is extended, but at a point where a next lexical contribution is due, both move to terminate the occasion. And in Fragment 21, offerer shifts topic as the affiliative utterance approaches its projected completion (lines 19 and 20-22), and recipient immediately takes up the proffered new topic (line 23). That is, just as affiliation is achieved, offerer moves to terminate the laughing-together. In each case, closure is activated subsequent to affiliation with no further lexical contributions to the laughing-together. In Fragment 19 it is recipient who moves to terminate, in Fragment 20 it is both parties who move to terminate, and in Fragment 21 it is offerer who moves to terminate. The activity-identities offerer and recipient appear not to matter for the termination of such a sequence. One way to account for the placement of the moves to terminate (i.e. prior to the point at which a next lexical contribution is due), and for the irrelevance of the categories recipient and offerer for the move to terminate, is to characterise the sequence in terms of the continuum proposed earlier (p. 160ff). In each of the three fragments, three response-types occur in the order proposed for the continuum (disattention, appreciation and affiliation). The response-type proposed as last on the continuum (escalation) does not occur in these fragments, and may be characterisable as not, yet, having occurred (see, e.g. Fragment 18, lines 7-9 and 14-15 for the relevance of affiliation to escalation). That is, termination of the expanded affiliation sequence cum laughing-together may be activated by reference to avoiding a lexical contribution which might well turn out to be an escalation. The prospect of escalation might serve as an impetus for either or both participants to terminate an otherwise valued and methodically constructed event (both laughing-together in its own right, and the



offering and acceptance of intimacy via an occasion of laughing together), since escalation, as a next possible breach, can project a next point atwhich any of the non-affiliative response-types (disaffiliation, declination to respond, disattention, or 'mere' appreciation) can occur singly (see Fragments 12-16). That is, if an escalation occurs, then it is possible that further talk will not have affiliation as part of it. Thus, having achieved a level of intimacy perhaps not present prior to the introduction of the impropriety, having stabilised at that level with the occurrence of affiliation, further pursuit of intimacy is abandoned before escalation and its possible consequences can occur. At the very least, the foregoing considerations suggest that (1) laughter is a methodically produced activity, which (2) can itself be a component of a methodically produced sequence of activities; i.e. it is socially organised in its own fine-grained particulars, and at a grosser productional level, as well. T h e sorts of analytic resources developed in the foregoing are now turned to the detailed analysis of segments of a single conversation, which constitute an extended pursuit of initmacy.

Case Study: Laughter in Pursuit of Intimacy Following are five fragments of a single telephone conversation in which an impropriety (mention of participation in nude swimming) is repeatedly offered. 9 Across the series of mentions, affiliation is pursued and eventually achieved, whereupon escalation occurs and a next cycle is engendered. The five mentions are considered in turn.

First Mention Recipient disattends improper component. (23) [NB:PT:3:r] [Timexa.Ol:30-02:03] 1 2 3 4 5 6 7 8 9 10

Lottie: Emma: Lottie: Lottie: Lottie: Lottie: Emma: Lottie: Emma:

Jeeziz Chris'shu sh'd see that house E(h)mma yih av no idea h.hhmhhi I bet it's a drea.m.With a swimming poo:l enclo:sedjhu:h = u= Oh:::: .Gho:d we-'hhh uh-hu we swam in the nude = = hh Sundee night = = u(h)ntil abou.t, e hhh,h i h h ,huh,huh,ha:, ha, two uh' clo: ck. 'HUH HA HA: ::. hhhhh Oh:


= I b e t ' n the moonlight'n the beautiful stars the wind blew


terribly t h o u . g h ,


13 14 L o t t i e : 15 16

First mention (line 5) is observably precipitate in its delivery. For one, it veers off a topic which offerer had just taken up (line l). 1 0 Secondly, it disattends that the prior utterance (line 3) is a request for confirmation (a request which, when subsequently reissued and attended (fourth mention, Fragment 30, lines 13-22) generates extended talk), and utilises mention of the swimming pool as an occasion for mention of the nude swimming. It is possible that first mention is an instance of 'triggered' or 'touched o f f talk. 11 Further, participants to the reported activity are introduced as 'we' (line 5). In general, participants to a reported event tend to be identified rather than initially pronominalised (and in particular, in these materials, nominalised introduction is done, cf. Fragment 27 line 20). The pronominal introduction here may be attendant to the precipitate delivery. And it may be problematic. It is not that recipient would be unable to infer from 'we' who the participants to the activity were, but that an inference could be made which would tend to enhance the impropriety. That is, the 'we' has a candidate complement of three, the guest and host couple, one of whom is male, and there has been reference to the couple in immediately prior talk (data not shown) which may contributed to a tri-partite sense of 'we'. It is observable that reported absence of the male is a recurrent feature of subsequent mentions (cf. Fragment 27, lines 16-21, Fragment 29, line 4 and Fragment 31 line 30). It is possible then, that a byproduct of the precipitous introduction of the impropriety is an unintended aggravation of its possible offensiveness. There is a momentary break in speech following delivery of the impropriety (lines 5 - 6 ) '. . . we swam in the nude hh . . .'). While the break is momentary, it may be consequential; recipient silence at that point constituting a potential disinclination to respond.12 It is possible that subsequent talk by offerer is oriented to, and remedial of, that declination. Specifically, it appears that an 'offerer's correlate' of recipient disattention to an improper component is produced. First, a continuer (in this case '. . . Sunday night . . .') 1 3 and thereafter a de-escalated alternative to the impropriety which preserves reference to the activity and re-offers it as a respondable (in this case an alternative description of ribald fun, '. . . until about two o'clock.'). A similar sequence is found in Fragment 17 (p. 162), in which there is a substantial silence following the impropriety. That is, a potentially offensive description of illness (line 1 '. . . I just threw up. dry.') is followed



by a one second silence (line 2).14 Following the silence is a continuer (line 3 '. . . and I pulled the car over . . .') and a de-escalated alternative description of illness ('. . . and I was covered with sweat.'). Such a sequence may be characterised in the following way: In the absence of response to the offensive component, offerer provides talk which serially (a) retroactively displays that no response was required; i.e. that this is not an instance of a proffered improper respondable to which a recipient declined to respond, but rather, merely a non-problematic item, with further talk simply continuing, and (b) remedies the fact that absence of response to the improper component has also constituted absence of response to the reported activity I event, by re-offering the activity/event, now cleansed of impropriety.15 It is the availability of the same type of respondable without its improper aspect which makes this sequence an offerer's correlate of recipient disattention. And in the cases under consideration, recipient responds in the course of the cleansing sequence; in Fragment 17 with affiliation (line 4), and in first mention with a possible pre-affiliation, laughter (line 8). However, recipient's laughter in this particular case may constitute disattention to the improper component rather than appreciation of it. It is positioned in such a way as to display anticipation of the projected (deescalated alternative) respondable. Specifically, it starts at a point in the continuing utterance at which the projected naming of a late hour is 'due'. In general, due-point is a standard locus of overlapping talk (see Jefferson, 1973), and in the following fragments laughter is initiated at, or around, due-point. 16 In Fragment 24, a first anticipatory appreciation occurs at due-point for 'box' (line 2), a next occurs at due-point for the contents of the box (line 4). (24) [Labov:BG:5] 1 2 3 4 5 6 7 8

Joan: 'hhh She wrapped u:p in a very ni(h)cely gift wrapped, = Betty: _,,ss::: Joan: " U bo(hh)o:,k Doris: hn-hnn hn-,hnn hn-hmnnn, = Joan: bo:x, hh Joan: = a load of shit. (•) ( ): ehh hnh huh huhihnh!

In Fragment 25, two recipients start almost simultaneously at due-point for the object which will constitute a joke's punchline. (25) [Goodwin:AD:58:r] 1 Cal:

"I gotta git outta dih mood befo'I c'n git outta



d.i.hcah," . a.hhaha .hahahaha u-hu-huh hu:h hu:h hu:h

The laughter in first mention has a similar configuration vis-à-vis duepoint for the projected late hour. (23) 7 Lottie: 8 Emma: 9 Lottie:

u(h)ntil a b o i y , ehhh.hihh .huh.huh.ha:,: ha: twouh' clo: : : ck.

Further, in Fragment 25 and the following fragment, the laughter is escalated at or near completion-point of the anticipated component. 17 (26) [Goodwin:AD:63:r] 1 Bart: 2 Cal:

". . . 'n took mhhy fif^t(h)y c(h)e(h)ents" . hnn-hnn-hnn- hah- ha:h-ha:h

That is, there are increases in pitch and amplitude, and/or stretchings and/or openings of the shape of the particles (in Fragment 25 from 'ha' to 'ha' and f r o m 'huh' to 'hu:h', in Fragment 26 from 'hnn' to 'hah' to 'ha:h', and, see Fragment in note 16, from 'heh' to 'ah!' It can also be noted about Fragment 24 that while the anticipatory laughter stops and the anticipated object is delivered in the clear, the laughter which appreciates it is distinctive from the anticipatory sounds). The same is true of first mention, the laughter shifting from 'huh' to 'ha::', as the stretched 'clo:::ck' arrives at completion. 1 8 Methodically, then, via placement and sound-shifts, the recipient of first mention's laughter is produced as anticipatory to, and subsequently appreciative of, the late hour until which the activity took place, thereby disattending the improper component. In effect, the laughter collaborates in the cleansing sequence undertaken by offerer. However, offerer's activities appear directed to exploiting the fact that the appreciation of the de-escalated respondable is done with laughter (in contrast, e.g. to a lexical which specifies that which response is directed to). Roughly, offerer at some point stops talking and starts laughing, and thus contributes to a laughing-together which (cf. Fragment 19 lines 3-7 and Fragment 20 lines 5-18) can eventuate in affiliation. Offerer's laughter provides that the laughing-together is a single, rapidly escalating burst. That is, it reproduces the shift in particle-shape via which recipient escalated (line



9 'huh' to 'ha:: ha:'), simultaneously re-escalating via increased amplitude (line 10 ' H U H ' to ' H A HA:::.'). The result is a display of, not merely laughing at the same time, but laughing in the same way. 19 But just as offerer initiates the laughter which will constitute a laughingtogether, recipient stops laughing and starts to talk, whereupon offerer stops laughing (lines 8-11). And recipient's talk now specifically attends the de-escalated aspect of the carrier-utterance; i.e. attends that the activity took place late at night, with a continuation of the earlier dream motif (line 3 'I bet it's a dream') with lines 10-11 'Oh I bet in the moonlight and the beautiful stars . . .'. That is, systematically in its productional details and as subsequently lexically demonstrated, recipient's laughter disattends the improper component. A n d latched to the utterance which demonstrates an ongoing attention to innocuous aspects of the carrier-utterance, recipient offers a shift in topic (lines 11-12) which is accepted by offerer (line 14), who uses the topic shift to close the report (lines 15-16). First mention, then, consists of a disattended impropriety with the carrier-topic mutually abandoned (cf. Fragments 13 and 14, p. 161).

Second Mention Recipient disattends improper component and tends to disaffiliate. (27) [NB:PT:14:r] [Time:ca. 11:25-12:15] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Lottie: I left theh:: (.) restr'nexacly a quarter t u h e i : g h t , = E m m a : _ ..Mm hmmm,m Lottie: hhhhhh En I called- Isabel ettwunny five minutes t u h t e : n that's in s , : - i : n uh ,Pa:lm Spri:ngs. = Emma: (en got there.) Emma: = °*aOh::::w'l that's wonderf.ul. 0 Lottie: En then:: uh tha'wz eighteen miles so 'e siz w'l jis come on down s u h : - til yuh see the tiki lights::,'n I'll be .the:re. Emma: M m m hm, (.) Emma: Mm hm, (.) Lottie: So I drove on they were waiting for m'went h o m e - ' hhhhh = Emma: = °A.hhah.°, Lottie: En the :n, course Dwighthadtuh get up (0.2) uwul'e u-hegotupet(0.2)si:x. ' (.)

LAUGHTER IN PURSUIT OF INTIMACY 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma: Lottie: Emma: Lottie:


M,mmhmm:, th'next morning, so Isabel'n I en(h)w(hh)e swam in th(h)at pool until two uh'cl(h)o,ck in the,morning. = l O h :: , 1 = i(h),in the n . u : d e . = l G o : d! = hh,u-(h)Oj(h)oh GOjd ih wz^:: Isn't she c u : : :te, = = fun. = = 'hh She still drinkin er liddle dri:nks? (0.4) Y e : a h ' n , t h e n - ,we swam a:ll day t'day = '"Yeah," 1 = I d - I never, (.) well I got out about erry (.) five minutes er so.'n then 'n . t a k e . a "Ohlbetch e rJta:nned. (.) hh Y e a h . K i : n ' a . Y e a : h . = = Mm hmm:, = = hhh En the-.n, a h j e f t t h e r e e t u h : : : : (0.5) exacly et three o'clo:ck. pt hhhh En I didn'git inna any traffig e'all. . .

Second mention (lines 20-23) has a series of features which stand in contrast to features of first mention, which may constitute revisions in response to first mention's reception. (1) In contrast to the off-topic, triggered, precipitous first mention, second mention emerges out of a chronologically marked course-of-events narrative (lines 1-23), as a next event in that narrative. 2 0 (2) While first mention refers to the event's participants with a pronoun which does not exclude a candidate male participant (Fragment 23 line 5), second mention first implicates his absence (lines 16-20) and then specifies that the participants were the two females (line 20). (3) While in first mention the location of the event is implicit or to be understood by reference to the trigger-component 'swimming pool' in the prior utterance (Fragment 23, line 3), in second mention there is a place formulation 'in that pool' (lines 20-21) which may operate both interactionally and sequentially. That is, it is a 'recipient-designed' formulation, acknowledging or proposing recipient's familiarity with the object, 21 and such a formulation as 'that X' can mark a candidate topic, 22 offering, by reference to the shared familiarity with the object, the possibility of some extended talk about this particular segment of the ongoing narrative. (4) While in first mention the impropriety is delivered immediately (Fragment 23 line 5), in second mention there is a progression. The activity is initially formulated innocuously in a possibly complete (and intonationally completed) sentence/utterance (lines 20-21 '. . . we swam in that pool until two o'clock in the morning.') and the improper component is latched as a post-completion 'kicker' (line 23 'jn the



nude.'). 23 Attendant to this progression from the innocuous to the improper, there is a replicated format, 'in the X', which in first mention carried the improper component 'nude' (Fragment 23 line 5), which in second mention initially carries an innocuous term 'morning' (line 21) and subsequently carries the improper component 'nude' (line 23). In a range of ways then, the second mention is artfully arrived at and presented as a something to talk about. Further, it is presented as a something to laugh together about, with a series of within-speech laugh particles (lines 20-21). Recipient, however, pursues the innocuous, declining the invitation to laugh with a lexical appreciation of the late hour (line 22). A next invitation (line 23) is likewise declined with a recognisable continuation of the lexical appreciation of the late hour (line 24). While declining to appreciate the improper component, recipient may be recognisably hearing it. Specifically, the initial token and the continuation only partially overlap the 'in the X'-formatted phrases, giving clearance to their key terms. 21 22 23 24

Lottie: Emma: Lottie: Emma:

until two uh'cl(h) in the,[ X ]. = *0 h : : , 1 = i(h) f in the (, X ]. Go:d.

While one resource in the management of overlap is the stretching of a word, and while both of recipient's appreciations are stretched, they may be characterised as, not merely stopping prior to the other party's stopping,24 but as stopping at due-point for the 'X' of the 'in the X' format, a term which, in first mention was 'nude', and in second mention is, first 'morning' and then 'nude'. It appears that recipient is, twice in succession, producing talk which simultaneously appreciates the innocuous and gives clearance to, listens for, hears, the impropriety. By appreciating the innocuous while listening for the impropriety, recipient's talk may be equivocal vis-à-vis the possibility of forthcoming affiliation. And in simultaneous next utterances (lines 25-26) each part offers an alternative direction. Both utterances are assessments, but offerer's is appreciative and recipient's, while it is conceivably appreciative, tends to disaffiliate. As in first mention, offerer reproduces the object with which recipient appreciated the innocuous aspect of the carrier-utterance (line 25 'Oh God,' cf. line 22-24, cf. Fragment 23 line 8-9 'huh ha ha'), in effect applying recipient's appreciation of the innocuous toward appreciation of the impropriety with 'Oh God it was fun'. Simultaneously, recipient



is in the course of applying the object to an alternative assessment; i.e. is producing a continuous utterance 'Oh God isn't she cute,' (lines 22,24-25). While the assessment term 'cute' might be appreciative, the utterance tends to disaffiliate in that it selects for assessment, not the activity itself, nor a relevant alternative assessable, the co-participant to the activity who is this conversation's co-participant, but the co-participant to the activity who is non-present third party in this conversation. The two simultaneous assessments, then, are competitive alternatives occurring in overlap. A series of overlap-management techniques are deployed via which each assessment is designed to outlast the other. Recipient starts to stretch the assessment term and offerer stops prior to completion, thus potentially yielding the turnspace, whereupon recipient brings the stretched term to completion. 25 Lottie: 26 E m m a :

h h . u - ( h ) o , ( h ) o h G o . d ih w z , : : Isn't s h e cu:: :te,

Offerer thereupon latches with a continuation (and completion) of the discontinued utterance. 25 25 Lottie:


26 E m m a : 27 Lottie:

Isn't = fun.

s h e



At line 27 both utterances have reached completion and offerer's has outlasted recipient's. Via turn-taking systematics, recipient expectably will talk next (see Sacks et al., 1974), and via the organisation of assessments, will expectably produce an acknowledgement/response for that 'first' assessment. (See Pomerantz, this volume, Chapter 9.) However, while recipient takes next turn, a procedure characterisable as 'skip-connecting' is employed, via which a next utterance is produced by reference to current speaker's own prior rather than by reference to a co-participant's immediately prior utterance. 26 In this case, recipient produces a question (line 28) which locates recipient's own third-party assessment (line 26) as its base. The question provides a disaffiliative candidate account for the activity being assessed by proposing it to be one of a series of cute-typable activities third party is wont to engage in, the second (drinking) constituting a condition under which the first (nude swimming) might routinely occur. 27 26 E m m a : 27 28 E m m a :

isn't she c u : : : t e , ' h h She stil drinkin er liddle d r i : n k s ?



Further, the question contains a 'still X?' format. While the 'that X' format (e.g. lines 20-21) refers to mutually familiar matters, the 'still X?' format refers to a mutually familiar matter which requires updating, where that updating might permit or require extended talk. Thus, the introduction of a 'still X?'-formatted question can serve as a topic initiator. So, for example, in the following fragment, a possible lapse in the ongoing talk is followed by a 'still X?'-formatted question and extended topical talk ensues (data not shown). (28) [Goldberg: 11:2:5] 1 2 Maggie: 3 Gene: 4 5

(1.2) Are y'still all teaching school, Yah. ( hhhhhhhh) Yeah we're teaching (in uh:) Oh I gottaI don'know th'las'time I talked (t'yuh) I'm out here et Kroft Highschool (now),

In second mention the candidate topic is updated with a minimal token answer which itself receives a minimal token acknowledgement. Latched to the token answer is a return to the course-of-events narrative (line 30-33). The return, however, is not to the nude swimming, but to a next event in the chronologically organised narrative, 28 and that next event is presented in innocuous form. 2 9 Thus, while declining to engage in extended talk on the topic proposed by recipient (the hostess's drinking habits), offerer accepts the fact that a change has been requested; i.e. talk about nude swimming is discontinued. Subsequent talk addresses conventional aspects of swimming (lines 32-37) and thereafter, offerer terminates the report with a component fitted to the start of the narrative (lines 38-39 'I left there at uh (0.5) exactly at three o'clock.', cf. line 1 'I left the restaurant exactly a quarter to eight,'). Second mention, then, consists of disattention followed by disaffiliation, with carrier-topic abandoned, as in first mention, with closure of the report (cf. Fragment 23, lines 15-16).

Third Mention Innocuous, possibly allusive reference; recipient appreciates. As at the end of second mention, third mention's reference is innocuous. It occurs in the course of a story in which it serves as a partitioning device. That is, mention of swimming provides a situation in which two of three otherwise co-present story characters are talking privately. 30



(29) [NB:PT:17:r] [Time:13:45-13:55] 1 2 3 4 5 6 7 8 9 10 11 12 13


en I told i m - uh so when she wen' t'the re:stroom I siz boy there goes a (.) great gal'n 'e siz boy I sure l:love 'er 'n I hope I g'n make'er happy so (.) when-'hh we came home why he wentuh bed'n then we went swimming again,

Emma: Lottie: Emma: Lottie: Lottie:

,°Mm hm,°, 'fore we(h h)nt


I told Isabel'e said'at 'e sezh yAw yer a liar. I sz well no: t h a t - h e said . . .

t ' h h h h

= Yeah. h h S o hh (0.2)

Like 'she went to the restroom' (line 1), 'he went to bed . . . we went swimming' (line 4) operates as a partitioning device for an ongoing story. The latter partition's two elements are themselves separated. 'We went swimming' is introduced as a discrete event with 'and then' (see note 28), and is followed by a silence (line 5). That is, it is set off as, not merely a story-partitioning device, but a respondable on its own. Further, it is tied to the prior night's 2:00 a.m. activity with 'again'. 3 ' The combination of features (that it is set off from the current story, and that it is tied to a prior mention) may constitute a recognisable allusive evocation of the manner in which that activity was explicitly said to have occurred; i.e. 'in the nude.' After the momentary break, both parties start to talk simultaneously; recipient with a minimal acknowledgement (line 6) which treats the reported activity as part of the story and prepares to hear further story events (cf. e.g. second mention, Fragment 27 lines 8-20), teller with a continuation which preserves focus on the reported activity and invites laughter by reference to it (line 7). Again, then, two alternative directions are taken; recipient, hearing in the silence that a comment is due, provides a comment which disattends the possible allusive aspects of the respondable, while offerer, hearing in the silence a potential declination to respond, preserves reference and invites response. While recipient's next utterance declines to laugh (line 8), it constitutes a shift in response-type; i.e. it now appreciates the activity as a discrete respondable. In its details, the response may acknowledge the allusive character of 'we went swimming again'. The appreciation is initiated with the terms with which 'in the nude' had been listened for (while officially disattended) and the assessment term is that with which offerer had appreciated the impropriety ("Oh God . . . fun' cf. F. 27 lines 21-27). The recur-



rence here of objects which had figured in the delicate negotiations by reference to an explicit formulation, occurring as they do here, by reference to a possibly allusive reference, may, then, allusively appreciate the activity which has been invoked. Offerer starts to laugh at a 'recognition point' in recipient's assessment term. 3 2 In that the laughter is produced by reference to a possible appreciation of an alluded-to impropriety, it may potentially activate a laughingtogether in the course of which affiliation may eventuate. However, perhaps by reference to a series of factors (that there is a story in progress, that the impropriety here is merely alluded to, and that recipient, by stretching the assessment term across offerer's laughter (lines 8-9), is recognisably again declining to laugh), offerer consecutively (1) terminates the invitation to laugh together, with an acknowledgement token (line 10),33 (2) indicates preparation to return to the story's telling with 'So' (line 10, cf. Fragment 11 line 17), and (3) provides a place for recipient to opt for laughter (line 11, cf. Fragment 11 lines 17-24). In this case, then, the possibility of a laughingtogether is offered, the offer terminated, but the possibility left open. That is, it may specifically be displayed that it is upon recipient's option that the story be continued now, or, alternatively, that the discontinuation for appreciation of 'swimming again' and its allusive reference to swimming in the nude be expanded. Recipient declines the option to expand (line 11) and offerer returns to the story's telling with no further reference to swimming (lines 12-13 and subsequent data not shown). Third mention, then, consists of innocuous reference which, over a series of negotiations, may achieve the status of allusive reference; that allusive reference itself allusively appreciated, and the carrier storydiscontinuation is abandoned with a return to the ongoing story out of which it emerged.

Fourth Mention Familiarised impropriety, recipient appreciates. Fourth mention appears to rely upon second mention's listened-for (although officially disattended) explicit reference and third mention's allusively mentioned and allusively appreciated reference. That is, fourth mention treats the nude swimming as a matter of record between the current co-participants to this conversation.

L A U G H T E R IN PURSUIT OF INTIMACY (30) [NB:PT: 19:r] [Time: ca. 15:45-16:45] Lottie: Lottie: Lottie: Emma: Emma: Lottie: Emma: 9 Lottie: 10 Emma: 11 Lottie: 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Emma: Emma: Lottie: Emma:

O h : G o d what a house. You have no idea. (•) I m e a n it wz jist (.) furnished yihknow = = you,jus' w.alk irr to it'n, 'hh ih wz a model home = Yeah. jl°M m h m,°j with all the furni j ture 'n evry,thing 'n, hhhhh j : t ' s jis:t = ° M m hm - oh yihknkow liddle (.) the flo.wer p o t s ,: = Is t h e s w i h = now'n t h e , : n they d,o, lo Yea:::h.° Yah,° = = h h h h e h : Is t h e swimming pool enclosed with the th'gla:ss bit? = = N o : : , it's u h : o u : t s - ( . ) eh no outside the big (.) j u h : : : : j °Mmhm,°


Lottie: Emma:

gla:ss Ah:hah,°


u - I g o t that w o : n g ,


' " (•) Emma: O h that's, that's.o.kay,. hhhhhhhhh, Lottie: m But the wahter is, eighty, fi:: :ve. j Emma: o h I k n o w it. = Emma: = Isn'it gorgeous, = Emma: )0 lf( ] But yihknow when yih git out it's kin'a co:ld. = Lottie: Emma: _ j[(Oh: o h : ) y a , : h . = Lottie: Wul ih was, Lottie: two uh'clock in the morning en,then.las' ,°night° j Emma: °huh haw h awh ha :w. = Lottie: nhhhh ,hn-hn-hn, _ Emma: = n;O o I ( h ) b e t that w'z Emma: ,(fu:n.) 35 Lottie: with n o : k - ' h h , h h 36 Emma: hhhjhh 37 Lottie: clozejOn.God it's good. , _ 38 Emma: ((f)) aa aaaaaaaaaaaaaa 39 Lottie: ,hu-uh h u h . h u h h , hh 40 Emma: Isn't that e x e r t i n g , 41 42 43 44 45 46 47 48

Lottie: Emma: Lottie: Emma:

(•) Uh:.:? O h : that's wonderfj'l, O h : : G o d we had. W e , I never had so much fun m.y li:fe., O h : I ' m gladju w e n t . G O D Lottie I wish you c'd meet somebuddy like that. (0.4)



182 49 Lottie: 50 51 52 Emma:

G o d I wz tellin Isabel we talked a b o u t the thing yihknow en she siz well it's probly ih lotta yer fault too you sh'd m a k e u p t'him en e r r y t h i n g e n ' « wpii t j u s ' c a : n't.

As in first mention, the impropriety is occasioned by recipient's question (lines 10 and 13, cf. Fragment 23 line 3). Unlike first mention's precipitate introduction (Fragment 23, line 5), the impropriety is approached gradually, via talk in which the two participants are comparing notes on matters familiar to them both (lines 23-28). The talk shifts from the general (line 27 'But you know when you get out it's kind of cold.') to the particular (lines 29-20 'Well it was, two o'clock in the morning'), with the time element which locates the first night's nude swimming (cf. Fragment 23, lines 5-9 and Fragment 27 lines 20-23). The temporal locator is followed by recipient's laugh-burst, a series of particles whose shapes are recognisably 'hearty' (line 31). By producing such a laugh for such an object, recipient can be demonstrating that the reference so far is thoroughly sufficient, and, as a correlate, that more explicit reference is unneccesary. In overlap with recipient's 'hearty' laugh, offerer produces an introduction to a next, separate event (line 30 'and then last night', see note 28). The event, however, goes untold; 34 there is a marked drop in amplitude in the introduction's last component ('°night°'), and a momentary silence by offerer as recipient's laugh is completed (lines 30-31). The discontinued next event is replaced with a laugh (line 32). This laugh is not fitted to the prior laugh (cf. Fragment 23 lines 9-10 for intensely fitted laughter). Rather, it appears that the particles are of a shape which regularly occurs in anticipatory laughter. 35 Thus, although laughter is followed by laughter, two distinctive and alternative activities may be occurring; recipient 'heartily' appreciating a prior and thereby proposing that no elaboration is necessary, offerer anticipating and thereby projecting the immenent occurrence of, something more. The anticipatory laugh occurs in overlap with an assessment by recipient (lines 33-34), which, although it is appreciative, tends to disaffiliate. That is, it is a prototypic no-access assessment (in contrast, e.g. to the prototypic access assessment vis-à-vis the heated water at lines 23-25) (see Pomerantz, this volume, Chapter 9). Latched to the anticipatory laugh, and still in overlap with the assessment, offerer produces the anticipated 'kicker' (lines 35-37, see pp. 125-26 and note 23) 'with no clothes on'. This phrase departs from that used so far ('in the nude', cf. Fragment 23 line 5 and Fragment 27 line 23), and the informality it displays is particularly appropriate for the familiarised manner in which the talk is now being conducted.



T h e impropriety, produced in overlap competition with recipient's no-access assessment, appears to have successfully overridden it. Specifically, recipient who has indicated that no explicit reference to the offence is necessary, w h o has issued a no-access assessment for the allusively referredto activity, now, for the first time, officially hears, acknowledges, responds to the i m p r o p e r c o m p o n e n t (lines 35-38) with a falsetto semi-shriek, semisong. T h e impropriety is now officially received. H o w e v e r , the reception may be equivocal; it might constitute some version of appreciation and it might express shock (cf. Fragment 5 line 3 for an a m b i g u o u s lexical). Perhaps by reference to the equivocal status of the i m p r o p r i e t y ' s reception, offerer pursues an especially appreciative assessm e n t across overlap with the response. The 'it'-formed assessment 'God it's g o o d ' (line 37) is of a type found routinely, not in the reporting of an activity, but in the partaking, and in the inviting of another to co-participate. 3 6 O f f e r ' s invitational assessment and recipient's equivocal response to t h e impropriety are co-terminous (lines 37-38), and simultaneously therea f t e r , o f f e r e r invites laughter (line 39) and recipient provides an assessment (line 40 'Isn't that exciting') which disambiguates the prior response. As in s e c o n d m e n t i o n , one party to an overlap appears to be giving clearance to, listening for, a projected key word. In this case, offerer's laughter stops at d u e - p o i n t for recipient's assessment term (lines 39-40, cf. Fragment 27 lines 2 1 - 2 4 ) . U p o n the occurrence of the assessment term 'exciting', the impropriety is, for the first time, unequivocally and officially appreciated. T h e r e a f t e r , while an opportunity may be available and offered to recipient for affiliation (lines 41—42), recipient provides another appreciative assessment (line 43) and may thus indicate disinclination to proceed f u r t h e r in the affiliation sequence. 3 7 Perhaps by reference to recipient's indication of disinclination to proceed, offerer provides a standard closing assessment (see Jefferson, 1978), initiated with a series of restarts (lines 4 4 - 4 5 ) which arrive at a pronoun replacement ( T for 'we') which may be the p r o d u c t of a sensitivity to the fate of second mention's reference to the n o n - p r e s e n t third party (cf. Fragment 27 line 20 and lines 26-28). While offerer's closing assessment may, via its positioning, refer specifically to the nude swimming, recipient's subsequent closing assessment, while it appreciates offerer's prior, expands reference in such a way as to e m b e d t h e improper activity into, as but a sub-event of, an ongoing report of t h e e n t i r e s o j o u r n (line 46 ' O h I'm glad you went.'). It thus tends to select i n n o c u o u s aspects of the carrier-topic, as does the talk following the impropriety in first mention (Fragment 23 lines 10-11). A n d , as in first mention,



recipient latches with a topic switch (lines 46-47, cf. Fragment 23 lines 11-12), which offerer accepts (line 49ff, cf. Fragment 23 line 15-16). Fourth mention, then, consits of an official appreciation of the explicitly formulated impropriety, with an indication of disinclination to proceed further, and the carrier-topic mutually abandoned.

Fifth Mention Re-issued impropriety, recipient affiliates. Offerer escalates, recipient appreciates and affiliates. Offerer re-escalates, recipient disaffiliates. (31) [NB:PT:51:r] [Time: ca.39:35-41:58]

The transcript fragment starts in the midst of an elaborate description of the house in Palm Springs (see note 38 for prior data). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Lottie: Emma: Lottie: Emma: Emma: Lottie: Emma: Lottie: Emma:

O h : : : yea: :h en the sisunken ba:thtuh I never took a ba,:th, _ ~ " ( ) = becuz th.e pool's ,so clean. = "huh huh o 1 kno:w.°j T h e sunke n baa-'hhh en then the toilet's way off in a.liddle , lo Yeah,o) = cubbyhole.yihkno.w en with a o h : : . G o, °Yeah° With the la vatories in the bedroom I spoze with (.) basins in the bedroom, (.) Lottie: Yeah let's see she',s got, °hmhh c Emma: ( )' (0.4) Lottie: four bathrooms. (0.2) Emma: °OhmyGhhod.° (0.2) Lottie: e - e n : : she eehh (.) She's a grea: t person a'run arou:n naygid yihknojw hhuh! • hh Well yih can there nobuddy yihknow great big walls nobuddy kin: see over'r anything.yihkno: w,. _ l (Emma): (° " " Emma: _ ,,heh-,heh-,heh Lottie: ~ l l h h h l k - l Christ yih kin: hh (hh)en so hh when Dwight le:f tihday we took off ar s - hh suits yihknow en, eoh en she gave me the (.) most beau:tiful swimsuit chu've ever seen in yer life. (STORY O M I T T E D ] Lottie:'sjist beautiful. = Emma: Mm hm, Lottie: = So then when Dwight le(h)vwe(h)e took the suits off(h) en swam


a r o u n ' n t h e . n u : ,ne. =

32 (Emma): 33 Lottie: 34 Emma:

okh1 = h h u h l ' n t o o k a sunbath in the n u d e ' n errything.


hhh.hh Y o u know


Elly'n I usetih d o that on the rivers if th 'fellerd go down get

36 37 38 39 40 41 42 43 44 45

g a : s ' l e e n fer their b o a : t s hh hh sh'd say dih you m i : n d w e ' d b e inna cOr. ve? b u t , w e ' d take it o u : . t , under ,th'wahter.= l 1 (°Yeh°) 1 (°Oh yeh 0 ) 1 = y i h k n o w becuz uh, (.) e e - w i r o u t in t h e o ^ p e n yihknow, h h h B u h w e ' d jis slip ar bathing suit ow, en g - e n swim a r o u n d in that r n v e r t h a t uh C o l o r a d o River til, h h h h ( . ) G h h o d what a thrill. (0.2) I always have like'tuh swim in the nu^de.j M e : too yihknow =

(Lottie): Emma:

Emma: Lottie:

4 6 Emma:_ , , ' h h h h 47 Lottie: e h w u h - hh en then hh right eh theh t h e r e ' s two placiss where 48 49

t h ' h o t w a h t e r c o m e s in'n yih g'n git rhight u p close to'm'n yis feels like y e r . t a , k i n . a ,dou.:che.. eh l -uh l -uH -uh 1 - a h - "

50 Emma: 51 Emma:

= ahh.-ahh

52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

hhhHHU:H HHUH H H U : HlHA:hha e - u - e ah ^ a h ' _ e h, = E.n we-, 1 1 X ' N S E E Y O U T W O j a : D ( S ( )| Ensh ewzononeen'nlwzo'th' o t h e r en'with ur legs ub y'know = = e n , J e e y i ( h ) s , h h h h shho . g h h o o j d . h n a ^ h ha^h, _ [ '"Oh:::::"1 °Go:::dls she [ c u : t e,° = hu-.uh-hu, O h : she's a cutey. = =0.h:G O : D she's uninhibitideh, (.) hhhh.hhhhh.She.'s, l e-Ye:la:h.' (0.2) Ye:.a:h.,

Lottie: Lottie: Emma: Lottie: Lottie: Emma: Lottie: Emma: Lottie: Emma: Emma: Lottie: Lottie:

68 Emma: 69 Lottie: 70 71 72 73 74

Emma: Lottie: Emma: Lottie:

75 76 Emma: 77 78 Emma: 79 Lottie: 80




. H U H - H A . - H A - A H h . - a h h.ah.agh.uh.

: that's w o n d e r f u l Lottie.l lo

' m , . s : s o happy, Yeah.°1

(.) t ' h h h . e n I've h a d d a .real, good t i : m e too = '"(That's good) 0 = I: thought,I w'z .gonna(m)"Good. (0.9) I: misstche, b'd I, I m e a n evry day's been a n h c e day. " (0.3) ~ ^ T h e r e ' s a : 11— , D o n ' t you w a . n t me t'come down getche t ' m o r r e n take yih down tih the beauty parlor?

. hhuhhh



Fifth mention occurs approximately 40 minutes into the call, at a temporal distance of some twenty-three minutes from fourth mention. 38 The impropriety is offered as if anew (lines 24-25 and lines 30-33, cf. Fragment 23 lines 6-10 and in particular Fragment 27 lines 18-23 for an in-the-courseof-events introduction, in contrast to Fragment 29 lines 27-37), and is received as if anew, with a response-type of a different order than those which had so far been offered, a 'second story' (lines 34—42, in contrast to Fragment 23 lines 9-13, Fragment 27 lines 22-28, Fragment 29 line 8, and Fragment 30 lines 3(M3). 3 9 Although fifth mention is distanced from the other mentions, and although the impropriety is offered and received as if anew, the story which receives it is fitted to prior mentions in range of ways. For example, as did the prior mentions (except the precipitous first and the familiarised fourth), the story indicates the absence of men (lines 35-36, cf. Fragment 27 lines 16-20, Fragment 29 line 4 and Fragment 31 lines 24-25 and 30-31). For example, as recipient's selection of non-present participant to the event for assessment can have indicated, it matters that the activity was instigated by another (line 35, cf. Fragment 27 lines 26-28), and, for example, attention is paid to whether anyone could see (lines 37-39, cf. lines 20-21). In its course, the story selects prim aspects of the activity and of prior mentions to the activity. Further, it is introduced as something its teller 'used to do' (line 35), which, in general, claims that one does not do it anymore. Thus, in various ways, the story tends to disaffiliate. While the story's assessment (lines 41—42) tends to affiliate, its placement is problematic. That is, while there is explicit reference to taking the suits off (line 40), the activity itself is not 'swimming in the nude', but an innocuous version of that 'in the X' format, 'and swim around in that river' (lines 40-41). Further, the occurrence of 'that' marks a possible topic (cf. Fragment 27 lines 20-21 and note 22), with the topical possibilities enhanced thereafter (line 41 'that uh Colorado River'). The assessment, then, might conceivably refer to swimming in the river, not to swimming in the nude. A n d it appears that this possibility is relevant for the interaction. In general, response to an assessment occurs very quickly, with no gap or minimal overlap. 4 0 In this instance there is a silence (line 43), within which storyteller (recipient of the initial impropriety) may be awaiting, and story-recipient (offerer of the initial impropriety) withholding, a response. And in the absence of response, storyteller offers an unequivocal affiliation to the impropriety (line 44). At precisely the point at which affiliation has become utterly unequivocal; i.e. at the point in the 'in the X' format at which it is not anything



else but, recognisably, 'in the nude', response occurs (line 45, see note 32). T h e response takes the form of a claim of affiliation (cf. e.g. Fragment 18 line 8 and 12). The claimed affiliation may be well designed by reference to its prior. That is, to that point offerer might have been, and has been treated by recipient as an initiate to nude swimming, with non-present coparticipant as instigator. Similarly, the story presents recipient in such a position. Now a claim of longstanding, ongoing enjoyment (if not practice) of the activity is made. Thus, an escalation has been offered (by recipient of the initial offence) and its recipient (offerer of the initial offence) has affiliated. Latched to the affiliation is a precipitously initiated escalation 4 ' by the initial offerer (lines 45-49, cf. Fragment 23 line 5 for precipitous talk). Its language is designed for two parties familiar with the phenomenon (cf. Fragment 30 lines 23-34). And it is followed by a perfect display of intimacy. The escalation gets laughter which is particularly exquisitely constructed, and in the following text it is decomposed for careful inspection. T h e improper component is first anticipated with laughter by its recipient (arrow 1) and thereafter appreciated with escalated laughter (arrow 2) (cf. Fragment 23 lines 8-10 and p. 173). 49 Lottie: 'n yis feels like yer.ta ,kin , a ,dou.:che.. 50 Emma: Jeh - u h - u h - u h - a h - a h h < (1)


Offerer joins the laughing-together with a next escalation (arrow 1) (cf. Fragment 23 lines 8-10). 49 Lottie: ,ta .kin .a .dou. :che. . 50-51 Emma: l e h - u h 1 - u h - u h - a h - ' a h h . - a h h ah 52 Lottie: 'hhhHHUiH 1

Recipient initiates a laugh termination, with a de-escalated particle (arrow 1) and an inbreath (arrow 2) while offerer continues in the escalated register (arrow 3). ^ ( 1 ) fc.—(2> 51 Emma: ahh.-ahh ah. hhh . 52 Lottie: hhhHHU:H HHUH

Recipient's potentially laugh-terminal inbreath becomes a precontinuation inbreath (see note 3) and renewed laughter is produced in the form of an escalation of recipient's own prior laughter (arrows 1) which is



fitted to offerer's escalation (arrow 2) (cf. the fittedness in Fragment 23, lines 9-10). Offerer is continuing in same register (arrow 3). ^

51 E m m a : ahh.-ahh a h . 'hhh , H U H 52 Lottie: 'hhhHHU:H^HUH'HHUiH (2)


Immediately after recipient's fitted escalation (arrow 1) there is a re-escalation, by recipient (arrow 2). ^ ( 1 ) 12) 51 E m m a : ahh.-ahh a h . hhh .HUH-HA.1^ 52 Lottie: ' h h h H H U : H1 H H U H ' H H U : H J

Immediately thereafter, offerer matches recipient's re-escalation (arrows 1). 51 E m m a : ahh.-ahh a h , hhh . H U H - H A . 52 Lottie: h h h H H U :H H H U H H H U : H H A : h N o

Recipient continues in the same register (arrow 1) while offerer initiates de-escalation (arrow 2). Recipient follows suit (arrow 3).

51 E m m a : a h h . - a h h - a h . hhh . H U H - H A - H A - A H h . - a h h^ [ 52 Lottie: h h h H H U : H H H U H l H H U : H 1 H A : h ha l e - u - e 1

Thereafter, offerer fits the shape of the de-escalative particles to recipient's (arrows 1). Each produces a terminal particle (arrows 2), and the display of hearty laughing with each other is marked as completed with an emphatic inbreath by recipient (arrow 3) (see note 2). ^(D



I 51 Emma: a h h , - a h h - a h . hhh . H U H - H A . - H A - A H h . - a h h.ah.agh.uh. hhuhhh l 52 Lottie: hhhHHU:H HHUH HHU:H H A : h h a l e - u - e ah : ah1 :e h ! V )

V )

Both start to speak, almost simultaneously (lines 53-54, cf. Fragment 8 lines 3-5); offerer with an indication that the report is continuing ('And w e - ' ) , recipient affiliating to the escalated impropriety.

LAUGHTER IN PURSUIT OF INTIMACY 51 Emma: 53 Lottie: 54 Emma:


hhuhhh = = E,n we-, 1 I C'N SEE YOU TWO KI: DS

This affiliation is of the type considered throughout the chapter, in contrast to the 'second story' at lines 34—42. However, it may be characterised as a weak version of affiliation, presenting recipient as merely a witness (although willing and able), rather than participating; that is, it is closer to the claimed affiliations seen in Fragment 18 at lines 8 and 18-19 than the displayed affiliations in Fragments 17 line 4,19 line 7,20 line 18 and 21 line 9. While it affiliates, it carries some potential for estrangement. It turns out that the escalated impropriety is a prefatory abstract, the details of the scene yet to be explicitly depicted. Having started in overlap with the affiliation (lines 53-54), having cut off to permit recipient's talk to continue in the clear (lines 53-54), offerer restarts at a completion point in that utterance (lines 54-55). Both the initial and the restart are formed as continuations of prior talk, specifically as intra-segmentally linked to an ongoing course-of-events description (line 53 'And we-', line 55 'And she was on one end . . .', cf. note 28). That is, both via its placement and its lexical components, the explicit depiction of the scene is produced as a continuing part of an utterance which had been discontinued for laughter; the laughter now serving as a pivotal appreciation/anticipation. But while offerer is intensifying the intimacy, recipient is activating the estrangement potential of the prior affiliation (lines 57 and 58). That is, simultaneously offerer produces an intimate assessment ('and Jesu(h)s it felt so good') followed by laughter, and recipient produces a sotto voce assessment like that in second mention ('°Oh God is she cute,°' cf. Fragment 27, lines 22-27), once again declining intimacy by assessing the activity via its non-present co-participant. In various ways this segment is similar to second mention. Latched to recipient's declination (both to laugh and to affiliate), offerer pursues the inivitation to intimacy (line 59). In second mention the latched continuation is the assessment term 'fun' (Fragment 27 line 27) and in this fragment it is a de-escalated laugh unit. 42 And, as in second mention, the pursuit of the invitation to intimacy is met with pursuit of declination (line 60). While in second mention the declination consists of initiation of a potentially topic-shifting question which also makes a connection between nude swimming and drinking (Fragment 27 lines 27-28, cf. note 22 and p. 178 and note 27), in this case it consists of an escalated recycle of the prior assessment (see note 37). The diverging treatments of the escalated impropriety (offerer's pursuit of intimacy versus recipient's pursuit of distance) are now into a second,



recycled round (lines 57-58 and lines 59-60). And, at least inasmuch as laughter is extendable, the fact that offerer stops laughing while recipient continues talking (lines 59-60) and, that offerer, having started to talk on completion of recipient's reiterated assessment (line 61) cuts off as recipient starts (line 62), indicates that offerer is prepared to yield to recipient. And there may be specific grounds for offerer to yield. The observable similarities between this fragment and second mention may be available to the participants. And in second mention, offerer's pursuit of intimacy (Fragment 27 lines 25-27) was met with the disaffiliative topic switch 'She still drinkin her little drinks?' As it turns out, the utterance at onset of which offerer has cut off (lines 61-62) is a milder version of second mention's disaffiliation. 'God she's uninhibited' is milder sequentially, in that it does not offer a potential new topic but maintains focus on the carrier-topic, and is milder interactionally, in that it is not an indictment but a compliment which, however, carries a sense of estrangement/disaffiliation. 43 There is a momentry silence (line 63, see note 40) which both participants move to resolve; recipient with a start on still a next third-party assessment (line 64), offerer with an acknowledgement token (line 65) which, by addressing recipient's prior utterance indicates that the pursuit of intimacy is relinquished (a version of the deployment of acknowledgement tokens in situations of overlap, see note 33). A n d , perhaps in part by reference to the fact that offerer has relinquished, recipient discontinues the next assessment. 44 The two utterances are almost co-terminous and a brief silence ensue (lines 64-66). And again, almost simultaneously, both parties occupy the silence with talk; offerer with a recycle of the prior acknowledgement token (line 67, see note 37), recipient with a closing assessment which officially abandons the series of disaffiliative third-party assessments but also proposes that the carrier-topic be terminated (line 68). Thereafter, recipient produces a topic shift (line 71) which offerer accepts with a re-issue of an offer made prior to introduction of the carrier-topic (line 79, cf. Fragment 61 lines 6-11, note 38). An overview of the materials in this section shows an enormously elaborate working out of a simple procedure: A target impropriety (the obscene play at the hot water inlets) is arrived at via introduction of, and after recipient-affiliation with, a lesser impropriety (nude swimming), and is introduced at just the moment that an optimum condition for its introduction has been achieved. It is in fifth mention that the optimum condition is achieved, with a base sequence: Impropriety followed by Affiliation. The affiliation is followed immediately by an escalation (the target impropriety),



which generates an expanded sequence: Impropriety followed by Appreciation (pre-affiliative laughter) followed by Affiliation. This sequence is followed by a re-escalation which, it appears, has pushed the intimacy too far, and a next sequence consists of: Impropriety followed by Disaffiliation, w h e r e u p o n the carrier-topic is abandoned as in each of the prior mentions. T h a t is, although the target impropriety is elaborately arrived at and its delivery is initiated at an optimum moment, only the prefatory abstract receives affiliation. The explicit description does not receive affiliation (nor any of the response-types on the proposed continuum which are empirically observable precursors of affiliation (declination to respond, disattention to i m p r o p e r component, or appreciation). And, subsequent to the occurrence of disaffiliation and abandonment of the carrier-topic, there is no further pursuit of affiliation. The conversation is terminated approximately two minutes later with no further talk about swimming, nude or otherwise.

Notes to Chapter 6 1. For an earlier pre-publication draft of this article see Jefferson etal., 1984. 2. Rather than display a series of post-laugh inbreaths followed by no further laughter, two fragments are shown in which an orientation to laugh termination is displayed. In the first fragment, a two-party laughing-together has co-terminous laughter (arrow 1) followed by inbreaths (arrow 2), one of which is louder and longer than the other. A t completion of the latter inbreath, the one who stopped first starts to talk. (32) [SBL:2:2:3:52:r] S

1 Chloe: 2 Claire: 3 Chloe:


.huh heh heh heh eh eh eh eh "eh. hhhhh ^ uhh 'uhh uhh # u h h uh uh uh hhhahhhhh = = Bill looked at me helpless(h)ly

In the second fragment, two parties are laughing together; one stops (arrow 1) and the other produces a pair of louder, higher pitched particles (arrow 2). A brief silence occurs. The party who had extended his laughter takes an inbreath (arrow 3) and upon its completion the other starts to talk. In this fragment it appears that the possibility of still more laughter was relevant, and the inbreath was treated as the object which announced termination. (33) [NB : ITB: 8) 1 2 3 4 5


Fran: Ted:

.ihh-heh-huh-huh-huhj ^ , ; neh-heh:huh-huh-huh -Thuh-Thuh

Ted: Fran:

° = (31 hhhhnh, ^ = She wz inna hu:: rry,

S e e also Fragments 7, 8, and 31 (lines 51-54).



3. O n page 155, see note 2, it was proposed that inbreaths after a burst of laughter are oriented-to as laugh-terminal objects. It is the case for these objects, as for other oriented-to features of interaction, that they are not inevitably and automatically that. While an inbreath can mark termination of laughter, and is routinely oriented-to as marking termination of laughter, it can also constitute the catching of breath to laugh some more. Thus, a laugh which may have been intendedly completed, marked as completed with an inbreath, may retroactively be renewed, depending upon contingent activities. 4. A weak catchall term, 'impropriety', is used throughout the chapter to identify the various instances of the range of interactional breaches under consideration. 5. Throughout the chapter, participants will be identified in terms of this invitational characterisation of impropriety. The speaker who produces the invitation to intimacy is identified as 'offerer' and the co-participant to whom the invitation is directed is identified as 'recipient'. 6. Appreciation provides no explicit indication that the one who is appreciating another's utterance is implicated in the mentality, situation, etc., exposed by that utterance. This is evidenced in the following fragment in which an assertion is appreciated with laughter (lines 11-13). The appreciative laughter is followed by an explicit query as to laugher's status vis-à-vis the asserted situation (line 14). (34) [TC:II:14:excerpt] Two men, complaining together about an acquaintance's shady business procedures. 1 2 3 4 5 6

J.R.: Seth: J.R.: J.R.: Seth:


8 9 10 11 12 13 14 15 16 17

Seth: J.R.: Seth: J.R.: Seth: J.R.: Seth: J.R.:

H e always comes out smelling well though. That's what gets m e : . We:ll?,(maybe it's) the way he treats'em. YihYehl, Igue:ss,I:,I'mnot,( ), l (It's) like thee: uh (•) two inspectors no:w er (.) great pals'viz. " (0.5) " Ohj? Well one o f m looks et iz pornographic movies. I've never seen a pornograhic movie (dis) in my li:fe. = = uhh hihh uhh Hevyou^? No I ha:ven't. We:ll, I'm ready any ti:me b't I've never been seen one,

7. While the first two fragments have the sorts of objects so far considered as types of impropriety (i.e. crude or obscene language or reference), the third is a matter of rudeness. Although presented 'tactfully', the rejection of an invitation, specifically on grounds of preferring one's own company to that of inviter, is a potential offence. 8. That an assertion about one party by another is not denied outright, but follows appreciation of the assertion seems to be a systematic occurrence. It may have to



do with such issues as, if the assertion is denied immediately, the denial may be seen as produced by reference to the assertion's interactive features (e.g. is declining to be intimate with the asserter) or by reference to the in-principle deniability of the assertion, rather than by reference to the assertion's actual truth or falsehood. A solution to such a problem is to interact about it first, indicate that it is not in principle a thing recipient would deny, and then deny it. For example, in each of the following fragments, a participant characterised as (1) miserly, and (2) a procrastinator, initially appreciates the characterisation and subsequently rejects it. (35) [Goldberg: 11:2:81 'hh Wul knowing you you'd have thirty one en, (.) thousan e:nd a nickel, hhh! (0.2) hheh-heh-heh-.heh-(heh) Shih yuh I think y'got the original nickel.

1 Maggie: 2 3 Gene: 4 5 Gene: 6 Maggie: 7 8 Gene: 9 ( ):

10 ( 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

(•) ' hh-heh-heh-heh-heh-heh-heh, = = 'hh = = 'hh =


Gene: Maggie: Gene: Maggie: Maggie: Gene: Maggie: Gene: Maggie: Maggie: Gene:-I Maggie: Gene:

26 27 Maggie:


= We're not puh- we're relaxed about it. = Penny: „We're not procrastinatihsehh-huh. , Pat: (That's it relaxed I didn't get it righ(h)t.

9. T h e participants are sisters in their early sixties. One of them has taken a trip to Palm Springs to visit a friend and her newly-acquired husband, also in their sixties. The phone call occurs on the night of the return home. 10. Specifically, taken up, rather than initiated. Recipient has already inquired about the house, and that inquiry gets 'Oh God Emma. Jesus how lucky. You have no idea' (cf. Fragment 23, line 1), and is followed by area-locational talk and discussion of how the host couple is spending the upcoming Thanksgiving vacation. 11. For some consideration of triggered or touched-off talk, see Sacks, unpublished lecture, April 17,1968, p. 16 (mimeographed), and Jefferson (1978). 12. While Fragment 13 is a dramatic instance, other possible declinations to respond can be seen, for example, in Fragment 12, in which recipient silence (line 3) eventuates in disaffiliation (line 6), and in Fragment 14, in which recipient silence (lines 6-8) eventuates in disattention to the improper component (lines 9-11). 13. Technically, continuation provides that the break is an intra-utterance pause, not an inter-utterance gap. See Sacks, Schegloff & G. Jefferson (1974). Interactional^., the continuation proposes the inconsequence of the break; i.e. no declination to respond has occurred since the utterance is still underway and response is not yet due. 14. While the silence which follows the impropriety (lines 1-2) constitutes and is treated by offerer as recipient's declining to respond (cf. Fragment 13), it appears that in this case the problem is not recipient's unwillingness to affiliate, but, specifically, an initial inability to find an appropriate affiliative response. It appears that in the course of offerer's continuation (line 3), an object occurs which solves the problem for recipient; i.e. the word 'car'. Just after the word 'car' has been uttered, recipient produces an explosive inbreath ''khh!' (line 4) and then and there launches the affiliative utterance. For a consideration of 'discovery points' in ongoing talk, marked by such things as inbreaths, see Jefferson (1978). 15. Once the retroactive cleansing sequence is initiated, it goes to completion. In Fragment 17 (lines 3-4) and first mention (Fragment 23 lines 6-9) it goes to completion in competition with recipient response. In the following fragment, recipient starts to appreciate just after onset of the continuer (line 4), offerer takes the continuer to completion and follows with a de-escalated alternative (line 6, in this case a description of feistiness). (37) [Goodwin :GR: 40] 1 Jan: 2 3 Jan: 4 Beth:

So I said [ook Gurney, yer just a big ass kisser, (074) en,yer getting yer wa:y, AAHh hah-uh hah-uh huh ~

L A U G H T E R IN P U R S U I T O F I N T I M A C Y 5 Beth: 6 Jan: 7 Beth:



,'hhhhhhhhhhh, 1(h) ju(h)st l a i . d i t a : ,11 on, hhah


9 Beth:

ehh huh uh-huh uh-huh

16. Anticipatory laughter can occur in a range of relationships to an anticipated object. For example, in the following fragment, laughter precedes a projected next utterance in which a particular component might occur (line 4 vis-à-vis lines 1 - 3 and 5). The utterance occurs but the component does not (line 5) and the one who laughed in anticipation now provides the anticipated component (line 7). (38) [Schenkein:II:36] 1 2 3 4 5 6 7

Bill: Ben: Ethel: Max: Lori: Ethel:

So you walked all the way up t'the Chinese theater. Didje g e t - like stop in a restaurant or something? Have summing t'eat? eh hee heejheh! Stopped inna Shell station. Where's the Chinesejrestaurant.j tihgopee pee.

A n d in the following fragment, anticipatory laughter starts before due-point (arrow 1). At due-point the anticipated component is produced simultaneously by ongoing speaker and anticipatory laugher (arrow 2). (39) [Frankel:US:98] 1 Joe: 2 3 Joe: 4 Carol:

B't he wannid duh dawg tuh bite iz wife. (0.4) So he C n m f l c hrtmp nTip nioht'n tlip cAnnfa hit^h Kit hi-m

17. Fragment 26 is an instance of 'recognition-placement' rather than due-point placement. For a consideration of recognition-placement as a systematic device, see Jefferson (1973). 18. While the escalated laughter in Fragments 26 and first mention may be initially seen as more-or-less at completion, it may be the case that it is targeted to occur precisely at completion but an ongoing word is stretched beyond its anticipated completion (Fragment 26 'c(h)e(h)ents,' Fragment 23 'clo:::ck.'). 19. Such laughter may have especially strong affiliative import. See, for example, Fragment 31 lines 49-54 and p. 46-68. 20. For a consideration of temporally organised course-of-events narratives, see Sacks, Lecture 9, November 3,1971. 21. For a consideration of recipient-designed place formulations, see Schegloff (1972). In this case, the 'that X' formulation may specifically invoke recipient's earlier expression of interest in the swimming pool (Fragment 23 line 3) and/or may invoke a mutual interest in the swimming pool apparently expressed in prior conversations (see Fragment 30 lines 13-22).



22. For a consideration of topical aspects of 'that X', see Maynard (1975). It appears that this formulation can mark not only topic initiation, as in Fragment 40 line 3 (a re-transcribed version of a fragment which appears in Maynard's thesis), but can mark a move into a sub-topic for an ongoing topic, as in Fragment 41 line 7. (40) [Zimmerman : TA : FT : alt : 41 1 2 3 4 5

Al: Al: Al: Bev:

hhhheh hhih-ih hhhh hh- hh-hhhhh hhhhhh hhhhhhhhhhh (1.4) Thatsna^ke wzkindaneatetwork the other da:y. = = Cz,lotta th'k,idshad'neverseenasna:ke? l Was it?1

(41) [NB:PWT:2] 1 2 3 4 5 6 7 8

Emma: Penny: Penny: Emma: Penny: Emma: Penny:

Oh honey that was a tavely luncheon I shouldaca:lled you s:soo,:nerbutl:,l: Jo:vedit. Ih wz justdelightfu . : l.,_ l ((f)) [ O h : : : : 1 U ( )° Well J _ = I wz gla.d y o u , (came), j 'nd yer f: friends 'r so da:rliMig, = = O h : : : , : it wz: , e-ThatP a:tisn'sheado:.:ll? . iYe hjsn't she pretty,

23. Although the activity has been mentioned before, in the same words, it is methodically offered here as a surprise, a 'kicker', the best of it. Routinely, kickers are positioned after a possible completion point in talk which in no way has projected that there is more to come. Two fragments are shown in which recipients treat a story as completed (Fragment 42 lines 5 and 7, Fragment 43 line 5) and are met with a kicker to the story (Fragment 42 lines 8-13, Fragment 43 line 6). (42) [Labov:CP:3] 1 2 3 4 5 6 7 8 9 10 11 12 13

Ardiss: Ardiss: Brenna: Ardiss: Brenna: Ardiss: Brenna: Ardiss:

. . . and then he just battered her to get her 'andbag. They said she was like a battered cucumber when they took her into the infirmary, Was she- uh = „AndtheIs she out, = =The.bestof itwa:s. '( ) Her husband's an ambulance dr- uh, a lorry driver, a driver for the, (the kashizin) infirmary, hh an' he had t'git the trolley'n wheel his own wife off the ambulance.

(43) [Labov:TA:4:r] 1 Rita: 2

She had a:, (0.3) abroijjedhambuhrger, (0.6) with no: (•) gravy awnnit, (0.5) she hadda serving of cabbage, 'n

L A U G H T E R IN P U R S U I T O F I N T I M A C Y 3 4 5 Marge: 6 Rita:


she hadda salad. ~ (0.3) Very- (•) It's terrific I because I'm tellin yih- , t n she c o u l d n ' e v en fini-ish:: i(h)t,

24. When deployed for overlap management, stretches tend to persevere until overlap is resolved by co-participant's either dropping out or reaching completion and stopping. (44) [NB:PWT:3] 1 Emma:

O h evrything's workin out so pretty he,re with ar,

2 Penny:

O h : : :

:: Tnnatgood.

(45) [GTS: 1:84] 1 Ken:

Heck .a lotta

2 Roger:


(46) [Rose:II:5] 21 Donna: 3 Karen:

En Donna, (•) T h a ^ t ' s what they sa:y,j

4 Donna:

w i l l s o o : : : : : : : : : : n learn.

(47) [Frankel:US:I:9] 1 Mike: 2 Vic:

. . . all th ,at junk is in the chair., W o : : : : : : : : : w I didn'know tha^t?

25. Although the elapsed time between the sentence parts is brief, the criterion for treating the talk as a series of separated activities is structural, not temporal. The following fragment is similarly structured, but the elapsed time between sentence-parts is substantial (lines 5-7), and thus transparent for the step-by-step competitive constructedness of the utterance in progress. (48) [HG:II:18] 1 Sandy:

Hadiyou feel, |tired?hh

2 Marna: 3 4 5 6 7 8

hh O

Sandy: Marna: Sandy: Marna: Sandy:

,,'hhh N O I wz very: (0.3) pleased thet I (^accomplish', j You really a ccomplishedalo:t.= = so much. = = What got intihyhhou.

26. For a consideration of skip-connecting in an environment of competing topical directions, see Sacks, unpublished lecture, April 9,1970, p.2 ff (mimeographed). 27. It appears that some activities have drinking as a prior activity. So, for example, in the following fragments, stories are told about 'mooning', a sport in which a car is driven around public places with one of the participants' naked buttocks sticking out the window.


TALK A N D SOCIAL ORGANISATION (49) [ G o o d w i n : M : l l 1 Cal: 2 (50) 1 2 3 4 5 6 7


Lenny: Cal: Bart: Bart: Cal: —> Cal: (51)

1 2 3 4 5 6 7 8 9 10 11 12 13

'n D o n n e g a n wz b o m b e d , well we were all b o : m b e d . Donnegan (said) we'll go m o o : n i n ' y i h k n o : w ?

Y o u w'r th'driver. " (0.4) N o : I w z : : m t h e f r o n t . s e a t H I : D ' ,n. ey(h)he u-he wz hidin'? = =heh,hahhohha:hha:hha-,u-huh-a-ha. , 1 V ) h h h ( h ) A h wz 1 : drunk = = b ' t ah wa(h)sn'that d r o ( h ) n k ,



—» ( ): Bart: Lenny: Bart:

((story about on the way h o m e from school on a bus)) the whole (0.7) u h : : : buncha the guys filled up the ba:ck there. (0.2) A n ' th'n evrybuddy else wz up front. (1.0) So w ( h ) e ' r e goin d o w : n uh- (0.9) I don'know some main highway out there I d'know where the hell it wa: s. (0.9) W e 'ad s'm b e e : r , (1.2) A : n uh (1.0) I ed mooned a lotta people b ' f o r e , t h a t . , ~ Slm-h'm,^ = Yihknow ah mean theh (.) th'guys knew it'n they, (0.2) Y a h , S t a : n to:ld ,me about it. j S'mbody said Grozak you h a v e n ' g o t t a h a i r o n y e r a : s s 'nless you (0.7) unless you moon that car buhind us.

28. That the talk will no longer be about nude swimming is displayed at onset of this utterance, with 'and then . . .'. Roughly, course-of-events descriptions seem to be subdividable into segments which can package multiple possible discrete events into sub-units of a single, larger event. So, for example, while departure from home and arrival at destination may include a collection of separable events, it is presented as a single unit with an intra-segmental 'and' (lines 1-4). O n the other hand, while departure from home and arrival at destination may be a single event, the arrival at the target city and arrival at the target household are presented as separate events with an inter-segmental 'And then' (lines 7-14). At second mention's line 30, 'and then' projects, and subsequent talk delivers, no further talk about the night's activities. 29. It is not that, following recipient's disaffiliative utterance, an innocuous event is reported, but that an improper event is reported innocuously. As it turns out, the activities reported as 'And then we swam all day today' were performed in the nude. In fifth mention, 'today's' swimming is reported as follows: (Fragment 31 lines 24-25) 'And so when Dwight left today we took off our suits you know and . . .' (the report is discontinued for a parenthetical story and then taken up again at lines 30-31),'. . . when Dwight left we took the suits off and swam around in the nude.'



30. For a consideration of these materials vis-a-vis storyteller's tracking of presences and absences of story characters, see Sacks, unpublished lecture 7, May 28,1970 (mimeographed). 31. A n examination of the entire transcript permits the following itinerary to be worked out: The 2:00 a.m. swim occurs on a Sunday night following Lottie's arrival. The events which form the story in which 'swimming again' is introduced as a partitioning device occur the next night. The all day swimming and sunbathing occurs on the following day, and on that same day, Lottie returns home. 32. In this case, that the laughter is recognition-placed (see note 16) may be utilising the prior occurrence of the word 'fun' to find that the word now underway is a possible recurrence, its recurrence enabling recognition prior to completion. It turns out that a standard locus of recognition-placed overlap is in rather more local occurrences of a repeated word. For example: (52) [SBL] 1 2 3 4 5 6

Still, when yer in a tour, traveling in a bus for six weeks, the same group, why that too is loh- plenty of opportunity I should think, I've never done it, Mm hm, But I should think that would be plenty of opportunity. l Yeah . . .

Bea: —> Maude: Bea: Maude: (53) [SBL]

1 Anne: —» 2 Joslyn: 3 4 Anne:

I'll bet she wishes she was a little more cooperative Yes, because she has less chance there then she would have if she'd of c o o p e r a t e d a little bit, ^ l Uh huh,

In third mention, the recognition-placed laughter may, by catching the recurrence of 'fun', catch also its possible allusiveness to the prior occasion of its occurrence i.e. the explicit mention of nude swimming. 33. A standard configuration for speech-speech overlap is a discontinued utterance followed by an acknowledgement token. With the acknowledgement token, a speaker withdraws his own utterance and retrieves the other's. (54) [SBL] 1 Martha: 2 Bea:

. . . because s h e w - you know, was^in the house^ so n e a r Yes.

However, the withdrawal may be temporary. It can be followed by a selfretrieval. (55) [TG] 1 Ava: 2 Bee:

3 4 Bee:

,, A'ri::ght?, I'll see wt's-

(•) Yeah. See what's going o:n.



And, crucially for the purposes of the current consideration, it can be followed by an other-retrieval by the initially retrieved speaker. In the following fragment, a Gaston-Alphonse other-retrieval series is resolved with a repetition by one speaker of the other speaker's term. (56) [Reilly] ((transcribed by Suzanne Reilly, University of Pennsylvania, 1973)) 1 2 3 4 5 6 7 8 9

Loren: Loren: Kate: Loren: Kate: Loren:

. . . people thet jus' come en leave their little kids et the skating rink's appalling. (pause) Y'know.drops them, dump them =Yeah. Yeah. Yeah dump. Y'know. These car pools just pour out five and ten kids.

34. There is a possibility that the event which goes untold at this point is the escalation which occurs in fifth mention (Fragment 31 lines 47-49). That it goes untold at this point may have to do with the fact that affiliation to the initial impropriety has not yet occurred. That it is almost told at this point may be the consequence of a trigger in the prior talk (see note 10). Specifically, there has just been talk about 'eighty five degree' water (line 23); i.e. reference has been made to the fact that the pool is heated. As it turns out, the escalation has to do with some playful obscenity at the 'two places where the hot water comes in' (Fragment 31, lines 47-48). The mention of the heated water may trigger an introduction to an event which is not yet to be told, and the introduction is withdrawn. 35. Following are some instances of 'hn'-formed anticipatory laughter. (24) [Labov:BG:5] 4 Doris: —> hn-hnn hn-hnn hn-hmnnn, = 6 Joan: = a load of shit. (26) [Goodwin:AD] 1 Bart: 2 Cal: —»

. . . 'n to< took mmhy fif, . . hnn-hnn-hnn

(57) [NB:PWT:9] 1 2 3 4 5 6


W'l you know i-yuh a:lmost say a-yih almost af:: :raid dihihh heh Inih = ' Emma: —> = nhh hHnnh -hhhn °hn Penny: Penny: =Dyih know uh y'know wuh I mea^n?' hh = Emma: =Ye:uh.

36. So, for example, in the following fragment, Ethel and Ben are eating herring, Max is not. The various assessments of the herring are attendant to invitations to Max to have some.



(58) [Schenkein:I:16:r] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

Ethel: Ethel:

MMmmm. (0.3) Ooo Max have a piece.

Ben: Ethel:

This.uh be), the bes.t you ever tasted. Gesch m a c h t.


MMmmm. (2.0) 't Oh it's delicious = = Ben w'dju hand me a napkun please,

Ethel: Ethel: Bill: Ben: Ethel: Ethel: (Max): Ben: Ethel: Ethel: Max:




Lemme cut up a'little pieces a'brea:d. (2.5) Is'nat good? = = It'sduh::licious. (0.2) It's geschmacht Max. (0.4)




Geschma::cht, (1.0) Max, one piece. (0.2) I d'want.

37. In general, repeated assessments (and other recognisably reiterated responses) appear designed to close down the talk by reference to which they are produced. The following fragment is excerpted from the same conversation in which the nude swimming story occurs. Lottie is describing the purchases she made at a roadside stand on the way home. (59) [NB:PT:7:46] 1 2 3 4 5 6 7 8


. . . en I got some casaba'n then I bought uhs:: uh Anna back a box a', Emma: Oh at's ni.ce. Lottie: yihknow. = Lottie: _ risheEmma: That's-nice Lottie, Lottie: sh'fed the ca:t,,'n, Emma: That's beaudiful.

In fragment 59 it appears that the specific sort of talk being attemptedly closed down is an explanation (here, of a gift to someone else — and it appears that no such gift was made by Lottie to Emma). In the following fragment, an explanation is received with a series of acknowledgement tokens.



1 2 3 4 5 6 7 8 9 10 11 12

Billie: Darlene: Billie: Darlene: Darlene: Billie: Bobbie: Darlene: Bobbie:

en I'm jis saying thet- thet any question thet we = =, theh we sh'd take. I,think, Okay. (0.7) [[O k a : y . , I think we sh'd take it from the group's perspective. (0.2) if,we're in,here tuh learn somethin. Right. (0.3) Y'know yer nah listenin tuh me.

38. Fifth mention is a last attempt to achieve affiliation before the conversation closes. Prior to fifth mention, pre-closing arrangements are underway (see Schegloff & Sacks, 1973), see below, lines 1-19. Offerer returns to talk about the trip in c o m p e t i t i o n with co-participant's efforts to close (see, e.g. lines 2 7 - 2 9 and lines 4 0 - 4 1 , in particular the repeated 'I'm so happy for you', twice overlapped prior to c o m p l e t i o n by more talk about the trip). In passing, a sequence similar to those n o t e d in note 36 can be seen, lines 16-24. (61) [NB PT:49:r] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

Emma: Lottie: Emma: Emma: Lottie: Emma: Emma: Lottie: Emma: Lottie: Emma: Emma: Lottie: Emma: Lottie: Emma: Emma: Lottie: Emma: Lottie: Lottie: Emma:

Well ah'll get that tih .morrow., You git t t h a : t , Y e a h . . I'll go up the = .drugstore t'mo.rrojw. hh Yee-° xih sure yih don'wan'me tuh come down'n getcha en take yih d o w ^ e : - j N o : sweetie = Ah:'llthe byu= hhjhh beauty jparl?) YouNo: you go getcher hair tfixed f if you wanna drive down'n see me ah'd love tuh see yuh, hhhj'hhh Okay well I gotta luh few thing Irgot- m.y blou:se ,s,tuh iron.e- Cuz I got, _ l I : : kno:w*::. l I : : J k n o : w . = Well uruu- run reas'n why I told Isabel I came 'ome cuss:: Thursdee'll be a lo:ng day.yihknow.'ll be, bout = Yeah. = Ye^ahj ni ne ten hours s,oYah, = hhhhh I wanduh I'm not a bit tire'tuhnight though. = °'hhohjhhhhhh° ((sounds like a yawn)) O h : : I'm so gladje hadda good ti: me, =

L A U G H T E R IN P U R S U I T OF I N T I M A C Y 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59

Emma: Lottie:

Emma: Lottie: Emma: Lottie: Emma: Emma: Lottie: Emma: Lottie: Lottie: Emma: Lottie: Emma: Lottie: Lottie: Emma: Emma: Lottie:


= hhl'.m so happy ", 0 EhE n Mondee I kept s:sleepin I thought Jeez I can't sleep'nymore I din'hear Isabel around er anything en finally et ten uh'clock I got up en I: si:z: (.) e*:: en then I yelled in et'er (tuh) Is'bel'n God shhhe'd p(h)in up ffer hou: rs yihdow en she's g' be,quiet fer Tme:.= °Oh° ~~ = h,hl °*Oh° = =hunh,'hh l °*blessherhea:rt.°Whadda *gal. (.) Well I'm so happy fo ^r yuh j °Ghhho d y o u s h ' d s e e " t h e b e : d s p r e a : d s you kno:w, en these g r e a t big hhhhh king si.ze(b) , ~ '"Mm: bedsyihknow = = with a^ll the(p) pillows on these gr::eatbigl:la:mpsyihkno:w jen hhhh Like a movie s: set. (0.3) Cheeziz e n - (.) en aMl the lights yihknow en the air c'nditioneen en goes o:n injthee: jUh:, Inter co: ml.suppose, u(.) uHu^h? (.) Intercom? (0.4) y'g'n talk-from one room,°t'the other?", hyuO h : : : yea :h en the s: sunken ba:thtuh

It is at this p o i n t that the segment designated 'fifth mention' in the text begins. 39. F o r consideration of the 'second story' as a device for showing understanding of prior talk, s e e Sacks, unpublished lecture, April 3 0 , 1 9 7 0 (mimeographed). 4 0 . For a consideration of the relationship b e t w e e n assessments and assessment r e s p o n s e s , s e e P o m e r a n t z (this volume, Chapter 9). T h e occurrence of overlap appears to b e a c o n s e q u e n c e of both participants' work; i.e. assessor extends the utterance slightly b e y o n d completion point, thus providing a broader target, w h i l e recipient aims at and hits the initial projected completion point. (41) [NB:PWT:2] 7 Emma:

e-thatPa:tjsn'sheado:,Ml? ,

8 Penny:

i-Ye h isn't she pretty,

(61) [Goldberg:II:2:5] 1 Maggie: 2 Gene:

W'my God it sound marvelou.s G e n e , , Yeah it is, it's a it's a good deal,



Sensitivity to the expectable occurrence of assessment-response within or upon completion of an assessment term may be seen in the following fragment in which a slightly stretched assessment term ('cra:p') goes to completion in the clear and is followed by a series of expansions. (62) [Goldberg: 11:2:8] 1 2 3 4 5

Maggie: Gene: Gene: Maggie:

W'wuh w'd that involve. More schooling? Oh:: yea: h. Jista bunch era: p. Y'know? Maggie en it's, (0.4) Iuh:: Wul is the money there though Gene tuh compensate, you?

41. The escalation skip-connects to the course-of-events narrative which preceded the second story (lines 30-33) with an inter-segmental link ('and then', see note 27), which offers a forthcoming event as a next in an ongoing series. In this entire forty-five minute conversation, the escalated impropriety is never explicitly temporally located. The 'and then' here might be treated as evidence that the event occurred as part of today's nude swimming and sunbathing. However, there is some indication that the event took place the preceding night (e.g. that was the big night out, with much drinking and friskiness, data not shown). The point is, one would not say with assurance that because it is introduced via 'and then' the activity took place as part of the course-of-events it is thereby proposed as linked to. In this case, the inter-segmental link may be deployed for local sequential work, as a way to provide the relevance of and account for the here and now occurrence of a report which has become interactionally appropriate to deliver, now, for the first time in this conversation. Its 'nextness', then, is by reference to local sequential considerations and not to chronological fact. 42. See note 24. In this case, the two discrete actions are latched. That is, while in Fragment 48 there is a substantial distance between them, and in Fragment 27 lines 25-27 there is very little distance between them, in Fragment 31, lines 57-59 there is no distance between them. While it is possible that the laughter simply started to de-escalate at some point, independent of other activities, it is also possible that the de-escalation, as a next activity, was initiated by reference to its overlapmate's arrival at completion. 43. While an assessment like 'uninhibited' can occur as an admiring compliment, when affiliation/disaffiliation is at issue the assessment addresses features of its referrent which do not belong to the assessor and thus tends to disaffiliate. Recipient is proposing that non-present third party, co-participant to the event, is someone who does things recipient does not (and perhaps would not) do. Attendant to the activity of estrangement from the assessed party, recipient provides an acoustic display of estrangement; i.e. the word is especially carefully, clinically pronounced; is produced as a layman using psychological diagnostic terminology to characterise alien behaviour. 44. Following is a candidate account of recipient's stopping just after initiation of a next third-person assessment. (1) The overlapping acknowledgement token by offerer is potentially not a complete utterance, but an utterance-initial term (see Fragments 41 and 42, note 39 for 'Yeah + ' in the environment of assessments). (2) The utterance initiated by the token might be an assessment fitted to the current series of third-person assessments (again, see note 39). That is, recipient



may stop by reference to the possibility that offerer is in the course of volunteering a type-fitted assessment; i.e. recipient may cancel a next of a series to permit a first response to prior components of that series, a response which may affiliate with, not merely acknowledge, the series of assessments.

7 Unilateral departure CHARLES GOODWIN University of South Carolina

Work on the analysis of conversation has demonstrated a multitude of ways in which talk spoken in conversation is intrinsically interactive. For example talk is exchanged through a sequence of turns at talk, turns themselves are constituted through the joint work of both recipients and speakers, and participants use the talk of others as a resource for the proper understanding of what is currently being said. 2 The present paper will analyse an utterance, an instance of what Goffman (1978) has called 'self talk', that lacks these features. 3 It is not organised with reference to turn-taking, and indeed is not situated within a speech exchange system. It neither responds to the prior talk of another or elicits further talk that is responsive to it. It is not addressed to another, and is not explicitly attended to by anyone else. Both its speaker and those in a positon to hear it treat it as a bit of talk that is irrelevant to the conversation in progress. The utterance thus provides an example of talk that appears to have no interactive import or organisation. However when we look at the production of this utterance in detail we will find that both the lack of attention to it, and its irrelevance to the conversation of the moment, are carefully organised through the actions of both its speaker and its recipients. In the analysis to follow we will begin by investigating how the talk is to be understood, focussing on the way in which this utterance is embedded within activities other than talk. Then we will look at what information the utterance provides to those in a position to hear it. Finally, we will examine the participation framework proposed by the utterance and the types of orientation it receives. Investigation of all of these phenomena will require careful attention not only to talk but also to a range of nonvocal activities that the participants are engaged in. In essence we will find that what is officially formulated as a unilateral, single party event in fact displays quite careful attention to others and is sustained through an ongoing process of interaction. 206



The data which are to be examined are taken from a videotape of a picnic held in the back yard of Pam and Curt. The adults present at the picnic have divided themselves into two conversational clusters. Curt, Mike, Gary and Phyllis are seated around the picnic table, and are the group that is being taped, while Pam, Carney and Candy are seated a short distance away. Analysis will focus on the talk of Phyllis in line 11:4 1 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.


Keegan usetuh race uhr u h - er ih was um, (0.4) usetuh runum, (2.7) Curt: Oh:: shit. (0.4) Curt: Uhm, (0.4) Curt: Fjsher'scar. Mike: Three en^na ^quarter? Curt: Thr,ee enna quarter. Phyllis: °Need some more irCe. [ Mike: Yeh, (1.0) Curt: (When I) wz foolin around. Gary: I usetuh go over there with my cousin (when he had a car),

Finding an Environment for Phyllis's Talk A first observation that can be made about the talk Phyllis produces is that it is not tied in any way to the talk around it. The subject matter of her utterance and the types of orientation it receives. Investigation of all of these (cars), and sequentially it is not a next utterance to some prior utterance and no subsequent talk is tied to it. Indeed it is begun at a point where not one but two other parties are already speaking. However, despite this it is not oriented to by either its speaker or the other speakers as a characteristic instance of overlap. For example, none of the parties relinquish prior to completion; no perturbations, cut-offs, or changes in volume or pace occur, and the talk produced in overlap is not treated as in any way requiring remedy, repair or re-introduction. 5 This talk could in fact be removed from the conversation and the organisation of the surrounding talk would be in no way changed. If this talk is in fact not tied to a local sequential environment the issue of how it is to be understood emerges as a serious, and perhaps unsolvable,



problem, not only for the analyst but also for participants. A relevant sequential environment is one of the principal resources participants utilise to make adequate and appropriate sense out of talk. For example, taken in isolation the words 'three and a quarter' can make reference to an indeterminate number of phenomena and perform an unspecified range of actions such as naming a price in a service encounter, providing the weight of an object, specifying the distance for a race, etc. However by taking into account the sequential position of this expression in these data (lines 9 and 10) participants on this occasion of its use can find that it is providing an alternative formulation of an object located in a just prior piece of talk, 'Fisher's car', and is doing the activity of checking and demonstrating independent recognition of that object. Unless Phyllis's talk is tied to a relevant sequential environment what would constitute adequate understanding of it remains problematic. Is it 'you', 'I', 'we', or 'they' who need more ice? What kind of ice? Why is it needed? etc. Given the words in isolation an indefinite number of accounts can be imagined, but criteria for deciding a relevant and appropriate understanding of the talk for these participants remains unavailable. The inadequacy of the talk in isolation suggests that, despite the absence of ties to the talk around it, a relevant environment for what Phyllis is saying does none the less exist. When the visual record of the conversation is examined it can be seen that Phyllis begins to talk just after she has poured the end of a bottle of coke into a cup in front of her and while she is still looking at the cup. If others present tie her talk to her actions with the coke an environment for making sense of the talk becomes available. For example, the embeddedness of the talk in these activities can locate the party needing ice as the speaker and the ice as ice for a drink. Looking further at the videotape it can be seen that as Phyllis speaks she picks up the cup of coke, gets up, and walks away from the table. Thus her talk, when analysed with reference to her actions with her drink, makes visible a reason for why she is leaving the conversational cluster. Tying the talk to these activities is however not something that occurs automatically but rather a task to be achieved by the participants. The data suggests that the speaker in fact goes into somewhat special work to make these activities, and the relationship of them to her talk, visible and available for her co-participants. First, the talk is noticeably sequenced as an event in these activities, occurring just after the pouring is finished and just as the act of leaving is begun.



Second, in large part by virtue of the placement of the talk after the pouring, the talk appears to be occasioned by the pouring, the finding of no ice emerging at that point. Such a finding need not however have been made just at that point, but could have been done earlier, and perhaps was (less than half a minute before her remark Phyllis takes a sip from the coke and then shortly afterward looks into the glass as she starts to pour). Moreover, ice is not always, or even characteristically, added to a drink after it has been poured, but rather is placed in the glass prior to the pouring. While the sequence that occurs here (pouring of the drink followed by a search for ice) is certainly possible, such observations invite consideration of what would have happened if the pouring had not been done first, but speaker had rather j ust picked up her cup and left. In such a case of the activity of leaving would have been both more sudden and far less explicable, the placing of soda in the cup not emerging as a seeable event. The sequencing chosen is thus consistent with the possibility that the speaker is doing special work to make her activity visible and available to her co-participants. Moreover, by virtue of this work the departure becomes both a locally occasioned and an accounted-for event. Ice is needed at the present moment (the coke has already been poured) and is not available in the present environment. 6 Third, the speaker carries the empty bottle, as well as the cup of coke, off with her. One of the few advantages of pouring first would seem to be that the bottle could then be abandoned, and indeed such an action would be appropriate at the present picnic; the table is full of empty beer cans. However by carrying the bottle as well as the cup, speaker again heightens the visibility of what she is doing. The speaker thus seems to systematically organise her actions so as to provide her co-participants with resources that will enable them to clearly see both the activities she is engaged in and the embeddedness of her talk in those activities. Indeed, there is good reason for such work since it is those activities which provide her talk with an environment within which it becomes comprehensible. However, while these activities inform the talk, and make it comprehensible, the talk simultaneously informs the activities, and makes them comprehensible, providing an account for what the speaker is doing and why she is leaving. Co-participants thus use the activities to find the sense of the talk and simultaneously use the talk to find the sense of the activities.



Orientation to the Talk of Phyllis by Others In many circumstances the talk that occurs within a task activity is crucial to actually getting the task accomplished. For example when a guest at a restaurant asks a waiter for more ice the request that is communicated through the talk is an important component of the process of getting ice from the kitchen to glass; were the request not made the task would not be accomplished. However the talk that Phyllis produces does not help her actually perform the activity of getting more ice; she could replenish her glass just as effectively by going to the refridgerator without saying anything at all. Rather than helping her get more ice this talk functions to make the activity that Phyllis is engaged in accessible to others. Thus, though the talk is embedded in a specific activity it is designed and placed there exclusively for observation by others. In view of this it is relevant to examine how others present attend to what Phyllis is doing and the talk she is producing. When the tape is examined it is found that none of the others present orient to the speaker, for example by gazing toward her, or attending to what she is saying in any way. Rather, they remain exclusively involved in the talk they are already engaged in. Speaker's talk thus passes without any displays of co-participation in it, or hearership to it, whatsoever. Speakers who find that they do not have a hearer have systematic methods for requesting such co-participation and delaying the onward production of their talk until is it obtained (see for example Goodwin, 1981: Chapter 2). In the present case, however, the speaker in no way treats lack of displayed co-participation as a situation requiring remedy. She neither interrupts her talk nor makes any effort to secure the orientation of a hearer. The data thus provide an example of a strip of talk produced without the displayed co-participation of a hearer and without speaker seeking such co-participation. 7 In essence the talk comes off as an instance of what G o f f m a n (1978) has termed self-talk. Goffman notes that one of the characteristic places where such talk occurs is at the 'interstice between a state of talk and mere co-presence' (1978: 796) and that, unlike talk addressed to someone within a particular conversational cluster, self-talk is available to the gathering at large (1978: 794). The talk that Phyllis produces explicitly accounts for a movement from a particular conversational cluster. Not only is such a movement visible to all who are present but it may in fact be relevant to them. For example if Phyllis is now to move to a different cluster those already within that cluster might have to re-arrange their actions to incorporate her. From this perspective it is interesting to note that the talk Phyllis produces not only states a reason for withdrawing from her current



cluster but also provides information about the actions she is about to engage in. All present are thus able to see, not only that a movement has been initiated, but where that movement is going and what it is doing. 8 However, despite the information that Phyllis makes available to others, both her talk and her departure are performed without the explicit collaboration of anyone else, and thus come off as single-party events. This is notable in itself in that departure from a state of talk is frequently, perhaps characteristically, performed as a multi-party event, and indeed an event achieved within the conversational sequence itself (see for example Schegloff & Sacks, 1973, and Heath, 1979). The fact that this does not happen in the present data helps maintain the integrity of the other events then in progress. If Phyllis had done her departure as a multi-party event others in her cluster would have had to set aside their talk to attend to her. But when departure is managed in the way it is here, others do not have to disrupt what they are doing, and the activities they are engaged in remain intact. O n e feature of the departure that may be relevant to the treatment it receives is that, unlike the closing of an encounter, it does not mark a definitive end to the current accessibility of the parties to each other. The picnic is still in progress and it is quite certain that Phyllis will make contact with the people she is moving away from within a rather short period of time. From this perspective it is interesting to note that the reason she gives for leaving and attending to her drink, shows an orientation to her continued involvement in the event within which the particular cluster that she is leaving is lodged (note the very different effect that would have been produced if she had said that she was leaving to attend to matters unrelated to the picnic, for example 'I'm gonna go read a magazine'). Her departure can be seen as a brief hiatus undertaken to take care of needs relevant to the gathering in progress, rather than a rupture with the gathering and its participants. However, though such features of the departure make it possible for it to be ignored, they in no way establish that this is the way it must be treated. For example a similar departure occurs in the following but it gets an answer in a next turn from a recipient: 9 Don: John:

I'll go get some more water ((Leaves with pitcher)) Okay.

Indeed, Phyllis's departure is attended to by one of the parties she is leaving, though that noticing is organised in such a way that the unilateral character of her departure is preserved. What happens will be examined in some detail. During her talk and the silence in line 13 Phyllis lifts herself around and off



the picnic bench she has been seated on. She actually steps away from the table during the word 'fooling' in line 14. Only after that has happened (i.e. when she is no longer physically part of the cluster and is in fact turned away f r o m it) and as his own talk reaches completion (over the word 'around' in line 14) does Curt move his gaze in the direction of Phyllis: 10. Curt: Thr^ee enna quarter. 11. Phyllis: °Need some more i f ce l 12. Mike: Yeh, 13. (1.0) Phyllis Steps Away From Table 14. Curt:

I When I wz foolin around. I Curt Starts to Move his Head in the Direction of Phyllis

Moreover, though Curt positions his head so that Phyllis falls within his line of regard, he mitigates that look in a number of different ways so that something less than official gaze toward Phyllis is visible. First, his hand is to the side of his head so that it is between his face and Phyllis. Second, his head does not track the movements Phyllis is making. Thus, his head movement toward her does not come to a complete stop (so that he could be seen to be gazing at her) but instead bobs slightly, and he lets her walk out of his line of regard without following her. He then moves his head sharply in front of her with the effect that she again passes through his line of regard while he appears to be looking past her. With his sporadic head movements Curt manages to take note of what Phyllis is doing without tying his gaze to her movements and thus making her the visible, official object of his gaze. The effect of all this is that what Phyllis is doing is noticed but the noticing is organised in such a way that it does not propose the relevance of either of the others attending to what Curt is looking at, 10 or of Phyllis interrupting her leaving to deal with it. Though her action is taken account of, it remains unilateral. Both speaker's lack of effort to secure a hearer, and co-participant's efforts to avoid making the noticing of her departure something speaker has to attend to, raise the possibility that non-coparticipation of others in speaker's actions is in fact something they systematically work to achieve. Noting the structural properties of self-talk (Goffman, 1978) sheds important light on the organisation of the events occurring here but it does not explicate the detailed interactive work participants are doing to have a strip



of talk come off as an instance of such a phenomenon. The organisation of the talk Phyllis produces will therefore be examined in more detail. First, though her departure with its talk is available to the situation as a whole, and not just the cluster she is leaving, these actions do have special relevance for that cluster. Not only would they be the ones to participate in a sequence with her if the event were to be formulated as a multi-party activity, but her departure might be seen as informative about their treatment of her. Indeed, for some time prior to the departure she has occupied a somewhat special position in that group. Talk within the cluster has turned to cars, a subject that this group treats as a male domain (detailed examination of the way in which this is accomplished is beyond the scope of the present analysis). Thus, although Phyllis is physically part of the cluster she is not included with the participation structures made available by the current topic. The other women present at the picnic have formed a separate cluster and it may indeed be appropriate for Phyllis to leave what is now a recognisably male cluster. However such a noticeable action may have the effect of focussing attention on the fact that those she is with are not providing for her inclusion in their talk, i.e. her departure from the cluster could be seen as responsive to the way that she is being treated by the others in the cluster. The talk that she produces while leaving undercuts such a possibility by providing not simply an account for the departure but the official account for why she is leaving." As the person performing the action she can be seen to have privileged access to the reasons laying behind it12 and by showing that she is leaving to attend to needs of her own rather then reacting to her co-participants she permits their activities to continue unhindered. Not only are her co-participants not left to puzzle about why she is leaving but they are provided with a reason that is not in any way relevant to their own actions, and thus not something to be dealt with by them. Just as the content of the talk avoids implicating others so also is its production and articulation carefully performed so as to display that coparticipation is not sought or even appropriate. First, the talk is produced with noticeably lowered volume (indicated in the transcript by the degree sign before the talk). This volume contrast not only sets off this talk from the other talk then occurring but also provides a means for displaying that the talk is not claiming space within that other sequence of talk. Through use of this technique that speaker is able to produce her account while simultaneously displaying that this talk is not being performed as an intrusion into the talk of the other participants. Second, while speaking Phyllis does not gaze at any of the others present but rather keeps her eyes lowered and is in fact moving them away from the



conversational cluster by the end of her turn. The only thing that she does look at is the cup in her hand as she is picking it up. By organising her gaze in this fashion she performs a number of relevant actions. First, the movement of gaze is shown to be a component of the act of leaving the table. The particular way in which gaze is handled thus helps m a k e more visible the particular activity within which the talk is embedded. Second, one of the principal rules organising gaze within talk argues that a recipient should gaze at the speaker when the speaker is gazing at the recipient. 1 3 By not bringing her gaze to a recipient the speaker avoids invoking the relevance of this rule. The parties who fail to gaze at her during this talk are thus acting in a way that the speaker herself has proposed that they should act through the way in which she has managed her own gaze. Third, but related to the last point, by not gazing at any of the others present the speaker avoids performing the act of addressing any one of them or all of them in general. With reference to this it can be noted that one way in which the talk she produces might be analysed is as a request, i.e. by stating that something is needed she might be heard as requesting that someone get her the needed item. A more formal description of how talk such as this can be analysed as a request is provided by Labov & Fanshell (1977). In their analysis, speech acts, such as requests for action, include a set of preconditions, for example that the requested action needs to be done, that the recipient of the request has the ability to perform it, etc. (Labov & Fanshell, 1977: 78). Someone can perform an indirect request by making a statement that refers to one of these preconditions (Labov & Fanshell, 1977: 82). The talk that Phyllis produces could be heard as referring to one of the preconditions they identify, the need for a particular action to be performed. 14 Moreover someone who might be seen as the proper recipient of such a request is part of the current conversational cluster. Curt is both the host of the picnic and the owner of the house in which the ice will be found (from this perspective it is interesting to note that he is the one who takes note of her departure). By making it clear that neither Curt nor anyone else is being addressed, and by carefully situating the talk within the act of leaving, Phyllis shows her recipients that the talk is not to be analysed as a request. T h e task they are posed is not recognition of a precondition presumed to underlie some speech act but recognition of an activity, and speaker provides her recipients with abundant resources to see this task and to accomplish it. Fourth, by moving her gaze, and situating it within the act of leaving, the speaker makes herself unavailable for co-participation with others present not only during the turn, but also after it. The speaker thus displays



unavailability both during her talk and in the position just after it where subsequent action might be addressed to it. Through use of all of these phenomena together the speaker manages to construct a piece of talk that does not propose the relevance of other parties' displayed co-participation in its production by, for example, acting as hearers to it. The lack of orientation by others to it is consistent with the way in which the speaker proposes through the details of its production that the talk is to be dealt with. When Mike overlaps it in line 12, he treats this talk in the way in which it carefully and systematically displays that it should be treated, and the departure itself can come off as a totally irrelevant and unnoticed event. The line of argument which has been advanced here suggests that though the other parties present do not officially display hearership they may none the less hear the talk and take it into account in the organisation of their actions, for example by not co-participating in what would otherwise be a noticeable event, the departure. The unilateral, single-party departure that is not attended to by others, as well as the talk which receives no official displays of hearership, are thus still interactive events, achieved through the collaborative action of multiple participants. What is at issue is not the distinction between interactive and non-interactive action, but rather alternative structures available to participants that propose the relevance of different types of co-participation in ongoing events. With the procedures employed here the speaker has constructed a strip of talk which passes as uninteresting and indeed irrelevant. This talk nevertheless has interesting properties. For example, it provides an example of talk that does not claim space in the sequence of talk when in progress, talk that invokes the sequential relevance of an organisation other than talk for its comprehension, and talk produced for the hearing of others who are simultaneously instructed not to act as hearers to it. Such properties are neither contradictory nor accidental, but rather sensitive in detail to the particulars of the local environment where the talk is placed, and the tasks the speaker is attempting to accomplish in that environment.

Notes to Chapter 7 1. A n earlier version of this paper was presented at the 74th Annual meeting of the American Sociological Association, Boston Massachusetts, August 1979. I am indebted to Marjorie Goodwin, Erving Goffman, Richard Holmes and Anita Pomerantz for helpful and insightful comments on that version of the analysis. I alone am responsible for the weaknesses that remain.



2. The most detailed analysis of the sequential organisation of conversation is to be found in the work initiated by Harvey Sacks and his colleagues. See Sacks, Schegloff & Jefferson (1974) for analysis of how turns at talk are exchanged in conversation and Atkinson & Heritage (1984) for a collection of recent research into organisation of conversation. For analysis of interaction within the turn see C. Goodwin (1981), M. Goodwin (1980), Heath (1979), Jefferson (forthcoming), and Sacks (1974). 3. For analysis of a range of different types of speech that fall outside a ratified state of talk see Goffman (1978). 4. I am indebted to Gail Jefferson for audio-transcribing the tape from which this sequence is taken. 5. For analysis of some of the procedures available to participants for the negotiation of overlap see Jefferson (1973). 6. I am indebted to Paul Drew for bringing to my attention the importance of such displays in the organisation of accounts. Work of his that is currently in progress provides detailed analysis of such structures. 7. For some analysis of other types of talk produced without the co-participation of a hearer see Goodwin (1981, Chapter 3) and Goffman (1978). 8. In view of the way in which the information Phyllis provides is relevant to others present the argument by Corsaro (1979:333-35) that providing a verbal justification for leaving a conversational cluster is a form of 'modality redundancy' that carries only ritual meaning (which Corsaro recognises to be quite important in its own right) does not appear correct. Corsaro also argues (1979: 333) that the nursery school children he observed left a cluster of co-participants without any comment or remark whatsoever. However in the one example he provides he describes the child's leaving as follows: While looking toward another child the party about to leave, Barbara, said 'I'm tired'. Following this 'there was then a pause, and Barbara turned and saw Rita at the swings; then she said " O h " and ran o f f (1979:333). Corsaro argues that the 'I'm tired' should not be heard as an account for withdrawal. However both the look toward the activity she is about to become engaged in, and the verbal 'Oh' which informs co-participants of that noticing (for more detailed analysis of the use of 'Oh' in conversation, including its ability to mark a shift to a new activity or topic, see Heritage, 1984) would seem to do much the same work that Phyllis's talk and activity do. For example they show that a new activity has claimed her attention and provide information about why she is leaving and where she is going. 9. Note that in this case the party who provides the answer is not concurrently engaged in another sequence with other participants. His action to the party leaving thus does not disrupt another line of action. 10. For analysis of how recognisable gaze toward something can make it relevant for others to join in that looking see Goodwin (1981: 98-100). 11. On this issue see Sacks (Spring 1966, Lecture 20). He notes for example that 'this phenomenon of if an explanation is available then its that explanation that is the explanation, and formulates what it is that's happened, is of course very general'. 12. For more detailed analysis of the differential access different types of participants are seen to have to events see Pomerantz (1980). 13. For more detailed analysis of this process see Goodwin (1981, Chapter 2). 14. See also Ervin-Tripp (1976) in which it is noted that directives which are 'need statements' are as baldly stated as the imperative form.

8 6 You want to find out if anybody really does care'1 HARVEY SACKS Late of the University of California, Irvine

I will begin with a quotation. 'B' is a suicidal woman, 40 years old, divorced, no children. A: B: A: B:

Well perhaps you want to tell me uh why you feel like committing suicide. ((sigh)) ((sigh)) Well it's the same old childish reason that everybody wants to commit suicide. Why is that. You want to find out if anybody really does care.

There are many interesting things here, and much we would have to do if we were to be able to generate this interchange. I am now mainly concerned with . . if anybody really does care', and not the particular objects by which this sequence gets done. But I do want to note the fact that this question, '. . . tell me why you feel like committing suicide?' is, in the first place, askable. That is, that it stands as a sensible and appropriate question to which there is expectably or reasonably an answer: Why you want to commit suicide is something that you would have information on, or could propose to know. That the question is askable can be accounted for in this way: Given that there are sets of question-forms which Members use, one of which is 'Why do you want to do X?' where 'X' is some activity, and given that 'suicide' is an activity-category, then, just by reference to the relevance of that question-form for any activity, 'Why do you want to do it?' can be applied to suicide. How it is that such a question can expectably or reasonably be answered is worth some consideration, since for professionals there are classes of 217



things which, if you do them, or want to do them, then ipso facto you don't know why. And psychiatrists — and psychoanalysts in particular — take it that a person who wants to commit suicide does not know why they want to commit suicide — in the sense that the psychiatrist could say why they want to commit suicide. And of course sometimes a person says 'I don't know'. Now that fact does not seem to stand in the way of asking the question. And the issue then is, what is the relevance of that question, and what would happen as persons come to know what it is they did not know? That is Socrates' classic problem: that one thing about knowledge is that you know what you don't know, and to the question 'Why?' the answer 'I don't know' is sort of a deeper answer. That is, it might have an awareness of the character of this knowledge as something only professionals have. Now, the notion of 'opinion' as contrasted to knowledge (and Plato made a great deal of the difference between them) and the sheer introduction of a notion of 'opinion', provides in part for professionals' talk to laymen. Because one of the characteristics of 'opinion' is that it is something which lay persons are entitled to have when they are not entitled to have knowledge — in the sense that they can offer it without ever proposing to have to then defend it. Like they say 'My feeling is such-and-such on that, but I don't really know', as a permissable way of talking, where one then does not try to find out what kind of defence you have for that statement. So in a way, 'opinion' provides for the continuing discourse between professionals and laymen. And I presume that it's a means or a mechanism by which, not just psychiatrists, but perhaps professionals in general can talk to clients — by the notion of the permission that 'opinion' gives to a person to talk. That is, under the control that one does not really know; which is to say, one is not entitled to know. And very frequently when you see 'I don't know' appended to some statement, that's what it seems to be doing — providing that 'I'm not entitled to say this', that is to say, 'I cannot defend it professionally', if it's a matter of professional information. But if it is the case that there is going to be discourse between clients and professionals, or between the public and professionals, then the fact of a distribution of knowledge which provides that professionals know and laymen do not know, might seem tremendously interruptive unless you had some mediating device, like 'opinion', which would permit laymen to keep talking even when they find out that they do not know. Otherwise they might not have any way, for example, of even turning to a professional. I will now move on to the matter I want to focus upon: Why is it that suicide seems to be a way to find out if anybody does care? The question I

YOU WANT TO FIND OUT IF ANYBODY REALLY DOES CARE 219 asked when I was sitting trying to puzzle that out was, what are the available ways in this society for going about seeing, and determining, that others care, or that one is relevant to others? What are the means available for seeing one's relevance? And while I had that stored at the back of my head, I was reading 'Witchcraft, Oracles, and Magic Among the Azande' by Evans-Pritchard (1937). And some of his observations can begin to give you a feel for what such a procedure might look like. Here is what he reports. Whenever anything goes wrong among the Azande — if an Azande feels lousy, gets sick, injures himself, is economically in trouble, etc. — he engages in the following procedure. He pretty much drops whatever he is doing and goes off into the woods with some oracle procedure. Like, say, one oracle procedure is, they take a chicken and give it a little poison and ask questions to the chicken, which the chicken answers by dying or not dying upon being given the poison. So the Azande takes a chicken and some poison and goes off into the woods with it. And he sits down and makes up a list, essentially composed of his neighbours. He considers what his state was before he got ill, and then goes through this list of neighbours, considering about each person how he takes it they feel about his situation. Are they unhappy that he just got married that week, that he just got some wealth, etc.? By going through this procedure he then locates some persons who he figures would like to cause him trouble. And for each person that he has in this way, he offers a name to the chicken and gives it some poison. On some giving of poison the chicken will die. The person whose name was offered on that occasion is the person who has done him the trouble — caused him to have some illness, caused the rain to fall before his crops were in, caused him to have a bad hunting trip, etc. And once the one who caused the trouble is found out, there is some procedure for getting amends. Evans-Pritchard reports that the Azande just love to do this. There is pretty much nothing that will stand in the way of them stopping and going off into the woods and making up a list, and sitting down and considering, for all the people around, how they are interested in his good or bad circumstances. Now, this is one rather nice kind of procedure, which is institutionalised in a society, whereby persons can take an occasion and determine for themselves properly — that is, there is proper occasion for doing it — whether anybody cares, and how they care. Let me make a parenthetical remark about the situation of the Azande as compared with this society. One of the things that lies at the basis of the availability of that procedure for the Azande, and which is not present in this



society — and which then provides that we do not do that in this society — can be stated in the following way. The Azande do not have an institutionalised notion of chance. Things like falling ill, and most particularly things like dying, do not occur by chance for the Azande. There is always somebody who is responsible. And there is a set of procedures, the purpose of which is to find out who it is that is responsible. And these are not random procedures, because one has some way of finding out, in the first place, who would be interested. Now it is not that the Azande do not have a good notion of 'natural causes'. They are perfectly well aware of the fact that you can get ill from natural causes. That does not exclude the fact that there is somebody interested in those natural causes occurring. Evans-Pritchard reports, for example, that somebody will stub their toe on a tree and then go off with their chicken. Evans-Pritchard says to the guy, 'Well after all, you know, it's your fault. You stubbed your toe on the tree.' And the guy says, 'I know perfectly well that I stubbed my toe on the tree, and that the tree caused that trouble, but I've been through this forest hundreds of times and I never stubbed my toe before. There must have been some reason, then, why it happened this time.' And that, then, provides for the responsibility. So it is not a matter of them not having a good notion of natural causes. It is that they do not use a notion of chance. That being so, you can come to see how rather special it must be for a notion of chance to be, in fact, enforced — and how easy it might be for it to break down. Because what a notion of chance involves is that something that happens to you is not a matter of inquiry as to how it came about. It just happened. You simply do not investigate why this or that trouble arises, for a great many troubles. And that might provide for people to do you ill, in more or less subtle ways. The notion of chance is a pretty tender one anyway, and persons suffering various troubles in our society will often feel that they have to shed it and begin to employ, for any given trouble, the question 'Who did that, and why? What do they have against me?' That is to say, they no longer feel able to — or they feel compelled not to — use a notion of chance where others use it. But in this society it is not proper, and in fact it is diagnostically significant, if you do not use the notion of chance. By 'diagnostically significant' I mean, persons who do not have a notion of chance are persons who have the symptoms of paranoid schizophrenia. When some trouble befalls them, they take it that is is some persons who are in the business of generating it for them.

YOU WANT TO FIND OUT IF ANYBODY REALLY DOES CARE 221 For the Azande, then, there is a device which is routinely employable for checking out how it is that others attend to your ill — or well-being. Once we have some idea what such a procedure looks like, then we can begin to consider what sorts of things there are that look like that in this society. What are the occasions under which one can make up a list like that and just sit down and consider who cares and what do they care? I think you can find that there are very few. O n e such occasion is the wedding. Before a wedding the parents of the bride sit down with a big list and have an enormous amount of fun considering, 'Would So-and-So be happy that my daughter is getting married?' 'How would So-and-So feel?' Some people give parties to occasion such a device; that is, they say 'I just gave a party to see who my friends are.' But I take it that the most prominent occasion in, so to speak, a person's life, is right after they die. In this society, on the occasion of death, people gather around and talk about how important So-and-So was to our lives, how much we cared about him, how much we miss him, what a marvellous guy this was. And that is what this suicidal woman reports. Later in the call she talks of how her father 'was aggravating everybody' but 'as soon as he died . . . they just said he was a great guy'. And anybody who has ever witnessed that scene has learned what an opportunity it is. And of course it is a well known fantasy, seeing yourself as the one who died, getting a chance to get those credits which persons never give you and that you cannot, yourself, collect — that is, for which there is no occasion to collect them. You can see how, for somebody in pain, that scene after death — which is known to everybody as an occasion for having persons propose that they care about somebody — may then come as something exceedingly attractive, and 'the only way'. And how, then, the 'attempted suicide' can be the attempt to actualise that scene. There are, of course, less dramatic devices for considering somebody's relevance by reference to missing them, or absences. For example, when somebody comes back from somewhere, the question is, 'Did you miss me?' as a way of deciding whether it is that one cares. The question of absence and loss, then, seems to be a basic way that one has of dealing with relevance. Now there are other, more specialised devices for doing a similar task. I will start considering one of them in a slightly tangential way. One of the things I came across several times in the telephone conversations I have been analysing, involved a widow or widower who was suicidal. They would say that time hangs heavy on their hands and what they find is that 'nothing happens'. Nothing happens to them. And I wanted to see if there was some



way of finding out how that comes about — that somebody sees that nothing happens to them. I also have conversations between young married persons. And one of the most exquisite kinds of things that young married persons do with each other is, they say things like 'Kennedy was assassinated two weeks after we got engaged'. I want to give the name 'private calendars' to that sort of talk. A n d I want to note that married couples, each one, by themselves, independently, construct these private calendars. And what private calendars do is to provide for the locating of, not only events within that relationship, but events of the world in general, by reference to the relationship. Further, these calendars are 'causally powerful'. What I mean by that is, there are all kinds of events which can be explained by references to the relationship. There is a generic statement: 'Because A did X, B did Y', where one can substitute for A, 'wife' and for B, 'husband', and substitute for Y the event to be explained, and for X the activity which can explain Y. This provides a large class of sensible statements which persons in units like husband-wife are able to employ. Indeed for many events, such statements have to be employed; that is, for many events, such an explanation is the only sensible explanation. So it is often said that while you can give a whole list of explanations for why it is that somebody succeeded, in the last analysis it is because of his wife. It's said without knowing the guy, or knowing anything else. Another sense in which the private calendar is causally powerful can be seen in the paradigmatic statement, 'That was before I met you and I was lonely then'. There is a class of logical statements which the logician Nelson G o o d m a n (1953) named, and pointed to as creating very basic problems for the philosophy and logic of science. He calls them 'counterfactual conditionals' . Of which an example, I think, is 'If one had lowered the temperature to such-and-such a degree, then the following would have happened' — where one has not lowered the temperature and the thing has not happened, but one has done something else and something else has happened. Many scientific statements are made that way, and Goodman argues that there is not currently a logic providing for them. But counterfactual conditionals are nonetheless routinely used, and they are, nonetheless, enormously powerful. Which suggests that perhaps a logic can be invented, or that they are building on something very strong. Many uses of the private calendar are such uses. See, one of the problems in developing a relationship is finding out that the states of the person you are with are to be accounted for by you, and not by the sheer fact that they are with somebody. That is, they want to be able to say that even if

YOU WANT TO FIND OUT IF ANYBODY REALLY DOES CARE 223 they were with somebody before, they would still have been lonely. And that is what one wants to do with these private calendars. There are ways of building up, in deep and repetitive ways, the relevance of 'you'. And perhaps one of the big things about marriage is that that is just what you are constantly doing for each other. The notion, for example, that marriage is made in Heaven, is kind of an underpinning to the use of these things. That is, it is an account that would provide the basis for saying 'That was before I met you and I was lonely then'. Our meeting was virtually guaranteed, and it is just a matter of, until then one drifted, and now it happened. By virtue of this causal structure, of course, persons who are members of such units have built in procedures for finding that someone cares. And for a lot of things it is the only way you can find the sense of what is going on. Let me point out something about the private calendar that turns out to be rather important. I don't have a very large set of features of these things, but one thing I have found out is that if we compare these private calendars to everybody's calendar, then there is one striking difference between the two of them. And that is, everybody's calendar has, and private calendars do not have, guaranteed continuity. Everybody's calendar runs on into the indefinite future, without regard to anybody in particular being present. Private calendars end when 'we' end. The end of a relationship, in one way or another, can provide that there are no more events on the private calendar. Now then, what we can see the widowed person saying, when they say 'Nothing happens anymore', is that with regard to the private calendar, whereby events between me and my spouse happen and the value of my life is found, no more events can occur on it. You can get, then, a sort of task that a therapist, or somebody else to whom one of these persons would turn, might have. The task is at least programmatically simple, whether it is easy to do is another question. It involves bringing them back to the use of everybody's calendar, whereon events can still occur sensibly in their lives. I will add another thing, and this is somewhat more conjectural, though not strictly made up, and it may be relevant for our materials here. For widowed persons, the fact that they have had a life with somebody is something that the other's death does not take away. And they can say 'we had a marvellous twenty five years together', pointing to all the things we did together, how it is that I was happy on this day because of what he was doing, because we were together, etc. Now, when persons get divorced, something quite different seems to operate. Apparently a divorce can provide for the fact that one cannot even retrospectively use the private calendar one had going. The fact of a divorce, perhaps with the reconsidering of whether one



ever did care, and what after all they were doing these last five years that led up to this, seems to involve that one cannot then use it for the past that one was 'together'. That the woman in our materials is divorced may then not only provide that she has no current access to the built in procedures for finding that others care which such a unit as husband-wife provides, but also that she is deprived of whatever retrospective use she might have had of that unit's private calendar. Via this sort of a sketch we can begin to see where the relevance comes of having others care. And that is that the whole class of causal statements that are built out of such units as husband-wife and the relationships between categories in these units, provide an apparatus in which everybody is supposed to be entitled to become a member of such a unit and thereby to have these things done for them. And if they don't become a member, given that they are entitled to become a member, they have a clear way of seeing that something is missing. It is not the easiest thing in the world to find a way to say that something is missing. But if you have some objects for which there is no rule of exclusion in the first place — everybody is entitled to them — then if someone does not have it at some point that one is entitled properly to have it, one can say that it, and its consequences, are missing. We can tie this up to some extent by asking what, then, is the consequence of not having persons care? Well, these lay theories — and all these causal statements and entitlement propositions are lay theories — have a rather interesting property. If you consider our prototype of a scientific theory, then, if some object does not conform to what the theory proposes about the object, then the theory has to be revised. This world has been constructed in a rather more exquisite way. What goes on is the following. A large class of lay theories are properly called 'programmatic theories'. If they do not describe your circumstances then it is up to you to change. And if they do not provide for you as a Member, then it is up to you to rid yourself of being a Member, for example to kill yourself. In that way you keep the theories going as descriptive. If you are a member of one of these units you have essentially automatic ways of finding that others care. It is built into the structure of ordinary discourse, and the ways persons see how events come off. If you are not a member of such a unit, it is still relevant, but its structure is not available to you. And you may then try that procedure which works for everybody — dying — either as a way to find that somebody does care, or as a way of providing that the theory — that people ought to care — is made correct by virtue of your no longer being a Member. And we will see constantly that persons talk of a whole range of things where if something is not so for them,

YOU WANT TO F I N D OUT IF ANYBODY REALLY DOES CARE 225 then that does not provide that what is supposed to be so is thereby wrong, but that they are wrong. Let me add one more device relevant to 'Does anybody care?' It is, of all things, trash mail. The next time that they have hearings about removing trash mail, I am prepared to go and testify against its removal. Because trash mail is a most interesting thing. Some of my students used to go to the park and sit and talk to people. A lot of those people were old ladies. They were all utterly isolated. They came to Los Angeles after their whole family died, or they came with their husband and he died. They live in apartments near a park and they spend their day in the park. But they regulate their lives in most interesting kinds of ways. Even though they have almost no money they, for example, never purchase at supermarkets and never purchase more than a day's food. Because if they did, they'd have nothing to do the next day. And they routinely will get up — you will be sitting in the park talking to them, the only person who has talked to them since God knows when, they nevertheless get up and say 'It's eleven o'clock. I have to go home and check the mail.' Now there is nobody who is writing to them. What it is, is that there's that trash mail coming, and that is something. Consider their situation: The mailman comes every day, and they know it. A n d that means that for them, they have to go check the mail every day. The only mail they do get is this mail that everybody gets. But for them, it is something. And if they had to recognise that he would come every day, and every day they would find no mail, and they could look forward to that day after day, then that situation of theirs, that isolation, would so be built into their circumstances and shown to them routinely, that it might become far more unbearable than it is — and it is pretty unbearable — because this is a device that happens every day, for whomsoever. You do not know who is getting telephone calls, you do not know how many phone calls are being made, but every day, everybody has the mailman go by. And if you just consider the comparative cost of trash mail versus an enormous mental health operation, then trash mail is not expensive. And for these people it is by and large the only means by which the routinely-used device of delivering mail does not become the kind of thing it would otherwise become — this persistent statement to them, nobody cares. Note to Chapter 8 1. This Chapter is an edited combination of two transcribed lectures, presented in Fall 1964 and Spring 1965, compiled by Gail Jefferson.

9 Descriptions in legal settings ANITA POMERANTZ Temple University, Philadelphia

Introduction Because the judge or jury has the responsibility for resolving factual disputes, the details of what a witness knows are usually preferred to the conclusions the witness has reached. . . . But . . . the line between facts, an the one hand, and opinions or conclusions, on the other, is not easily drawn, and what are analytically opinions may be more helpful to the trier of fact than the perceptions they summarize. (Lempert & Saltzburg, 1982: 4.) W h e n litigants describe 'the facts' during a hearing, they often present different versions of what happened. Their testimony may be a large part of the evidence that substantiates their respective claims. In a civil case, for example, a plaintiff-consumer might report events that document how a merchant refused to either repair or replace defective merchandise. The defendent-merchant, on the other hand, might report events that document that the problem with the merchandise was caused by the consumer's improper use of it. In institutions that handle conflict, there is an expectation of the possibility or probability of discrepant versions, particularly concerning matters that bear on the outcome of the case. Part of that outcome may be an award of compensation. As compensation is largely determined by the loss or damage suffered, a plaintiff frequently seeks to maximise the damage and the defendant to minimise it. There are, however, matters that are neither in dispute nor are they centrally relevant to substantiating either side's claims. These straight226



forward indisputable facts nevertheless are described differently on different occasions. These variations are not due to the conflicting interests of the parties but to the differing uses of the descriptions within the course of the hearing. Understanding how descriptions normally are shaped for particular interactional uses gives us a base from which to look at discrepant descriptions arising from the parties' conflicting interests. What is meant by a description's use? During a hearing, a litigant may 'tell what happened', 'correct misinformation', 'make arguments', 'concede points', 'attribute blame to others', etc. These activities often are performed with descriptions of 'facts'. The descriptions through which these activities are performed are shaped so as to perform these very activities, for example 'giving facts', 'disputing false allegations', or 'arguing'. Interactants perform multiple activities simultaneously: a litigant may 'conclude' by 'giving facts' that 'dispute the opponent's allegations'. Activities are singled out in the following analyses if and when they are related to the type of description used. To illustrate some of the activities in which descriptions are used and to show how the descriptions are shaped to do these activities, descriptions of a single 'fact' are compared. The data comes from a hearing in a Small Claims Court in London. The case involves the following: the plaintiff lives in a Council-owned block of flats. Water overflowed from a bathroom tap in the flat above his and caused damage to his flat. The upper flat was vacant and undergoing renovation at the time. The plaintiff is claiming damages to his flat, naming the Council as the liable party. The Council is denying liability for the damage because they had hired outside plumbing contractors. They claim that the contractors left the tap running and that the contracted firm is liable. The adjudicator, in the beginning of the case, explicates the basic law relevant to this case — if owners in good faith instruct outside contractors to do work and something goes wrong with the contractor's work, the contractors and not the owners are liable. T h e participants in the case refer to when the flooding occurred on four occasions. Excerpt 1 Adj:

on the eleventh of March last year

Excerpt 2 Adj: Pla:

at two o'clock in the morning. on the eleventh


228 Excerpt 3 Adj:

early in the morning of the eleventh of March

Excerpt 4 Adj:

in March last year where early in the morning

T h e r e was no dispute over when the flooding occurred. There was only one source reporting that information, the plaintiff, who is treated as a credible witness. He gave a first hand account of the flooding. When compared, the four descriptions are interestingly different.

Analysis In the remainder of the chapter, two sets of alternative descriptions are discussed in relation to the conversational activities in which they are used. O n e set, Official versus Relational identifications, consists of alternatives for identifying persons, places, objects, time, etc. In identifying a person as 'Linda Pomerantz', I am using a Complete Official identification, giving her name as it appears on official records. A Complete Official identification is designed to identify a referent 'decontextualised': without regard to the circumstances in which it is used.' Complete Official identifications are treated as the definitive way of identifying a referent, unless there are indications to the contrary. In identifying a person as 'my sister', I am using a Relational identification. With a Relational identification, a referent is identified in terms of how it is related to a point of focus. The referent that is identified is dependent upon the circumstances in which the Relational identification is uttered — when I say 'my sister', I am referring to person A; when someone else says 'my sister', (s)he is referring to person B. A n Official identification consists of proper names and numbers. A Relational identification consists of a referent identified in terms of its relationship to a point of focus. In the description, 'on the eleventh of March of last year' (Excerpt 1), 'the eleventh of March' is an Official identification: a number for the day and a name for the month are used. 'Last year' is a Relational identification in that the year is identified in relation to the speaker's present time. The second set, Numbers versus Characterisations, consists of alternatives for describing quantities. 'Two months' and 'two or three months' are examples of quantities described with Numbers; 'several months' and 'a few months' are examples of quantities described with Characterisations. When a quantity is described with a Number, the



Number is taken to be 'representee' (or not) of an actual circumstance. When a quantity is described with a Characterisation, the Characterisation is taken to be influenced by the judgement of the person who is characterising the quantity. Excerpt 2's 'at two in the morning' is a description of the time of the day using a Number for the hour; Excerpt 3's 'early in the morning' is a description using a Characterisation for the time of the day. Each of these alternatives is selected for particular activities. Some of the activities that have been singled out as relevant for these types of descriptions are: introducing matters into the hearing, clarifying, soliciting and giving testimony, providing a sense or coherence to an event, and showing an understanding of facts. The following discussions attempt to display the relationship between these activities and the various types of descriptions.

Instance 1 Excerpt 1: Flooded Council Flat In Small Claims courts, adjudicators have a fair amount of discretion regarding how they conduct the hearings. In Flooded Council Flat, the adjudicator begins by summarising the plaintiff's account of the events. Adj:

Pla: Adj:

If I can just run over the basic facts to start with I believe that on the eleventh of March last year, your flat a flat owned by S Council was damaged because water poured through the ceiling from the flat above flat nineteen That is exactly true And as a result of that you had damage to carpeting . . .

This is the first occasion in which the 'basic facts' of the case are described: what happened, when it happened, who is named as defendant, and the extent of the alleged damage. In describing them, the adj udicator introduces them into the record and gives both plaintiff and defendant access to a version of the facts that the court provisionally accepts. The adjudicator describes when the flooding occurred by giving the date, 'the eleventh of March last year'. The date is given with a Complete identification. Each unit [day] + [month] + [year] has a specification [the eleventh] + [March] + [last year]. In the remainder of the case, no subsequent reference to when the flooding occurred has the Complete date



specified. Making an initial identification in the hearing is an occasion in which a Complete identification would be done. The written forms that the adjudicator has in front of him undoubedly contain an Official identification of the date: a number given for the day, a name for the month and a number for the year. The adjudicator, however, identifies the year as 'last year'. In the delivery, he has transformed an Official identification on the written records to a Relational one, identifying the year that the flooding took place relative to the present. The transformation of the year into a Relational identification is consistent with a powerful practice in conversation, that adjacent temporal units are referred to with Relational identifications, e.g. 'yesterday', 'tomorrow', 'last year', 'next year'. In saying, 'last year' then, the adjudicator has adopted the conversational practice of using Relational identifications for adjacent temporal units. With the Relational identification, he has a sense of the time of the flooding relative to the time of the hearing, i.e. how long ago the flooding took place. The Relational identification of how long ago, in contrast to a year indication, may be used to answer other concerns, e.g. whether the hearing was taking place within a reasonable amount of time, whether the damage most likely already would have been repaired, etc. Another set of identifications in Excerpt 1 provides a further demonstration on the differing uses of Relational and Official identifications.

Excerpt 1-A: Flooded Council Flat Adj:

Pla: Adj:

If I can just run over the basic facts to start with I believe that on the eleventh of March last year, your flat a flat owned by S Council was damaged because water poured through the ceiling from the flat above flat nineteen That is exactly true And as a result of that you had damage to carpeting . . .

The adjudicator identifies the flat with the running tap in two ways. The first reference, 'the flat above' is a Relational identification: it locates the flat in relation to the plaintiffs flat. By referring to the flat with the running tap as 'the flat above', the adjudicator provides coherence to the event being described. It is by virtue of the location of the flat as right above the plaintiff's flat that the damage was caused to the plaintiffs flat. Part of our



u n d e r s t a n d i n g of water pouring through the ceiling of the plaintiffs flat c o m e s by conceiving the location of the overflowing water as in 'the flat above'. A f t e r identifying the flat as 'the flat above', the adjudicator adds a second identification: 'flat nineteen'. While 'the flat above' is produced with respect to the event's coherence and is referentially adequate for identifying t h e flat, this reference will not do for other purposes. C o n s i d e r the following situation. I am giving directions and say: 'When you get o n the Boulevard, go to the second traffic light and turn right. That's R i v e r Street.' W h e n I identified the street at which to turn right as 'the s e c o n d traffic light' I gave a Relational identification. I provided a way for a driver coming f r o m a particular place to identify the street. When I added ' T h a t ' s River Street' I gave an Official identification, the street name. In giving the street n a m e , I supplied the driver with a way of checking out w h e t h e r (s)he was on the right street. In E x c e r p t 1, it is not the case that all important referents are identified with Official identifications. T h e plaintiff's flat is not introduced into the h e a r i n g with its flat number. Because the identity of the plaintiffs flat is a s s u m e d to be unproblematically known with no chance of confusion or e r r o r , its introduction in the hearing with an Official identification a p p a r e n t l y is not called for. In contrast, the identity of the flat with the running tap is not assumed to b e u n p r o b l e m a t i c . In giving the flat number, the adjudicator is orienting to the c h a n c e of confusion as to which flat is being referred to as 'the flat a b o v e ' . 'Flat nineteen' is an Official identification, its own (non-relational) n a m e , which is treated as a definitive determination of the flat's identity. 2 'Flat nineteen' is only part of the address, not the whole address. The rest of the address would include the address of the block of flats. The a d j u d i c a t o r seems to assume that the parties unproblematically locate the flat with the running tap within the same block as the plaintiffs flat. Since the identity and location of the block of flats is treated as unproblematical, t h e r e is n o identification of the block. T o reiterate, to remedy any possible confusion or ambiguity concerning t h e identity of the flat, the adjudicator introduces its Official identification, 'flat n i n e t e e n ' , into the hearing. T h e Official identification allows the flat to b e r e f e r r e d to independently of the p l a i n t i f f s flat. T o summarise the discussion of Instance 1, Official identifications and R e l a t i o n a l identifications are used for different purposes. A n Official identi-



fication is called for when an identity is possibly problematic and needs to be determined or established. On such occasions, a Relational identification is treated as not a definitive identifier. A Relational identification provides a sense of a fact. It provides a way of viewing a fact as would be relevant for a current purpose or project. One of the uses that Relational identifications are put to is to assess how this case stands in relation to what is normal for this type of case, e.g. whether this hearing is occurring within a normal and reasonable period of time.

Instance 2 Excerpt 2: Flooded Council Flat Part of the adjudicator's job is to have the relevant details told, clarified, and elaborated. In this excerpt, the adjudicator prompts the plaintiff to report what he saw when he went upstairs to investigate the flooding. Adj: Pla: Adj: Pla: Pla: Adj: Pla: Adj: Pla: Adj:

How if I can just ask you please one or two points in clarification Sure. Yes that's what Wre're here for of the issue: ,s Yes, (0.9) Mm Hm This flooding I think occurred at two o'clock in the morning (0.4) On the eleven rth On: the eleventh ((clears throat)) This is true. And what- Did you go upstairs to find out what was happening?

When the adjudicator says, 'This flooding I think occurred at two o'clock in the morning', he is leading up to having the plaintiff give his testimony, i.e. tell what he did and what he saw in the flat with the running tap. The adjudicator leads up to and solicits the plaintiff's testimony by setting the scene. The scene is: it is two o'clock in the morning and flooding is occurring. In setting the scene, the adjudicator provides a starting point for the plaintiff's testimony. The plaintiff places himself in the scene and recalls his actions for testimony.



In providing a starting point for the plaintiffs testimony, the adjudicator formulates the time of the flooding as 'I think occurred at two o'clock in the morning'. In saying 'two o'clock in the morning', the adjudicator probably is repeating the time that was specified in the plaintiff's report. In including the uncertainty marker, 'I think' he is recognising the plaintiff as the participant in the event and as having first hand knowledge (see Pomerantz, 1980). By including an uncertainty marker, the adjudicator names the time as provisionally correct subject to confirmation by the plaintiff who is being treated as the authoritative source on the matter. In identifying the time of day with a Number, 'two o'clock in the morning', the adjudicator uses the preferred type of description for testifying. 3 Numbers are treated as representing an actuality, as 'factual'. The adjudicator gives the preferred type for testifying in setting the scene and soliciting the plaintiffs testimony. Because a Number is used in this instance, other materials are needed to show that Characterisations are not preferred and Numbers are in testifying. The next two instances demonstrate that Characterisations are seen as vague and second best to Numbers in testimony.

Excerpt 2-A: Sandals Adj:

Then you say you wore the shoes for a few weeks at the end of last summer. Wha. what's a few weeks. Ho- How many weeks roughly

The adjudicator is gathering facts concerning the length of time the plaintiff wore her sandals before they became defective. In her statement, the plaintiff had described the length of time as 'a few weeks at the end of last summer'. With this Characterisation, she was indicating that she was approximating the time and that it was notably short (in weeks, not months, and at the end of the summer, not the whole summer). The adjudicator asks the plaintiff to replace the Characterisation with a Number. In the question, 'how many weeks roughly', he asks for an approximation, perhaps appreciating that 'a few weeks' was offered as such. Yet for the purpose of the hearing, the approximation of 'a few weeks' is a different kind of object than an approximated Number of weeks. In asking 'what's a few weeks', the adjudicator is asking for clarification of a term that is regarded as 'vague'. 4 He is asking the plaintiff to specify a



Number (or Numbers) where the Number is treated as 'raw datum'. To the extent that 'a few weeks' is seen as the plaintiffs Characterisation of the raw datum, it may be treated by the court as something like an opinion or conclusion with an attempt made to solicit the raw datum in the form of a Number.

Excerpt 2-B: Flooded Council Flat When the defendant testifies, he describes the date that the Council completed renovation of flat nineteen as: 'a few days prior to the tenth of March'. Adj:

Def: Adj: Def:

Adj: Def: Adj: Adj: Def: Adj: Adj:

Well now if I can ask Mr F ehm If you can please uh tell me about the work that was being done and who was doing work. It seems to f me to be th' basic p r oint Yes Certainly. in this mat,ter Well. I think it's true to say that all the flats in this block were in the process of being renovated. Uhm Number nineteen, which was the one in question, had in fact been renovated and the renovation had been completed a few days prior to the tenth of March we can't say exactly but a few days before. That work has been carried out by:: the: Works Department of S Council. And that was done by the Works Department of ^Council of S Council Yes (•) Right. And that was finished on the Well a few days before the tenth is the nearest we can get to completion date Yes. Have you no work sheets which show where people were

On the two occasions that the defendant describes the date of completion, he offers something other than the date. The date, itself, is claimed to be not known to them: 'the renovation had been completed a few days prior to the tenth of March we can't say exactly but a few days before' and 'a few days before the tenth is the nearest we can get to completion date'.



T h e date of completion is oriented to as factual. A report of the date, using Official identifications and/or Numbers, would seem to represent the actuality. Instead of reporting the completion date either [day + month] or ' N days before the tenth', the defendant gives a seeably vaguer or less precise description: 'a few days before the tenth'. His accounts, 'we can't say exactly' and 'is the nearest we can get to completion date', recognise his r e p o r t as inferior to naming the completion date. In describing how many days prior to the tenth, the defendant might have used either Numbers (e.g. 'somewhere between 2 and 4 days prior to the t e n t h ' ) or a Characterisation as he did. In Characterising the number of days as 'a few days', the defendant proposes that while the completion date was not a large n u m b e r of days before the tenth, it was more than one day. T h e court's project is to assess, who most likely, left the tap running. In using a Relational identification, the defendant gives his sense of the import of the completion date. The import of knowing the last day that the Council w o r k m e n were working in Flat 19 is in how it relates to the time the tap was left running. To show the time when Council workmen were last in the flat in relationship to the date the tap was left running requires the selection of a Relational Identification the defendant offers a Relational Identification that proposes a gap between those two times of a 'few days'. The gap is an a r g u m e n t for the Council's not being the ones who left the tap running. In using the Relational identification, the defendant has analysed the import of the completion date for the issue at hand — establishing liability— and describes the completion date in a form designed to be used for that purpose. In Excerpt 2-A, the adjudicator attempted to have the plaintiff replace 'a few weeks' with a Number equivalent. In Excerpt 2-B, the adjudicator asks about the existence of work sheets, i.e. records that would contain items like the completion date. In both 2-A and 2-B, the Characterisations ('a few weeks' and 'a few before the tenth') given in testimony are treated as lacking and as only vaguely or imprecisely representing the facts.

Instance 3 Excerpt 3: Flooded Council Flat T h e adjudicator announces the judgement after both sides' evidence has b e e n heard. In Flooded Council Flat, the adjudicator prepares to announce



his judgement, which will be against the plaintiff, by presenting a case that supports the conclusion that the contracted firm, not the Council (the defendant), is responsible for the damage. Adj:

And Mr F states that those contractors came on the tenth of March, collected the keys from the caretaker in the morning, he believes, then went in the flat and returned the keys in the afternoon (0.5) of the tenth of March and the flooding of course occurred early in the morning of the eleventh of March. I think it is quite clear from the description of the flooding that was given by the caretaker to Mr M that the water could not have been flowing for too long because it seems that it flowed first . . . And that is presumably something that uh could have only occurred over a space of a few hours, rather than, say a few days. And so it is my finding . . .

In this excerpt, the time of the flooding is described as 'early in the morning of the eleventh of March'. 'Early in the morning' is a Characterisation of the time of day. The adjudicator previously had referred to the time of the flooding as 'two in the morning' using a Number for the hour. It was argued in Excerpt 2 that Numbers are preferred types of description in testifying, being seen as factual, and that Characterisations frequently are seen as vague, imprecise, and relying on judgements. The puzzle for this analysis is to answer why on this occasion the adjudicator becomes 'vaguer' and less precise, if you will, in Characterising the time. The adjudicator's concern at this point is to display reasoning that supports the conclusion that the contracted firm is responsible for the damage. The reasoning uses the factor of time: (1) figuring the time that the tap most likely would have been left running, and (2) determining who was there at that time. In the first part of his description, the adjudicator presents evidence that argues for the conclusion that the contractors were in flat nineteen on the tenth of March. He is presuming that the tenth of March was the critical day, the day that the tap was left running. And Mr F states that those contractors came on the tenth of March, collected the keys from the caretaker in the morning, he believes, then went in the flat and returned the keys in the afternoon (0.5) of the tenth March.



H e then backtracks to work out when the tap would have been left running. This involves subtraction. [The time of the flooding in the plaintiffs flat] - [The amount of time the flooding would have taken] [The time the tap was left on] The time of the flooding in the plaintiffs flat is recalled as a fact already established and known in the hearing: 'And the flooding of course occurred early in the morning of the eleventh of March'. The adjudicator approximates the amount of time the flooding would have taken to occur:'. . .the water would not have been flowing for too long . . . . And that is presumably something that could have only occurred over a space of a few hours, rather than, say a few days.' Putting his terms in the subtraction problem yields: [The flooding occurred early in the morning of the elventh of March] - [Water flowing . . . a few hours rather than say a few days] [ANSWER] The adjudicator does not articulate an answer. He is treating the tenth of March as the only reasonable possibility of when the tap was left on. His calculations are designed to have his recipients conclude that the tenth of March is the only plausible time for it to have happened. 5 This interest governs the type of identifications and descriptions used in his calculations. To make the tenth of March into the only plausible time, the adjudicator dismissed the possibility that the flooding took more than one day. In saying, 'over the space of a few hours, rather than, say a few days', he is claiming that any number of days would be an inappropriate measure of the amount of time that the flooding took; that the appropriate measure would be a number of hours. T o describe the time of the flooding in the plaintiff's flat, the adjudicator uses a Characterisation for the time or hour of the day, 'early in the morning'. For the length of time the flooding took he again uses a Characterisation for the number of hours, 'a few hours'. In describing these hours with Characterisations instead of with Numbers, he eliminates the possibility of figuring the time to hours that the tap was left on. He structures the figuring to be rounded off to the nearest whole day. Had he used Numbers, the answer may have looked something like this:



[The flooding occurred at two in the morning of the eleventh of March] - [The water flowing . . . between two and eight hours] [The tap was left on between 6 p.m. and midnight of the tenth of March] To set up the conclusion that the tenth of March was when the tap was left running, he dismisses any number of days as an appropriate measure of the flooding time and eliminates the possibility of calculating the time of the day in hours by his use of the Characterisations, 'Early in the morning' and 'a few hours'. The Characterisations are used as 'rough approximations' and are not to be used in calculations. By Characterising the hours, he leaves it for the recipients to round off to the next larger unit, days. Although the times were characterised with hours, the answer is given as a day. Rather than having doubt about the approximate time that the tap was left running, the adjudicator structures the problem for a conclusion with certainty as to which day it was left on — namely, the day before the night of the flooding.

Instance 4 Excerpt 4: Flooded Council Flat The adjudicator ruled against the plaintiff due to a procedural matter. The plaintiff had named the Council as defendant rather than both the Council and the contractors as co-defendants. The adjudicator ruled against the plaintiff on the grounds that the Council was not liable for the damage. Before announcing the judgement, the adjudicator describes the plaintiff as a blameless victim who experienced considerable undeserved loss and suffering as a result of the flooding. 6 Adj:

Now this is a case where one must have a good deal of sympathy for Mr M. He's been a tenant of S Council for a very long time, and here we have an incident in March last year where early in the morning water starts coming in through the ceiling of their flat and causes damage to their carpets and their decorations and perhaps causes uh further ill health to Mr M's wife. And clearly whoever is to bla:me for this occurrence, it's not Mr and Mrs M and they are the people who've suffered from it. But the law unfortunately cannot always take the side of those who are blameless . . .



T h e adjudicator begins describing the plaintiff as victim by formulating an appropriate reaction: 'one must have a good deal of sympathy for Mr M'. H e then goes on to describe what happened to Mr M. The description starts as follows: H e ' s been a tenant of S Council for a very long time, and here we have an incident in March last year where early in the morning water . . . (emphasis added) W h e n the flooding occurred is described as 'in March last year. . . early in the morning'. 'In March last year' is not an Official identification, despite its resemblance, but a Characterisation. 'Early in the morning' also is a Characterisation. These Characterisations are produced as part of the adjudicator's showing sympathy. T h e three underlined descriptions will be discussed in turn.

'for a very long time' In saying, 'He's been a t e n a n t . . . for a very long time', the adjudicator Characterises the amount of time rather than reporting a Number of years or a d a t e that the tenancy started (see Excerpt 4 - A ) . With the Characterisation, he displays a view or attitude towards the plaintiffs circumstances. The formulation of the plaintiff as tenant in the flat 'for a very long time' is a favourable one: it suggests a person who is stable, has shown commitment to his place of residence and is, in short, a good tenant. By portraying the plaintiff as a very good tenant, the adjudicator provides for 'injustice' to be seen, where the suffering and loss has happened to a person who particularly does not deserve it, i.e. this very good tenant. T h e differing uses of Characterisations versus Numbers and Official identifications is sharply illustrated in this next excerpt drawn from an earlier portion of the same hearing.

Excerpt 4-A: Flooded Council Flat Pla:

Def: Pla: Def:

because these flats have been tenanted by so many people My wife was an original one in wa in 1934. That's um I've been a tenant since That's r i f g h t ( ) about 1939 We know you you've been tenants for very long standing



The plaintiff is presenting facts that argue for their being exceptionally good, solid tentants, unlike so many of the others who were short term. To this end, he offers Official identifications of the years that their tenancy started. 7 T h e year designations are presented as evidence that substantiates a point that is not articulated. If a point is not recognised, there is an incompleteness in the interaction. T h e defendant claims to already know and appreciate the import of the p l a i n t i f f s tenancy having started in 1934 and 1939: 'We know you you've b e e n tenants of very long standing'. In giving a sense of the dates and in claiming to have already known it, the defendant indicates that the plaintiffs point has been made and no further elaboration is necessary.

'in March last year' In introducing the facts, a Complete identification of the date is made (Excerpt l ' s 'on the eleventh of March last year'). On subsequent occasions, the date is identified with either just the day (Excerpt 2's 'On the eleventh') o r the day and the month (Excerpt 3's 'on the tenth of March'). The description on this occasion, 'in March last year', includes no day designation. T h e question arises whether 'in March last year' is significantly different f r o m 'on the eleventh of March last year'. Just as 'a few' as Characterisation of a quantity is an alternative to a N u m b e r , 'In March last year' as a Characterisation of when is an alternative to naming the date. 'In March' is an approximation of when it happened, a vague or imprecise way of describing when. With 'in March last year', the adjudicator proposes that exactly when it occurred is not relevant. In Characterising when it happened, he is giving an indication of when it happened without naming the date. In so doing, he provides for the when of the event not being central for the event. He is providing for the when of event to be irrelevant for an understanding of the event. Giving the date of the flooding and characterising when the flooding occurred are used quite differently. Giving the date is part of being factual and precise, part of presenting evidence. Characterising when it happened is d o n e when the adjudicator is understanding what sort of experience it was f o r the plaintiff, when he is showing sympathy. It appears that speaking factually is inappropriate and, indeed, may be incompatible with being sympathetic.



'early in the morning' In Characterising the time of day that the flooding took place, the adjudicator provides an understanding of what 'two in the morning' was for the plaintiff, i.e. early morning. 'Early in the morning' is a particularly inconvenient time to have flooding occur in one's flat. In Characterising the time of day, the adjudicator is able to provide a sense of the event, where that sense adds to the portrayal of the plaintiff as having suffered. In selecting a Characterisation, the adjudicator displays an understanding of the time, where showing an understanding of 'what it was for you' may be part of what constitutes displaying sympathy. Again, being factual may be incompatible with showing sympathy.

Conclusion The aim of this Chapter has been to provide some analytic distinctions that relate to courtroom activities. The organisation of the Anglo-American legal system is built on a division of responsibility : the litigants present their evidence, i.e. describe the facts as they know them, and the decision maker evaluates the evidence. The analytic distinctions offered here may help to refine the accepted dichotomy between fact and opinion. In testimony, descriptions representing the facts are looked for. When a quantity is described with a Characterisation, it is recognised as vague and imprecise. When a reference is made with a Relational identification, the referent is not seen to be properly and definitively identified. The preference for facts involves a preference for description types that are seen to represent an actuality. Numbers and Official identifications are seen to do that. This view, however, does not appreciate the work that Relational indentifications and Characterisation do in discourse. Both Relational identifications and Characterisations are a fundamental part of sensemaking machinery. They are used to provide a sense to, or coherence in, what is being described. Relational identifications are used to show how people, things, events, actions, etc. are related to each other. They are used to display a speaker's orientation to the purpose at hand, the current project. Characterisations are used to show understanding of events. They are used as focussing devices to indicate, from the speaker's point of view, what is relevant and how it is relevant.



W h e n litigants arrive in t h e c o u r t r o o m , they bring their resources f o r p r o v i d i n g c o h e r e n t a c c o u n t s , p r e s e n t i n g e v i d e n c e , m a k i n g a r g u m e n t s , etc. T h e i r u s e of t h e s e r e s o u r c e s clearly is n o t limited t o c o u r t r o o m settings. 8 T h e a n a l y t i c d i s t i n c t i o n s p r e s e n t e d in this C h a p t e r h o p e f u l l y illuminate types of d e s c r i p t i o n s in social action across a variety of settings. T h e social actions i n c l u d e : i n t r o d u c t i o n s a n d clarifications, providing f o r a recipient t o arrive a t ' t h e o b v i o u s a n s w e r ' a n d showing s y m p a t h y .

Acknowledgements I w i s h t o t h a n k J o h n B o a l f o r his m a n y t h o u g h t p r o v o k i n g c o m m e n t s , W i l l i a m O ' B a r r f o r his n u m e r o u s editorial suggestions, a n d J. M . A t k i n s o n f o r his e n c o u r a g e m e n t .

Notes to Chapter 9 1. In fact, the form of an Official identification is sensitive to the circumstances in which it is used. For example, in initially identifying the date, the adjudicator used a Complete Official identification. In clarifying, the participants offered only the component that was subject to confusion, e.g. 'on the eleventh'. Moreover, in general, components that are presumed known are not included, e.g. what is the rest of the address of 'Flat 19?' 2. Both types of identifications may be called for, each to accomplish its own work. Q: What time was that you put this in the locker? A: This would be approximately 10:20,10:30 that evening. Q: Prior to putting in the locker, had that can been out of your possession at any time? A: No, sir, it hadn't. Q: When did you next see this can? A: At approximately 10 o'clock the following morning. Q: That would be on January 26th? A: Yes, sir. (Lempert & Saltzburg, 1982: 35.) In this excerpt, A uses a Relational identification ('the following morning') to identify an event temporally in relation to the event previously talked about. In giving the events a sequence, A provides a sense or coherence to the actions described. Q transforms the Rational identification to an Official one ('January 26th') to provide a definitive, unambiguous identification of the date for the record. 3. Descriptions seen as other than reports of facts are produced in testimony, but reportedly these are tolerated or accepted on the grounds of expediency. In large measure the opinion rule is a rule of preference. Where specificity is possible, it is preferred to less descriptive conclusions. Where specificity is not possible, more conclusory statements will be allowed so long as they promise to aid the trier of fact. (Lempert & Saltzburg, 1982:43.)



5. 6.

7. 8.


An opinion is an inference from observed facts. Since most human discourse is largely made up of opinions, an insistence that no statements of opinions be made would be unworkable . . . non-expert opinions must be admitted where this is convenient in the interests of a reasonably normal prose during the giving of testimony. (Heydon, 1975: 369.) Sacks proposed that it is through the use of 'vague' characterisations that people exhibit their competence at observing scenes. . . . there can be various ways to characterise how long you were parked in a traffic jam — 'just a second', 'seemed like for hours', 'quite awhile', and various things like that. When you pick one of these, for example, instead of saying 'we were parked there for twenty five minutes', then what you're doing is involving the other in appreciating that the 'how long you were parked there' was known by you, by reference to how long people are parked at various sorts of traffic j a m s . . . It is, in a fashion, better not to use a 'precise' characterisation, which can be equivocal in the sense of, is that a long time for a traffic jam . . . ? Instead, what one does is offer the product of an educated a n a l y s i s . . . So that this possible 'vagueness' of the report — 'Quite a while', well how long was it? . . . — is not a defective kind of vagueness but is the way to show that you measured the thing in an appropriate way to measure being caught in a traffic jam . . . (Sacks, 1983:17-19.) Paterson (1982) proposes that when a hearer makes the inference himself, he is more likely to be convinced of the soundness of it. The adjudicator's sense of justice may not have been satisfied by his ruling in this case. In addition to expressing sympathy prior to announcing the judgement, he closed the case by encouraging the plaintiff to seek legal counsel and take further legal action. In giving the years, the plaintiff provides the 'raw data' with which to get a sense of length of time that they have been tenants there. Giving the year (without day or month) suffices for getting the sense. Even something like reporting 'just the facts' or the raw data is a conversational practice when participants talk about sensitive, incriminating, or controversial matters. (Pomerantz, 1981.)

10 Work flow in a paediatric clinic WES SHARROCK University of Manchester and BOB ANDERSON Manchester Polytechnic

By and large, discussions of the role which Conversation Analysis could play within sociology have been grouped around three general themes. First, there is the question of method and data. Here, it is sometimes felt that devotees of Conversation Analysis seem to be arguing, or at least implying, that the analysis of transcripts and similar materials is the only proper procedure for sociology to adopt. And, to be fair, there are some reasons for this impression. Occasionally, expositions and defences of Conversation Analysis do lend themselves to being read as claims for the general applicability of its methods. In our view, such claims when they are made, are both misguided and misleading. As we have made quite clear elsewhere (see this volume, Chapter 12 and Anderson, Hughes & Sharrock, 1985), we are very suspicious of anything which smacks of methodological imperialism. The relationship between 'object of enquiry' and 'investigative method', both in sociology in general and Conversation Analysis in particular, is a tangled clump of barely glimpsed, let alone solved, problems. However, enticing though it might be, for the moment we will set this particular Gordian knot on one side. Second, there is the concern over the legitimacy of focussing exclusively on talk and conversation pure and simple, without reference to thematics, parameters or variables such as race, class, gender, power and so on which, it is held, provide the institutional nexus in which ordinary talk is located. Although we will not seek to elaborate or defend it here (a defence can be found in Sharrock & Anderson, 1986), our view on this question is quite straightforward. 244



Conversation Analysis' disregard of 'the wider social structure', as the issue is often, we think confusingly, put, follows directly from the analytical goals which it sets itself. Given those goals, the strategy is perfectly acceptable since it is adopted to solve only problems faced in Conversation Analysis and not all the problems to be found in sociology. Third, and here we come to the issues which we do want to take up in this Chapter, are the supposed limitations of Conversation Analysis. If we grant the extension of 'ethnomethodological indifference' (this expression is defined and explained in Garfinkel & Sacks, 1970) to Conversation Analysis, is it thereby precluded from addressing any of the topics generally studied by sociology? Is there no way that the concerns and orientations displayed by Conversation Analysis can be turned towards questions tackled by more conventional sociologies? We will argue that there is. However, and the point cannot be over-stressed, this is not to say that Conversation Analysis can be incorporated wholesale into the rest of sociology by means of flexing a few frameworks and a little conceptual jiggling. To think that is to misunderstand throroughly and comprehensively what motivates Conversation Analysis in the first place. Our aim in this Chapter will be to show what we mean by suggesting that Conversation Analysis can be turned towards familiar sociological topics. We will do so by exploring in a preliminary manner what is involved in opening up one specific area in the sociology of occupational life to the sensibilities displayed in Conversation Analysis. To do this we will need to begin by sketching very briefly what we take such sensibilities to be. We will then show the purchase they give on the investigation of one particular occupation. The case we take up is the managing of work tasks in a paediatric consultation. Two stock ideas are by now almost emblematic of studies in Conversation Analysis. 1 One is the notion of there being a 'detailed orderliness' to the organisation and operation of what are called 'speech exchange systems'. These systems are to be seen in the characteristic forms of talk found in ordinary conversation, judicial proceedings, class-room teaching, telephone calls to service agencies, therapy sessions, and the like. The other is the proposal that this detailed orderliness is 'oriented to' by the participants to the talk and that their orientation to it is visible in the talk itself. The claim is that the duplets, triplets and even more elaborated structures which analysts have documented, are both the resource for, and the products of, the activities which the co-participants to the talk engage in. The work that has been done to elucidate the nature of turn taking systems, remedial exchanges and repairs, the composition of more global structures such as openings, closings, story telling and so on, and the connection of these features to other sets of conversational objects like topic organisation and



category usage, is directed to drawing out the principles of organisation whereby speakers operate the speech exchange systems and produce the structures which they do. O n e of the consequences of pursuing the demonstration of this detailed orderliness, and its 'oriented' to character, in this way has been the introduction of a separation between what can be observed about the talk and associated activities simply because one has a transcript to hand, and what we might think of as the talk's 'real world' contingencies. Such a separation has meant that consideration of stretches of talk has tended to move toward what are essentially transcriber/analyst observable features, and away from features readily observable by actors. The result is that Conversation Analysis necessarily disattends to what actors may see as the business of their talk, in favour of the activities which actors engage in solely by virtue of their character as operators of a speech exchange system. It is the orderliness which this entails which increasingly has become Conversation Analysis' topic. We should emphasise, though, that this does not mean we think there is a need to look at other data in addition to transcripts. Rather, what we are pointing to is the way that the attitude shown towards the transcript has changed and evolved. It is important to notice that we describe this analytic separation as a necessary one. It would not be possible for Conversation Analysis to attend to the fine grained texture of the structures of speech exchange systems and at the same time maintain an overall view of the character of the particular sets of interaction in hand. It is simply because analysts were able to compare fragments of talk independently of their original locale that Conversation Analysis had been able to develop as rapidly as it has. The question to be asked, then, can be formulated like this: 'Can the motivations underlying the separation we have identified be brought to bear upon other activities in daily life?' In seeking to answer this question, we hope to be able to show that it is possible to extend the analytic interests which inspired Conversation Analysis beyond conversation and other speech exchange systems to different forms of interaction and different types of activities. W h e n this has been achieved, it becomes possible to identify and display an orientation to the production and reproduction of detailed orderliness in the routine organisation of activities and settings. As we indicated above, we will begin by drawing out in more detail what it is that motivates Conversation Analysis. Only then will we be able to relate those motivations to the aims we have just set out. How is the idea of 'an orderliness' conceived in Conversation Analysis and just what does 'oriented to' designate? We can get some grip on these issues if we turn back



to the materials which Conversation Analysis utilises and the approach which it adopts to them. For the sociologist, and that is a very important qualification since linguists, psychologists, discourse analysts are all very different from sociologists in their approach both to the nature of language and the character or ordinary talk, what more than anything else Conversation Analysis has been able to do is to demonstrate the analytic possibilities inherent in the data of daily life. Yet again the terms are important. Conversation Analysis directly addresses data generated in daily life not data on daily life. Transcripts cull talk from social interactions of various kinds and are not strictly comparable to the reports, résumés, indices and summary variables on daily life which are more usually the sociologist's stock in trade. What Conversation Analysis has been able to do is show how much can be done with materials like transcribed talk and just what a fertile locale for sociological reflection it provides. But, and this is central to its analytical attitude, it has achieved this by taking what it regards as an unmotivated interest in its materials. That is to say, it does not try to specify in advance what range of analytic problems a set of transcripts will be pertinent to, nor just what phenomena will be revealed in the analysis of the talk. Each transcript is inspected for what it contains, what its structures are, and how its features can be made visible and analysable. The orderliness which Conversation Analysis depicts is that discernable in the unnoticed, taken for granted, flotsam and jetsam of talk in all our ordinary, daily lives. Showing how that orderliness underpins the routineness of this aspect of our daily lives is precisely to show just how valuable such materials can be for sociological analysis. The approach to the materials is one which does not minimise their mundane character but explores and exploits it. They do not stand to the analysis as evidence or grounds for inference, but as the object of enquiry. In the last analysis, the question we have asked is how this same attitude can be extended beyond talk in therapy sessions and classrooms, calls to the police and radio chat shows, to such other common or garden activities as organising the things to be done during the working day.2 We said a few moments ago that Conversation Analysis treats all talk as the instantiation of one or other speech exchange system. The concern is to bring out the texture of the system's operation. It does so by insisting that each transcript exemplifies an idiosyncratic and highly contextual specification of what are very general structures. The context to which reference is almost often made is that of the conversation-so-far and the activities both that have been undertaken, and those that are projected within the talk. The structures in the talk are, therefore, designated as 'local productions' or 'locally produced'. What this means is that the remedial exchanges, the greeting given, the stories told, the explanations undertaken all have the



shape which they do as in situ applications of general principles of organisation to universally available formats. The axiom with which Conversation Analysis begins is that the social organisation of any speech exchange system is managed by co-participants within the settings in which it is found. Ordinary conversation, trial by jury, classroom teaching are, then, locally managed productions. There is a further aspect to all of this which we can now bring out. Following and applying one of Garfinkel's 'study policies' (Garfinkel, 1967), Conversation Analysis treats the locally managed production features of talk, that is the structures observable in the transcript, as routinely matters of course for participants. Indeed, it was the emphasis which could be given to the sheer orderliness of these structures and their organisation which drew ethnomethodologists to studies of conversation in the first place. It is visibly and routinely a matter of course that questions are asked and answered, corrections are given, requests made, descriptions offered, and the like. Conversational events such as these constitute an unnoticed part of the daily round. They are unnoticed simply because in daily life they are essentially unnoticeable. The system only works because the structures and forms of organisation which comprise it are taken for granted. What the analyst takes to be problematic, i.e. how it all works, the participants take for granted. In so doing, the participants endow the systems with an objective facticity which the analyst must, perforce, scrutinise. From the point of view of every participant to talk, the social organisation of conversation as a speech exchange system is experienced as a normative social order. For things to get done, for aims to be achieved, activities have to be carried out in certain, predictable ways. In orienting to their order, in accommodating it, managing it within daily life, social actors, the speaker/hearer operators of the speech exchange system, reproduce that system as a system of socially organised activities. In demonstrating this through the examination of actual cases, Conversation Analysis has made a unique contribution to sociology. By addressing its materials in the way which it has chosen to do, it has documented the specifics of a working solution to the problem of social organisation as the outcome of the routine features of our daily lives. All of the elements which we have just been discussing may be crystallised in what is, by now, a famous contra-distinction, namely that between context sensitivity and context independence (Sacks, Schegloff & Jefferson, 1974). This pairing is used to catch how standardised forms and general resources may be shaped within particular circumstances to achieve, there and then, the precise outcomes they do. One could say that the crux of any analysis is to bring to the fore through the examination of particularities, just what this localising work consists in as a 'this time through' phenomenon. 3



T h e latter term demands some explanation. Whatever orderliness a set of data may display has to be seen as the achievement of a specific collection of participants in a defined set of circumstances with the resources which they have to hand. Every interaction is, in this analytic sense, unique. Its unique character is what makes it recognisably what it is for the participants. The notion of 'this time through' points to this recognisability. A call may be the first or the nth that the speakers have had; the topic may be being raised for the first or the nth time, and so on. What they know about each other, what they know about the activities to be engaged in, all reflect the 'this time through' character of the interaction which the co-participants produce. It is precisely this which is meant by the suggestion that the setting or context of talk, or any activity, is 'self-explicating' (see Pollner, 1979). What motivates Conversation Analysis, then, is the determination to make the local production of the routine features of speech exchange systems visible as forms of social organisation. The character of this orderliness is, as we said, mundane, taken-for-granted, ordinary. It is recognised and oriented to as such by participants. This is the sensibility we will carry over to the study of occupational life. Armed with it, we will seek ways of depicting how occupational activities are performed within an organisational environment. At the same time, placing emphasis upon the importance of treating settings as 'self-explicating', we will endeavour to do so without counterposing the performance of the activities and the environment in which they are to be found. What has been 'done so far' and what has 'yet to be accomplished' constitute part of the environment of the activities 'presently being engaged in'. Approaching things in this way, we will be able to show how the ordinary orderliness of the distribution of a set of work tasks is the display and management of the constraints of organisational life. As we will see in the case we take up, such constraints are visible in the way that the schedule is kept ticking over, the notes and files are kept in order, legal and ethical requirements are respected, colleagues are inconvenienced as little as possible, and so on. The transcript to which we will refer is taken from one of a number of videotaped recordings which we made of a paediatric clinic while we were collecting materials on the utility of video-tape for sociological analysis.4 Even the most general glance at the tapes would show two obvious features. First, the consultation is composed of a number of concatenated episodes. 5 A form is signed; the child is examined; the mother asks questions about some possible symptoms; and so on. Second, and equally obviously, all of the activities to be done are, for the doctor at least, part and parcel of 'just another working day'. That this consultation is 'just another one' is what its orderliness consists in. This consultation is no different, in the broadest



terms, from many others which he will have carried out. There is nothing special about it. And yet, despite its absolutely routine character and its similarity to so many others, it does not have the sense of being 'just run through' or that the doctor and the mother are merely 'going through the motions'. What comes across powerfully is how the medical tasks are arranged so that this child's development can be discussed and this mother can have time to talk about any possible problems which she has. In this instance at least, medical practice is organised so that the doctor can routinely deal with the child, and the mother can make the most of her time there. The 'just another one' character of the consultation is produced through management of the specifics of 'this instance'. The organisation by which this is achieved is what we have called 'the flow of work tasks'. Since it is the doctor who must, albeit in collaboration with the mother, achieve this work flow, in our discussion we will focus upon the work which he does. We have no doubt that the transcript could be used, from the mother's viewpoint, to give an analysis of the work of 'bringing professional expertise to bear upon your problems'. But that is not our concern here. Let us think for a few moments about the management of a flow of work activities.. We said that we would be interested in this flow as a selfexplicating, locally managed production. The orderliness which such a flow can display may take many different forms. As we shall see in a moment, this one consists in running through a collection of tasks in a quite routine manner. Such tasks have no particular order built-in to them, but nonetheless, on any occasion, they will have to be done in some specific order. T h e order in which they are taken, their distribution, is their organisation. Showing what is involved in organising a set of work tasks so that they can be achieved in as efficient and smooth a manner as possible, means directing attention towards the situational contingencies which bear upon the activity of making the work flow. We suggest that it is in the opportunistic handling of such contingencies that the routine character of the work resides. For medical practice, the contingencies we have referred to might be encapsulated as the spectrum of demands which patients and doctors can make on one another, the sets of medical and non-medical tasks which have to be carried out; the allocation of time to particular phases and tasks; as well, of course, as the requirement to keep this 'case' and all the others 'moving'. These contingencies are visible in the transcript as a set of oriented to constraints and are managed by the distribution of tasks which, in turn, constitutes the normative order of 'good-medical-practice-on-this-occasion'. It is in accommodating the constraints that the orderliness of the flow of work tasks becomes visible.



We said that the immediate character of the consultation was that it was 'just another one' in what is a routine day. This is just another child to be seen, just another mother to be talked to. Business is pretty much as usual. The processing of these routine cases as routine cases take place within a particular institutional setting which we will call 'bureaucratic medicine'. By this term we mean to pick out both of the primary features of these encounters, the fact that a particular sort of occupation is being carried on here and that this occupation relies to some extent upon non-occupationally defined procedures. The institutional locale of the consultation means that there are other children to see, other mothers to talk to, and each has the right to just as much medical attention and concern as any other. In order to be able to get through the cases and satisfy the demands, the clinic has instituted sets of non-medical procedures. Mothers must check in when they arrive; the appointment rota must be adhered to; files are 'got out' in the order patients are dealt with; there is a division of labour in which nonmedical and paramedical staff undertake certain tasks. These procedures carry their own requirements. The doctor has to keep the files 'up to date' and 'in order'. The notes have to be read and returned. To allow them to manage the tasks which they have been allocated under the division of labour, the para and non-medical staff have to be able to predict the distribution of patients through the day and thus the times when their services will be called upon. The appointment system with its pre-defined turn-taking distribution allocates both order of patients and rough timing. All of these 'obvious' things comprise the larger context of work routines on any ordinary working day for this doctor. Keeping the appointment system working, that is processing the cases in the order and in the time allocated, while at the same time attending to the needs of each child and mother is what paediatric medical practice comes to on this, and on any, occasion. It is routine bureaucratic medicine; no more, no less. Part of the demands of bureaucratic medicine are, of course, those of professional expertise. All of the appointments in this clinic are checks of one sort or another on infants. Such checks involve a number of things being done both to mark the child's progress and to monitor for early developmental difficulties. As we indicated, not all of these are carried out by the doctor. Others will weigh and measure the baby and give it immunisations. Processing any individual case means more than simply 'doing things to the baby'. It means passing the mother and baby 'along the line' from stage to stage. Only one of these stages is the consultation with the doctor. What we have here, then, are specific work tasks carried out within an overall process, namely the 'appointment at the clinic'. Things are done before and after the mother and baby see the doctor. In like manner, the consultation



with the doctor for the check up, consists in sets of things to be done. The accomplishment of these things as part of the overall processing of mother and baby is what the consultation is properly about. In addition to all the matters which we have mentioned, there is another set of constraints which are equally as important. These are the ethical and legal requirements to which the doctor must submit. These requirements allocate expectations to both participants. The mother can expect the doctor's attention and interest; the doctor can expect to be given information which is relevant to efficient diagnosis. The ecology of the clinic can be viewed as designed to foster these expectations and hence to help satisfy the requirements. The consulting room is separated from the waiting room by a closed door. Other staff enter the room by another door and tiptoe in to deliver files and instruments. Telephone calls are held at the switchboard. The doctor sits facing the mother who is positioned at the side of his desk. M o t h e r and babies are seen one pair at a time. O u r aim is to show that just as the ecology of the clinic could be analysed to show how its organisation displays the requirements of bureaucratic medical practice, i.e. is self-explicating, so these requirements are also visible in and taken for granted by the managed organisation of the activities which make up the consultation. We will suggest that while the activities which are carried out are just what anyone would expect and give the encounter its routine character, nonetheless their organisation on this occasion is contextually specific and idiosyncratic. It is a locally managed production. We will try to show that the work of this local management can best be summarised as the application of a general principle of opportunism. This principle has much the same status as the principles of simplicity and economy do within Conversation Analysis (see for example, Sacks & Schegloff, 1979: 15-21). It allows us to make sense of the specific character of the activities on view in the transcript by providing a rationale to the order they display. If we use the principle as a guide, we can see readily recognisable formats and structures embedded in the pattern of the activities as a flow of work tasks unfolds. Under the principle of opportunism, social actors are deemed to seek to take advantage of whatever resources are available in the interaction to achieve whatever can be accomplished at any point during the encounter. What we see the principle achieving is the characteristic shapeliness that the flow of work has. Each activity seems to merge with those that precede and follow it effortlessly and naturally, producing what appears as an almost 'seamless' construction. 6 T o illustrate what we mean by the achievement of this natural flow of work tasks, let us take just one section of the transcript and try to bring out



its self-explicating and locally managed character. The example we have chosen is the work involved in 'getting the consultation underway'. From the following fragment, it is apparent that by utterance 2.2, the business of the consultation has begun. 1. 1 D:

1. 2 M: 1. 3 D: 1. 4 M: 1. 5 D:

1. 6 1. 7 1. 8 1. 9 1.10 1.11

M: D: M: D: M: D:

1.12 M: 1.13 D: 1.14 1.15 1.16 1.17

M: D: M: D:

1.18 M: 2. 1 D:

2. 2 M:

It's just e r - e r by way o f - u m - c o n s e n t for t h i s - t o be done . . . OK? ((1.00)) just ( ) ((laugh)) I'll hold it and you can write ((7.00)) right thank you very much ((4.00)) ( ) Mrs-*** is it. . . n o t M r s - . . .thatthat's her - that's the baby's name A *** -yes-//yes Oh I see - I got it the wrong way round ((3.00)) it's usually the case we always seem to do that ((short laugh)) ((5.00)) r i g h t - now then ((1.00)) this young lady is now? ((1.0)) he's six months Six months Yes ((1.00)) it's a young man Yes Six months ((both laugh)) we're going to get it sorted out eventually // don't worry Yes ((6.00)) Right - now - um - you - he he's supposed to be having a six month check today is he Ye yes And you were also thinking of starting the immunisations Yes I s e e - O K ((1.00)) right-well let's start we'll do t h e e r have a look at him first Mmm And then we'll think about the immunisations ((14.00)) Any problems at all - Y e s - y o u know he's sweating a lot on the h e a d - j u s t the head - on the sides - and some days um - it's not even warm he's sweating



A t utterance 2.2 we could say that proper medical topics in the shape of a set of potential or candidate symptoms have been introduced. What we are interested in is the work which the doctor engages in to be able to 'get to' 7 these symptoms, that is the organisation of the serial ordering of the activities which are undertaken. This is the work of getting the consultation underway. W e could hazard that whatever things a doctor has to do in a check up of this sort could be taken in any order, even though, as medically relevant m a t t e r s , they might vary enormously in their import. The child's eyes and ears might b e examined before its breathing or motor skills, or afterwards, without very much turning upon that placement. This, of course, might not b e t r u e of other lines of medical enquiry where a critical path of diagnosis might be involved. So, although any task might be done first, there is still the crucial organisational problem of selecting one to do now, getting it accomplished, and then fitting others in with it. This is the nub of the 'getting to' p r o b l e m . T o refer back to the organisational constraint of speedy and satisfactory processing of cases for a moment, we can see that if the doctor can develop methods of solving the 'getting to' problem in collaboration with the mother then that solution would also contribute towards the more general requirement of efficient and personalised processing of cases. Collaboratively getting things underway and keeping them going prevents either a backlog of cases building up or interruptions to the scheduling of para-medical tasks, and much more besides. All of this is, of course, staring us in the face in the transcript and is no news to anyone. That ready recognisability is its character. However, what is of interest in any specific case is how the 'getting to' problem is solved as a locally managed accomplishment. The solution the doctor and mother produce in our data, for e x a m p l e , satisfies two sets of constraints which have primacy for this occasion. They are those of medical and professional ethics and bureaucratic efficiency. Satisfying the requirements of proper professional practice and efficient processing of cases achieves the serial organisation of the medical tasks. Let us now see how. In a research environment like that of a doctor's consulting room, it is s t a n d a r d practice to elicit the patient's permission for a consultation to be observed and recorded. The precise procedure which the research team had developed was to have those who were happy to have their consultation v i d e o t a p e d sign a consent form. If permission was refused, the camera crew withdrew from the room. The signing or otherwise of the form is, then, a pre-condition for what happens next. Either camera keeps rolling or the researchers leave the room. But, more important than this, although signing



the form is one of a number of things which have to be done, it is something which should properly be done first. Without the patients's permission being asked and given, the consultation-as-envisaged, that is as 'another one to be recorded', could not proceed. So, from among the range of things which could be done first, there is one which necessarily must come first. Without it, the consultation's format as the efficient processing of 'this case on this occasion' would have been threatened. The second aspect to all of this is the administrative locale. The mother and baby's physical progress through the clinic is paralleled by a bureaucratic processing of their file. The doctor has to see if this processing has been carried out correctly. That is, he has to check that the file he has to hand is the one which contains the baby's history and notes. The principle of opportunism and economy which we outlined above is evident here. The doctor tries to use the signing of the form as an opportunity to check the file. T h e name on the form is matched against that on the file. What makes the activity visible in this case is that the matching goes wrong and a repair sequence is initiated. The child's name is corrected. The need to engage in a remedial exchange to determine the child's name brings out both the importance of achieving relative efficiency and the interactional dexterity needed to do two things at once. The name is corrected and the file is checked. At the same time it provides an interesting instance of how far Conversation Analysis' concerns and the ones we outlined as belonging to more conventional sociology, mesh. The generalised structure of correction solicitor/ correction/acknowledgement is used here to satisfy a requirement of bureaucratic medicine. We will come back to this in a moment. Once the name has been determined, the doctor can use the materials available in the file to get the consultation underway. However, rather than do this, the doctor asks for information which he might easily have gleaned from the notes, namely the age of the child. It is important not to get over-Machiavellian with observations of this sort. We do not want to offer a strategic reason for everything the doctor does, as if all his moves were planned out in advance and plotted together as a set of interconnected ploys. H e may not have any reason for asking the mother the baby's age rather than looking for it in the file. What we want to say, though, is that having asked for it, the answer he is provided with is a resource for undertaking the next relevant activity namely the determination of the reason for the appointment. The determination of the age provides the medically relevant reason. This is a six month check up and the course of immunisations could be started today. Once again, the organisational detail will show how Conversation Analysis' resources can be turned towards the familiar sociological topic of medical practice.



We take it that while 'six months old' is an age category,8 it is not just offered as an age for the child. 'Six months old' is also a stage of development category similar to 'new born', 'crawling', 'just learning to speak', 'toddler' which can be used about children. Some of these stages are medically relevant ones, others are not. What the mother offers is a category which is medically relevant because at 'a few days old', 'six months', 'pre-school' stages medical checks are made on the child's development. When asked for the child's age, the mother offers one of the medically relevant stages and so provides the reason for the appointment, and also why she is seeing the doctor. The pair of correction solicitors, in the form of questions, at utterances 1.13 and 1.15 confirm the finding. These structures, amply described by Conversation Analysis are being used to provide an organisation to work routines. 1.13 D:

Right - now - um you - he he's supposed to be having a

1.14 M: 1.15 D: 1.16 M:

six month check today is he Yes yes And you were also thinking of starting the immunisations Yes

We said, before, that we do not want to make too much of these observations. We make no claims on their behalf. They are what anyone can find in the talk and what the participants did find in it. But that is the important point. The recognisability of 'what is being done' in what is said as routinely part and parcel of 'just another consultation' is what allows us to talk of this readily comprehensible and visible organisation of activities as self-explicating. The orderliness is also oriented to by them. They arrange the tasks in the order in which we find them and so get down to 'proper medical business' and 'get the consultation underway'. What we are pointing to is not the fact of what they are doing, but to the effortlessness of the organisation and the use which is made of generalised conversational devices to achieve it. Two things have to be completed before the consultation can be got underway. The permission for recording has to be sought and the reason for the visit has to be provided. In the transaction of these preliminaries on this occasion, doctor and mother collaboratively get the consultation underway. Thus far, we have talked in broad outline about this transcript and how it displays a flow of work activities. Tasks are just arrived at. They merge into one another. We have also said that this flow is a managed production. What we have to do now is show in what, in actual detail, this production



consists. T o repeat, there are a number of medically relevant things to be d o n e in the consultation, and that any one of them could be done first. There is n o given ordering. Again, we have seen that, this time, there are things which ought to go before medically relevant matters. A set of administrative and professional requirements have to be satisfied before the consultation can b e got underway. Further, responsibility for getting things underway lies with the doctor. H e has to decide the order in which medical matters will be t a k e n . But whatever he decides to do, he will still have to get from the preliminaries he has to go through to the first medical task whatever that turns out to be. In the event, the immediate mechanism which is offered as a candidate solution to the 'getting to' problem is the proposal made at 1.17 and 2.1. 1.17 D:

I s e e - O K ((1.00)) r i g h t - w e l l let's start we'll do the e r have a look at him first

1.18 M:


2. 1 D:

A n d then we'll think about the immunisations

H e r e a putative ordering of activities to be done is offered by the doctor. H e utilises the first part of a standard two part structure, namely proposal/acceptance-rejection. The achievement of agreement on what to d o first and what to do next acts as a bridge from necessary preliminaries to the first properly medical tasks to be undertaken. In making the proposal the doctor guides things forward from preliminaries to the consultation proper. In the transcript, the proposal at 1.17 and 2.1 does not elicit an overt a g r e e m e n t . T h e mother's silence at 2.1 can, however, be heard as indicating a g r e e m e n t . Studies have shown that in many sequences there appears to be a p r e f e r e n c e for agreement (see Pomerantz, 1980: 57-102 and Sacks, this volume, Chapters 2 & 8). The failure of the mother to offer a rejection of the proposal can be taken as an agreement to it. What we think of as the consultation relevant contingencies of the mother seeking to reschedule the things to be done, would be quite profound for the relationship between herself and the doctor. She would have taken over a responsibility which, as we have just seen, is properly his. To do so we would have to raise for examination sets of considerations which showed that at this particular point h e r concerns outweighed his and, in addition, were not being attended to. She would have to take over the directional role from the doctor. We are not saying she cannot do this, merely that on this occasion the opportunity to do so is let go. In their collaborative scheduling, the mother and doctor produce the 'everything's-gone-routinely-so-far' character of the consultation. From the proposal/agreement pair, the doctor has indicated what he takes to be



the next relevant thing to do, namely examining the baby. Beginning the examination of the baby gets the consultation underway. By agreeing that this is indeed the next thing to be done, the mother and doctor collaboratively produce the sequentiality of these activities and its recognisable ordinary orderliness. Now let's step back from this particular fragment and look at the sequences as a whole. With what we have just said in mind, we can step through the data picking out those points at which the doctor uses similar structural forms to make what can be thought of as directional moves, thereby giving the activities a phased organisation. Each consists of a disjunct m a r k e r and the first part of a two or three part structure. Here they are: 1. 1.1 D:

It's just e r - e r by way o f - u m - consent for this - t o be done . . . OK? ((1.00)) just ( ) ((laughs)) I'll hold it and you can write ((7.00)) right thank you very much ((4.00)) ( ) M r s - * * * is i t . . . n o t M r s - . . .that that's her - that's the baby's name 2. 1.5 D: Oh I s e e - I got it the wrong way round ((3.00)) it's usually the case we always seem to do that ((short laugh)) ((5.00)) right - now then ((1.00)) this young lady is now? 3. 1 ..13 D: R i g h t - n o w - u m - y o u - h e he's supposed to be having a six month check today is he 4. 1.17 D: I s e e - O K ((1.00)) right-well let's start we'll do the e r have a look at him first T h e point about the use of standardised structure is, as we have just said, that the appearance of a first part makes the second structurally relevant, i.e. the next relevant thing to do. If the second part is not produced then that ommission is 'accountable'. That is, some repair or other sequential work is necessary. So, in offering first parts, the doctor gives the mother clear-cut, next things to do. The use of the disjunct markers 'right', 'OK' etc mark off activities from one another, the signing of the form from the determination of the name, and that in turn from the reason for the visit. That is to say, these disjunct markers demonstrate for the mother that one activity has been successfully completed and that another is 'upcoming'. Utilising a structure of disjuncture markers and first parts, the doctor picks his way through the necessary preliminaries and gets the consultation underway. H e directs the moves, so to speak, by which the form is signed, the name is determined and the reason for the visit made known. He does so in a way that preserves their serial organisation's effortlessness. The disjunct markers identify where activity completion points are to be found and the



structural format provides a means of moving things along. The natural flow from one thing to another is not a contingent feature of this encounter. It is essential to its oriented-to character and is the product of the collaborative work of both the mother and the doctor. Once the preliminaries have been settled, the same structural device is used to get the consultation's proper business started; that is, he begins the examination of the child. 2.1 D:

And then we'll think about the immunisations ((14.00)) Any problems at all

This has been only a very brief and initial analysis of the work of getting the consultation underway. In it we have tried to show how, in getting the consultation underway, the ordinary orderliness of the serial organisation of work tasks is oriented to and made visible in a work setting. We have also indicated how solutions to the problem of achieving that organisation might be described as following a preference for opportunism in the efficient handling of things to be done. Our intention in examining these matters has been to show how the sensibilities which gave rise to Conversation Analysis can inform the sociological depiction of the organisation of work tasks. Conversational resources and objects are deployed by participants to talk to bring about particular ends as routine elements of daily life. The same ordinary orderliness which characterises their organisation in talk, is on view in the way in which they are used to organise work tasks. We have shown that the orientation which we glossed as 'the description of localising work' and which we said underpins Conversation Analysis, that is, the achievement of the context sensitive-context independent character of social structures, can be carried across to the analysis of work routines. We do not claim that, at present, such analysis is directly comparable to the detailed specification of the orderliness of talk available in Conversation Analysis. But it does show, we would argue, some of the ways that studies of occupational life can learn from Conversation Analysis and in what directions such learning might lead.

Acknowledgements We would like to thank John Lee for his comments on an earlier version of this paper. The research on which part of it is based was supported by a grant from the Social Science Research Council. We would also like to thank Digby Anderson, Diana Bowes and Nick Spencer for their help in obtaining the data.



Notes to Chapter 10 1. Recent collections of work in the field are Atkinson & Heritage, 1984, and Sociological Inquiry, Vol. 50, No. 3/4,1980. 2. A somewhat different approach to a similar order of problem is to be found in Garfinkel, Lynch & Livingston, 1981. 3. This notion has its origin in Garfinkel etal.'s (1981) description of their transcript as 'first time through'. 4. This research is reported in Sharrock, 1982. 5. To see how the approach we outline differs from others, and thus the pervasiveness of Conversation Analysis' distinctiveness, compare our analysis with Byrne & Long, 1986. 6. Roy Turner used this phrase to describe the accounts produced by lawyers in trials. 7. 'Getting to' as a phenomenon is analysed at length in Schenkein, 1971. For an examination of the methodical determination of a reason for a visit in a medical setting, see the unpublished paper of that title by Katz, 1972. 8. The locus classicus for the exposition of category organisation.

11 Interdisciplinary considerations in the analysis of pro-terms D . R. W A T S O N Sociology Department,

University of Manchester

In this Chapter, 1 1 shall outline an elemental grammatical model of the set of English personal pronouns. This is a model which, for all its simplicity, is so pervasive in linguistics that a great many elaborate conceptual edifices have been constructed using it (tacitly or explicitly) as a basis. After giving two brief examples of the use of this model in linguistics, I shall move on to outline some drawbacks of the model as a heuristic device. Two points must be clarified at the outset. Firstly, I shall be arguing against the exclusive use of this model where what is meant by 'exclusive' will, I hope, be progressively elaborated as we move through the empirical analysis upon which the argument is built. After having paid these dues, we can then perhaps afford some observations on interdisciplinary work on this topic. Secondly, the heuristic purposes and criteria to which I shall be directing analytic attention concern whether the model can be used to address issues in the selection and use of pronouns in naturally-situated communication events — events which are typically social-interactional in character. The interactional nature of such communicative events is given in co-participants' collaborative orientation to: (a) the occasioned and situated nature of the communication, where the communication is a non-extractable feature of the occasion in which it is embedded, and (b) the shared conventions in terms of which 'orderly' (i.e. mutually intelligible) talk is generated and traced. The grammatical model can be sketched out as follows. The pronouns are arranged in terms of first, second and third persons singular and plural. The pronoun T constitutes the first person singular (i.e. as a self-reference), and cannot be used to constitute, for instance, the second or third person singular. Similarly, 'You' may not be used to constitute, say, the first or third persons though it can invoke the singular or plural of the second person. 261



Thus, to choose to invoke person and number is to choose between personal pronouns, whereby (with the exception of 'you') the choice of one pronoun precludes the choice of any other pronoun, since no other personal pronoun could invoke that person and number. Choosing between pronouns is, then, choosing between mutually-exclusive linguistic alternatives. Personal pronouns partition one person and number reference from the other person and number references — though as we shall see, other kinds of partitioning (i.e. partitioning with regard to sexual status) is also provided for as a subsidiary feature of the model. When sociological rather than linguistic studies do pay any attention to personal pronouns — a rare ocurrence — they too seem to predicate their analyses on a conception of the scheme of personal pronouns as a set of mutually exclusive alternatives. One such analyst is Norbert Elias who, in a couple of papers containing some undeniably challenging heuristic insights, seeks to establish correspondence between the scheme of personal pronouns, considered as a set of mutually exclusive alternatives, and some system of elementary roles, each role with its own 'attached' perspective. Thus, the perspective from the 'We' standpoint is different from, say, the 'They' perspective, the 'I' a different standpoint from the 'You' or the ' H e / S h e ' perspective, and so on, (of course, these days he's and she's may not always claim, or want, a single perspective). Elias' concern is not with lay society-members' ordinary talk; in his 1966 paper he is concerned to make explicit a series of analytic standpoints from which sociology can be conducted and from which, according to Elias, various major sociological theories have operated, albeit with varying degrees of implicitness. He is also concerned, in his 1969 paper, to use the scheme to analytically formulate the relation between sociology and psychiatry. Broadly speaking, his analysis can be characterised as providing a conceptualisation of the varying types of involvement and detachment which can be engendered in the sociologists' observations of society. Elias regards the scheme of personal pronouns as usable in describing and indeed in furnishing a set of analytic standpoints on society for the sociologist. He regards this scheme as having three major properties, and it will be instructive for us to examine these properties. The first property of the scheme of personal pronouns is that of mutual exclusiveness: e.g. However one uses it, 'I' means 'I' not 'You' . . . just as 'You' means 'You' not 'I' and 'He', 'Me' not 'You' and 'I' and so forth throughout the whole series. (Elias, 1966: 2)



The second property is that of complementarity; the alternative personal pronouns within the scheme may be seen to be complementary, in some sense, e.g. The personal pronouns are a series in the sense that each of them pre-supposes and implies all the others. No 'I' without 'You', without the ' H e ' and 'She'; not 'I' without 'We', 'You' without 'They' and so forth. (Elias, 1966: 2) The third property is the one to which I have alluded above, namely that each of the personal pronouns comprises an elementary role furnishing its own, distinctive perspective on the social world. The use of grammatical properties of the language to conceptualise social persepectives is, of course, a not uncommon feature of sociological analysis; G. H. Mead's distinction between the 'I' and the 'Me' in order to formulate the distinction between the role-taking 'self as an acting subject and as an object conceived from the standpoint of others respectively, is perhaps the best known case in point. These three properties (particularly the first two), are typically assumed in the apparently more formal and elegant linguistic models of the type to which we may now turn. Certain sociological studies would, in addition, seek to establish correspondences between the series of personal pronouns considered as mutually-exclusive alternatives and some system of elementary roles, with their 'attached' perspectives, in a communicative (social) order. There are many examples in linguistics of studies premised upon the above apparently simple, obvious model. Kenneth L. Pike (1973) uses that simple model in order to construct his diagrammatic and other representations. Thus, in one figure, subject A refers to himself in the first person, B is referred to in the second person and C is referred to by means of a third-person pronoun. Thus, an utterance such as the following can be constructed: 'I (A) am speaking to you (B) on her (C's) b e h a l f . Another example can be found in an article by Ivan Lowe (1969, especially pages 416-18), who abstracts and simplifies certain properties of English pronominal system for algebraic representation. These properties are again founded upon the properties of the simple grammatical model of personal pronouns. At one point in his article, Lowe simply represents the persons as 1, 2 and 3, with a binary operation (represented by + ) for the construction of plurals. Lowe notes that the model cannot handle 'ambiguities' such as the inclusive use of 'We', although the model can handle



exclusive uses of 'We' as in 'we saw you'. He says that ambiguity with, for instance, the pronoun 'We' occurs when the other pronoun in the same stretch of talk is not the second person pronoun. Thus, 'we said to him' is ambiguous in that the 'We' may be 1 + 1 or 1 + 2. These formal considerations are, of course, all based on the grammatical model outline above, whereby a given personal pronoun partitions one person and number reference from other person and number references. My objections concerning the heuristic value of the grammatical model as a model for examining issues in the naturally-situated selection and use of personal pronouns must be carefully delineated. Clearly, the grammatical model may on occasion be oriented to by co-participants as a template for the selection and use of personal pronouns, as in this example which manifests what Speier (1973) terms a simple 'pronominal transformation'. See, for instance: A. B.

'How are you?' 'I am fine'

— where 'I' is a transform of 'You' and B hears 'you' are referring to himself. Speier makes the important point that even such straightforward seemingly grammatically-based transformations have interactional significance in that they can be used to achieve the articulation of social interaction by providing for perceived 'stability of reference' to co-present (especially) and nonpresent persons which, in turn, is one facet of the attainment of 'referential adequacy'. Thus, pronouns may, says Speier, be said to be the 'cement' of conversational structure and of its interactional management and administration by co-participants. Thus, personal pronouns — and the transformation operations which constitute (in part) the procedures for use of personal pronouns — are part of society-members' jointly-held conventional apparatus for achieving orderly, intelligable communication; in short, personal pronouns and procedures for using personal pronouns constitute an interactional tacitly-held, unexamined conception of 'what everyone knows to be the case'. These assumed typicalities work as (or substitute for) 'data'. We are urged to accept these analysts' claims by invoking that selfsame item of knowledge as well as by virtue of being committed to what is somewhat optimistically taken to be a body of shared professional knowledge, or even (dare it be said?) a shared 'paradigm'. Harvey Sacks and others (Atkinson & Heritage, 1984: 21-27; Mehan, Jr., 1978: 3,11-38; Jefferson & Lee, 1981: 399-422) have pointed out many of the shortcomings of 'data' furnished by unaided and unconstrained intuition. These purportedly 'typical' occurrences and settings frequently comprise purpose-built 'gloss-overs' of an analytically all-too-convenient



and possibly misleading kind. The only constraints on their formulation are prima facie 'common-sense plausibility' and what a professional audience will accept. Sacks proposes that we replace their intuitively apparent 'hypotheticalized-typicalized' models by recorded and transcribed data so that the constraints on analytic claims are grounded only in very close scrutiny of the (record of) naturally-occurring instances themselves. Sacks claims that this approach reveals phenomena which might well be unavailable to intuition and that the matters which others model in an aprioristic way frequently take on a very different shape and texture. Put crudely, Sacks and the others recommend that we base our analytic statements in (what was recorded as) 'what actually happened' rather than in what supposedly happened. We can see that the sociological work of Elias and the linguistic work of Pike and Lowe share a common basis in intuitively apparent conceptions (or preconceptions) concerning the nature of personal pronouns and how they fit together and work. These studies comprise cases of representatives of different disciplines jointly taking a most unfortunate analytic track; there is no example of multi-disciplinary cross-checks or controls to be found here, just a reproduction of the same problematic approach. O u r approach will be to examine audio-recorded and transcribed data to see whether the simple grammatical model works when it comes down to cases of pronoun selection and use. To provide an initial orientation to the findings of the analysis, the following statements were found to hold. Any analytical assumption of the exclusive omnirelevance of the grammatical model will inevitably do violence to the large range of member's conventions and methods of personal pronoun selection and use. This is important because both Pike and Lowe's analyses can be taken as variously addressing, premissed upon, or at least generally relevant to issues of pronominal selection and use; certainly, Pike's work in general seems to be informed by the aim of treating language under the rubric of 'human behaviour'. Indeed we can say that any exclusive use by the analyst of a formal version of the simple grammatical model is likely to obscure at least three m a j o r issues in personal pronoun selection and use: (a) the multi-faceted ways in which (a commonsense version of) the grammatical model is used in naturally-situated talk, and (b) the range and complexity of other interactional methods and conventions of selection and use, and (c) since the commonsense grammatical model is part and parcel of the larger range of conventions for selection and use, the issue arises of the combinatorial use of



conventions within that range, e.g. the combinatorial use of the grammatically-based pronominal transformations with other conventions relevant to pronoun selection and use, such as organisational references2 of pronouns. Although we shall be dealing with these organisational uses at greater length below, we can make an early comment that the selection and use of a given personal pronoun may well be effected with regard to the typical and the particular characteristics of the social/interactional setting or occasion in which the co-participants are presently situated, or are part of, or are referring to, etc. A formal or bureaucratic organisation is, as we shall see, a case in point. Through the examination of such social or interactional conventions in personal pronoun selection and use, I hope to show that, at best, the grammatical model may only be relevant on some occasions and in some respects; any attempt, such as those by Pike or Lowe to treat the grammatical model as exclusively omnirelevant can, therefore, be seen as analytically aprioristic attempts at decontextualised stipulation. In short, they are instances of 'constructive analysis'. Indeed, one cannot help sustaining the suspicion that it is only through decontextualised stipulation that the 'set' of personal pronouns can be seen as a bounded 'set' at all. Indeed, even when the decontextualisation process is effected, we are left with the problem of what organising or theoretical principle underlies this alleged 'set' such that (a) the distinguishing boundaries of the set are established, and (b) the various personal pronouns 'fit together' in the complementary way as alleged by Elias? What is the principle of this complementarity? We have, for instance, the issue of the third person singular pronouns 'he' and 'she', which do not just partition by person and number but also by membership categories from the collection (or device (See Sacks, 1974:325^5)), 'sex'. There is a strong sense in which these pronouns explicitly map onto a set of descriptions in a way which is not the case for the others of the set. Indeed, 'he' and 'she' share these properties with other sets, not their 'own'. Moreover, as we shall see, 'you' works as an address term and in that respect is unique amongst personal pronouns; this is important since, as indicated above all analyses of personal pronouns must refer one way or another to how they work, so this feature must be seen to operate in any principle unifying the personal pronouns into a 'set'. If the collection of personal pronouns does not seem to operate as a 'set' at the level of theory, can we say at least that it comprises an explicit model of members' uses? Indeed, for many linguists (especially in the sphere of pragmatics) and sociologists (especially in the sphere of ethnomethodology and conversation analysis) this would be criterial as an intitial defence of the model — depending of course on the precise way in which the modelling was effected. This is the issue to which we can now turn, by reference to



naturally-occurring instances. In the absence of theoretical guidance, we might turn to A1 Capone for advice: 'why, lady, you know it all comes down to cases in the end'. Instances of 'interactional' issues in the selection and use of pronouns or o t h e r pro-terms can readily be found, and these issues indicate that pron o u n s are not all of a piece. As we have just observed, the personal pronoun 'you' may (not 'must') be used interactionally as an address term in, for instance, the process of selecting a next speaker in a conversation, as when the present speaker turns to some other party to the conversation and says 'You do agree, don't you?'. However, the personal pronoun 'he' or 'she' is not routinely usable in this way; rather, 'she'/'he' are what we can call 'designators' . My examination of the data corpus has yielded many examples, such as 'do you keep having depressions?', or 'She asked if I would talk to you', and so on. It is also notable that, as Sacks stresses, personal pronouns are not, as such, substitutes for proper names for persons. The work done by the conventional use of the term 'you' in addressing persons is in principle as well as in practice an 'independent' achievement in the sense that the speaker may not even know the name of the person being addressed. And, in turn, n a m e s (particularly surnames) can work to constitute 'same-person' in a way that pronominals cannot. 3 T h e r e are other, more intricate, interactional relevances in the selection and use of personal pronouns (not to mention other pronominals and pro-terms), and analytic references to some of these interactional relevances m a k e it virtually impossible to sustain a strict or simple conception of the relation of 'mutual exclusiveness' between the various personal pronouns, as proposed in the simple grammatical model. Examples for analysis can be culled f r o m our transcribed communications to the crisis intervention centre, and such analysis can cast light on these as on other conversations. O n e such way is to use the term 'we' as what Sacks describes as a 'warm' way of saying 'everyone', or 'anyone' and this, by implication, can in some respects do the work of 'you'. For instance, a psychiatrist may tell a client complaining about her husband's insensitivity to her requests: 'Well, you k n o w , sometimes you have to tell people straight out. We can be very blind to t h e things around us.' This is a 'warm' way because the psychiatrist's 'counting himself in' is readable as partly gratuitous — after all, he is not primarily making a point about himself. We have a more complex example in the following data excerpt: 255:


'I am not needy, I am not one of them type that come screaming.'


268 256:

Co. Well, no, I, we, you don't have to go screaming . . . if they just know, I, I, mean people do care but they've got to know before they can care, haven't they?'

I take it that one way of redressing the 'we' on line 256 is to read it as being a 'warm' way of saying 'anyone' or 'everyone' and — especially in view of the fact that the client has talked about herself not needing help, or not being prepared to 'scream' for it — by implication can do some of the work that use of the term 'you' can do. Indeed, the counsellor's self-repair from 'I' to 'we' might be read as making his 'anyone' statement 'warmer', and as intensifying the warmth of the utterance as a whole. Certainly, 'we' is followed by a 'people (in general)' reference in the same utterance, such that 'anyone' or 'everyone', as indexed in a 'warm' way by the pronoun 'we', might stand in the same relation to 'people in general' so far as responsibilities and rights to obtaining 'care' or 'help' for oneself are concerned. Again, then, in the above sequence from extract D, we see that an operation on the ongoing conversational sequence is required to redress the 'we' as standing for 'everyone' and, by implication, 'you' without providing for the isolation of the client in such a way that, for instance, she might be blamed 'as an individual' for needing help and/or for not actively seeking help. The 'we' can work to establish that everyone (not just the 'you' mentioned immediately after the 'we') needs help sometimes, and that everyone has to seek it sometime — thus, the client need not feel herself to be 'different', 'inferior', a 'failure', etc., in this respect when contrasted with other people. Of course, the 'lack of isolation' effect can be analysed, at least in part by reference to the simple grammatical model; indeed, this example manifests the 'inter-relation' between the grammatical model and other interactional conventions. O n e may further redress the 'we' in this sequence by treating the counsellor's whole utterance beginning on line 256 as being recipientdesigned, as being tailored to address the caller who in the foregoing utterance has been speaking of her need for help (or lack of it), and unwillingness to actively seek help. Such an operation will also help us to redress the 'you' as standing for 'everyone', though another operation allows us to see the counsellor's use of the term 'you' as being transformative of (or also transformative of) the caller's 'I'. Indeed, we might even read the counsellor's utterance as gaining additional force from the ambiguity of 'you' as 'everyone' and 'you' as an address term, for ambiguity can do interactional work too. It must be noted that the ambiguity of the personal pronoun 'you' is, as Sacks often notes (e.g. lecture 11, Fall 1966), an in principle ambiguity —



and, we might add, that it possesses this 'in principle' ambiguity for lay persons and analysts alike. The 'in principle' character of this ambiguity points to the fact that interlocutors know, as a matter of commonsense, that 'you' can potentially be used to indicate 'anyone' or used as an address term. This is not to say that interlocutors invariably treat the term 'you' as ambiguous on each and every occasion of its use; consequently, any analyst would be most unwise to assume such ambiguity as an invariant matter. Very many actual uses of the term 'you' are treated quite straightforwardly and unproblematically as either an address term or as standing for 'anyone', and are treated as such by interlocutors' consultation of the particulars of the context (particularly the intra- and inter-utterance placement) of the pronoun. If there is, for interlocutors, an ambiguity in the use of the pronoun 'you' it does not inhere in any abstractly-formulated linguistic properties of their pronoun considered in vacuo, but in the specifics of its placement. What interlocutors treat as an ambiguous use of 'you' is, then, an occasioned matter, not one which can be decided in advance of particular instances of use. This data excerpt is, perhaps, a case in point. The general case is elegantly argued (though not with regard to pronominal use) by E. A. Schegloff (in Atkinson & Heritage, 1984). At one point in conversation L (not included in extracts D and E of this conversation), the caller forcefully selects a reading of the counsellor's use of the term 'you' as 'everyone' rather than herself in particular: 436: 437: 438: 439: 440:

Co. 'Well everybody needs help sometime Mrs. R., in different degrees'. CI. 'I know they do I know that'. Co. 'You know you have to ask for it sometimes'. CI. 'I know t h a t . . . have you, yes well I am not going on my knees to ask for 'elp no more'.

H e r e , the counsellor's 'you' in line 438 can be 'redressed' by an operation tying the 'you' to 'everyone' in line 436 (and Mrs R's 'they' in line 437 attests to her orientation to an 'everyone' reading); or, alternatively, the caller's 'you' may be redressed by reading it as an address term tying to the caller's name, Mrs R., mentioned by the counsellor on line 436. Again, one might see that there is room for ambiguity in the reading of 'you'. 4 However, the caller legislates between the two readings by using the personal pronoun 'I' to separate herself from any simple 'everyone' assertion. Just as the personal pronoun 'we' can work to prevent the isolation of a member (from, for instance, a category-identified collectivity), so some uses of the personal pronoun 'I' can, so to speak, include a person out.



However, the caller does not always separate herself out from 'everyo n e ' . N o t e that the 'anyone' use of 'you' ('you' as indicating 'one') following what might be read as a 'confession' that she is 'too proud' (to ask for help). 408: 409: 410:


' . . . I wouldn't do it, I am too proud, I know you can be too proud'. Co. ' Y o u can b e too proud and there are sometimes . . .'

N o t e that the counsellor preserves the caller's 'everyone' use of 'you' by reduplicating the caller's phrase 'You can be too proud'; such reduplicaton can d e m o n s t r a t e to the caller that the counsellor is doing the same 'operation' on the 'you' as is the caller. One piece of interactional work that can be achieved by such sequentialised activities is to 'take the sting' out of such acts as confessions or disclosures. In the above sequence, for instance, the ' e v e r y o n e ' reading of 'you' (providing as it does for the prevention of 'isolation' of persons) may be read as presenting the caller's failing as being 'typical', 'normal', or 'common failings', 'nothing special', and so on. The caller's failings, then, can through such work be seen as being nothing more t h a n the collectivity's failings. In Sacks' terms, the utterance has a built-in d e f e n c e , so the use of 'you' can stand for 'I', or 'me' also — something that s o m e o n e trying to work a language via the use of a simple grammar might have some difficulty in handling. M a n y other strictly interactional relevances in the use of personal p r o n o u n s can be cited, and instances can be found in the transcribed calls u n d e r scrutiny. As Sacks has shown, the personal pronouns 'we' and 'they' can have an organisational reference, where their organisational status may w o r k independently of whether the number reference (singular or plural) h a p p e n e d in fact to be present for the persons being designated by the terms ' W e ' or 'They' (see Sacks' Lecture, May 10,1971:2). 'We' and 'They' can on occasion be used, for all practical purposes, in the making of singular refere n c e ; they d o not invariably or only have a plural use. For instance, one may r e f e r to a single agent of some organisation as 'they' rather than 'he'. S u p p o s e a person telephoned the crisis intervention centre, spoke to a single agent of that centre and received no help, he could later report to his friend: 'I p h o n e d the centre for help', or even 'I spoke with a member of the crisis intervention centre but they couldn't help'. Indeed with regard to the first instance quoted above, it would probably sound strange if the person said: 'I talked to the crisis intervention centre but he couldn't help', or, perhaps, 'I talked to the crisis intervention centre and he said they couldn't help', even if t h e person only spoke to one agent of the centre. Similarly, an agent of the c e n t r e can say 'sorry, we can't help you' when conducting the 'proper work' o r 'business' of the centre.



An organisational use of the pronoun 'we' can be found in another excerpt from our data. The caller states: 18:




'I'm just simply looking for a girl, (I'm afraid)' and the counsellor responds: 'Oh we can't do that for you, I'm afraid'. 5

'We', or 'They', when used by a single individual (or to refer, for instance, to transactions with a single individual may warrantably be used, then, when referring to oneself as a member of a unit or a duplicativelyorganised device, particularly as a member of an organisation (in a fairly catholic sense of the term) — membership being constituted by and through such activities as 'speaking in one's organisational capacity', e.g. as an incumbent of a unit of the duplicatively-organised device, the 'crisis intervention centre'. 'We' and 'They' as organisational references stand in systematic alternation to each other, and in doing this they can constitute an 'insider-outsider' relationship vis-à-vis the organisation. For instance, in reporting a conversation with the crisis intervention centre, in which a caller was told 'We can't help you', she can report to a friend 'They couldn't help me'. Personal pronouns may be artfully used to an incumbent of a category such as 'counsellor' in order to disavow any assertion, or potential assertion, that s/he has made a given decision in a 'personal capacity', but has made the decision 'purely' in his/her capacity as one of the centre's counsellors (i.e. as an incumbent of a category of this unit of a duplicatively-organised or 'team-like' device). In this sense, the counsellor may count on the caller's understanding of 'organisational constraints' in order to provide a built-in and available excuse for, e.g. denying some request from the caller. For instance, this sequence extracted from transcribed conversation B shows such a use of pronominals. 19:


20: 21:


22: 23:

CI. Co.

'now a: a blue Ford came for her, and I think she 'phoned you before she came, would you have any record of it, you know?' 'er, well, I'm sorry but we just can't divulge any information at all.' 'You can't do anything' 'I'm sorry I can't no'.

H e r e we have two interesting and related phenomena: (a) on line 21, the counsellor turns down a request, whereby part of the 'turning down' procedure counts on what one can read as an organisational



use of the pro-term 'we', and (b) the counsellor can be read, on lines 21 and 23, as being able to count on a conventional first-person present tense presentation of an apology (in accordance with the conventional construction of a performative utterance, as we shall discuss below), to warrantably introduce an 'I' as well as the foregoing 'we' — if, indeed, such a warrant is required. Note also the in-principle ambiguity of 'you' in the singular and 'you' in the plural in 'you can't do anything', and conceivably the counsellor in line 23 could either answer 'we can't, no', or 'I can't, no' — and each response could be read off as 'unsurprising'. However, the apology, cast in terms of an 'I' not only provides even more strongly for the second 'I' in '. . . I can't no', but can also work to provide for such potential imputations as (and extenuations, as are frequently found around apologies) 'It's not my personal choice, I'm only doing my job' or 'it's more than my job's worth to divulge such information' and so on, should an accusation, complaint or whatever ensue from the turning down of a request; after all, the turning down of a request may make relevant just such a complaint from the caller. The counsellor's prefatory apology (that is, prefatory to a denial of the caller's requests) on line 21 and the repeat apology and request-denial on line 23 can work to provide a 'pre-emptive move', i.e. providing for a built-in, prospectively invocable excuse for the request-denial. This built-in excuse or potentially-available excuse, may be attended to by the co-conversationalist — in this case the caller — and may indeed work to pre-empt the introduction of a complaint or accusation, or at least one cast in personal terms by the caller. If those utterances do not serve as a pre-emptive tactic, then the counsellor may be able to explicitly invoke an excuse such as 'I'm only doing my duty', and so on. It has already been indicated that the personal pronoun 'they' may involve an organisational reference whilst, for instance, reporting conversations, negotiations, etc., with a single individual acting in the capacity of an agent of the organisation, as in 'I 'phoned the police about the matter, but they could do nothing'. 'They' can also serve as an 'everyone', 'anyone' or 'someone' kind of reference, as in Sammy Cahn's well-known quip about growing old: 'Never lie still. They will throw dirt on you'. Such 'people-ingeneral' uses of the personal pronoun 'they' involves, and counts upon, a considerable 'vagueness' — a vagueness that may or may not be 'filled in' later, but a vagueness which has at least provisional practical adequacy in terms of, for example, its placement of that point in the conversation. However, the organisational use of 'they' may also be used with a prospective, yet-to-be-specified vagueness — though whether this vagueness can be considered as in some sense 'borrowing' from that embodied in 'people-in-



general' uses of 'they' is debatable and in any case is only decidable with regard to specific instances of use; moreover, as we have shown above, it is a built-in feature of personal pronouns that they are 'open-ended'. A prospectively-vague organisational use of the personal pronoun 'they' can be found on line 19 of the following sequence: 12 13 14 15 16 17 18 19


CI. Co. CI. Co.

Well I don't think that we can do anything for him but if you can just hang on I'll just have a look and see if there's a number that I can tell you where to ring, there is somewhere er:m, can you hold the line a moment? e:yes I'll just see if I can find the number Yeah Where they will be able to help you.

This use of 'they' is not readily redressed as an 'anyone' or 'people-ingeneral' use of the term — after all, if 'anyone' can help, then a search procedure such as the counsellor embarks upon above would be redundant. O n e plausible reading of 'they' is an organisation, group, association, etc., with 'a' telephone number (i.e. a category-based 'they' might not — at least in all cases — be a possibility since, e.g. 'businessmen', 'doctors', etc., considered together do not have 'a' telephone number). The 'somewhere' on line 14 might also be redressed in terms of the 'telephone number' cue, indicating again that an organisational use of 'they' is involved in this case. Again, of course, we have conducted, as analysts and as lay members, an 'operation' on the utterance and the sequence in order to come up with an 'organisational' reading of the pro-term 'they'. Having constructed such a 'they' utterance so as to tacitly invoke an organisational reference in this instance, the counsellor needs, for the present (or for his present) purposes to make no further specification of the organisation in terms of designating its precise nature or function, giving its name or title, and so on. Indeed, in the above conversation, neither the organisation's name nor its number is ever provided by the counsellor, and the project of finding the particular group/organisation is later abandoned. However, the point is that the construction of an utterance implicitly premised upon an organisational use of 'they' — this implication being established by the counsellor's utterances on lines 12-15 and line 17 — is sufficient to provide for the counsellor's embarking on a search procedure where he can count on the caller's understanding of, and acceptance of, both the reason for and nature of the procedure. That is to say, the embarking on such a search procedure is, in practical terms, unproblematic and unsurpris-



ing for the co-participants to the conversation. Having set up 'they' as an organisation, the counsellor can now take the time to try to find an organisation to 'attach to' the pronoun, an organisation that purportedly could help t h e caller. T h e linking of the activity-descriptor 'help' to the organisational ' t h e y ' also establishes another dimension of the orientation which informs the search procedure, 6 i.e. what kind of organisation to look for. In a n o t h e r p a p e r (see W a t s o n , 1986), I outlined two tacit procedural a s s u m p t i o n s which provided for the sustaining of the telephone call to the suicide prevention centre, in spite of the fact that a 'solution' to the p r o b l e m has not yet been reached, nor will it ever necessarily be reached. T h e s e two background assumptions are: (a)'provisional relevance', whereby the counsellor in practice will listen to any problem-disclosure as if it were relevant to the concerns of the centre (even though, e.g. a caller looking for a prostitute may not be possessed of such concerns) and (b) 'prospective circumstantiality', which involves an 'it-all-depends' property, i.e. that for t h e time being, the counsellor can make no promises of help. These two assumptions address the temporal organisation of the conversation; they w o r k to sustain the conversation until, e.g. 'something (relevant to the c e n t r e ) turns up' or until the caller decides to close off the conversation, or whatever. . T h e organisational use of the personal pronoun 'they' in the utterance and sequence under consideration may be read as helping to provide for and highlight the maintaining of the two background assumptions outlined a b o v e , and perhaps particularly (though by no means exclusively), that of 'provisional relevance'. However, the background assumption of 'prospective circumstantiality' is also highlighted; after all, the counsellor does not g u a r a n t e e that he will be able to find the organisation's telephone number, so neither can he guarantee that 'help' is necessarily forthcoming. However, t h e linking of this organisational 'they' to a 'yet-to-be-explicated' activitydescriptor 'help', provides a 'sufficient-for-all-practical-purposes' reference to allow for the counsellor to be granted time to search for the organisation's t e l e p h o n e number. Had the counsellor been 'successful' in finding the n u m b e r , he might well also have found the organisation's name or title and he may also perhaps as a consequence of learning the name, have established some reading for the caller of 'what it turned out was meant' in more specific terms, by the term 'help' in the same utterance. T h e sustaining of the background assumptions of provisional relevance and prospective circumstantiality through this instance of the organisational use of 'they' is a 'two-tiered' matter as regards the structural organisation of the conversation. As we have said, the use of 'they' is embedded in an



utterance which invokes these assumptions with specific regard to some local relevance within the conversation, namely, that 'local segment' of the conversation which involves the provision for, and transaction of, the search p r o c e d u r e as localised activity within the conversation. However, the organisational reference of 'they' (and the utterance of which it is a part), also helps sustain and highlight the use of these background assumptions to inform 'what is going on in the conversation perse'. An example is the way in which this utterance and sequence may be seen to be informed by the particular use of what we have referred to under the notion of 'prospective circumstantiality', mentioned above; whether the conversation per se may b e seen to have been 'of help' depends in part on whether the counsellor can find (or put the caller in contact with) the organisation which is presently only vaguely designated. It is notable a pro-verb such as 'do' is also potentially very useful in sustaining or reconstituting the 'prospectively-oriented vagueness' eng e n d e r e d in the two procedural assumptions outlined above. In their use of such pro-terms such as 'they' and 'do', members are aware that such pronominals may provide for the relevance of 'filling-in' activities of a retrospective-prospective kind; that is to say, the use of these pro-terms occasions 'consultative work', 'inspections' or 'operations' of the kind we have indicated above. T h e use of pro-terms, then, as part and parcel of co-conversationalists' glossing particles constitutes one dimension of the way in which retrospective and (in the above cases predominantly though not solely) temporallysensitive assumptions such as provisional relevance and prospective circumstantiality can be established in respect of particular issues, such as the achieving of an instance of a collaborative reading of ('proper') help. Of course, the very abstractness and vagueness of a pro-verb such as 'do' can also work to cut off any hope of, e.g. 'direct' help (howsoever it is conceived) f r o m a counsellor in the ensuing talk. See, for example, line 12 of transcribed conversation J. 12:


'Well, I don't think we can do anything for him, b u t . . .' (and the counsellor goes on to recommend other sources of help).

H o w e v e r , it must of course be added that the qualification introduced by ' b u t ' does sustain some prospect of some kind of 'indirect' help, e.g. referral to a n o t h e r organisation, which later in the conversation turns out to be a ' s o m e w h e r e ' with a 'telephone number' as we saw above. W e have already indicated that the issue of which personal pronoun is selected may be crucial for the producing and assessing of what we might



guardedly (and with due respect to Roy Turner, 1974: 197-215) term a ' p e r f o r m a t i v e utterance'. 7 'I' plus the present tense of a verb such as ' p r o m i s e ' constitutes the doing of an action, namely 'promising', whereas t h e personal pronoun 'he' with the verb 'promised' is usually readable as a report of an act of promising. Linking the verb 'promise' with different personal pronouns involves 'splits' or distinctions in performative relevance. V e r b tense, of course, is also crucial. 'I promised' is a 'self-quote', a reference to an already-effected act, as in 'I promised you that I'd do it, didn't I? Such self-quotations may work to indicate that the commitments are still operative — that is, such a self-quotation may work as a comment on an existing state of affairs. Moreover, 'self-quotations' indicate that performative utterances have consequences for the 'doer' and the 'recipient'. This is also the case when a pro-verb such as 'say' is used performatively. As Sacks points out, the utterance 'I still say . . .' indicates a consequentiality that can, for example, be expressed in notions such as 'I still say x, and you can call m e on this', i.e. I am willing and bound to accept the consequences of what I said. Such uses are crucial interactional resources; these uses constit u t e (inter alia) part and parcel of the moral dimensions of social interaction — a dimension which is central to members' ongoing procedures for monitoring the coherence of the interaction.

Pronouns, Possession and Problems Let us now move on to make some brief comments upon the uses of possessive pronouns ('my', 'your', etc.), which themselves have been t r e a t e d as a 'set'. W e may start from the utterly unsurprising observation that possessive pronouns can constitute relations between 'parties' and 'possessables', and as such may be of crucial relevance in the categorising of persons as 'problem persons' as opposed to what Goffman terms 'normals'. 8 T h e r e exist a vast number of various phenomena which might be listed as treatable as 'possessables'. However, such 'possessables' are, commonsensically speaking, not 'all of a piece', and the claimed 'possession' of such p h e n o m e n a may involve a wide diversity of interactional work. For instance, there exist a variety of relationships between the usability of p r o n o u n s possessively in an 'ownership' sense, and their usability in an affiliative sense ('my country', 'your family', and so on). However, one might see that such a use counts upon a different notion of the 'possessability' of a 'country' or 'family' from that implied in the 'possessability' of, for instance, a 'problem', a 'set of personal troubles', or, yet again, for a 'car' — and so on. 9



An example of the affiliative use of a possessive pronoun-plus-category can be found in the following data excerpt; lines 4 2 2 ^ , where the caller says: . . . no, charity begins in the church (doesn't it), (that) my church, what they've done for me I'm very thankful for, nothing at all. It might be ventured that such affiliative use of possessive pronouns is an especially powerful technique when the 'category' mentioned in combination with the possessive pronoun is one that can also be taken as a unit of a duplicatively organised (i.e. team-like) device, such as a 'family', or also as a duplicatively organised device in itself ('country'), rather than a 'simple' category-reference such as 'bus driver' — though this, of course, depends precisely on the practical relevances to which the particular utterance is addressed. In the lines quoted above, it may be seen that the caller's complaint against the church gains its purchase from the affiliative use of a possessive pronoun in combination with a category that can also be taken as a unit of a duplicatively-organised device, and, additionally, a duplicativelyorganised device in itself. It must be said, however, that neither lay nor professional analysis of the affiliative uses of possessive pronouns in relation to (for instance) church units provide for which possessive pronoun should be selected. Indeed, what might be termed the 'selection problem' again involves a whole variety of interactional relevances, and in particular, may be contingent on a membership analysis and perhaps a circumstance analysis of the persons co-present with the speaker. For example, whether one says 'my church', 'our church', 'your church', or whatever, may be contingent on whether there are other members of the congregation present, or other members of the faith in general, or on the status of fellow church-members within the church; for instance, one may even say 'your church' to the vicar or priest of one's own congregation. Indeed this affiliative use of pronouns might also work to formulate place, 10 such that in the course of a search for help procedure a particular person seeking help might through that apparently simple place reference made by the counsellor, literally know 'where to go' (without the counsellor even knowing where, in geographical terms, that is). Obversely, 'personal troubles' or 'problems' may be seen as being 'owned', with possessive pronouns constituting that 'ownership'. This reading of 'ownership', however, does not involve using the possessive pronoun affiliatively. Persons 'owning' problems might, for example, be seen as having certain rights and obligations concerning those problems — rights such as, for instance, that of deciding whether to disclose the problem or not, also rights concerning knowledge and definition of the problem, and obliga-



tions such as the obligation not to 'saddle' certain persons with the problem. The nature and relevance of these rights and obligations is, again, contingent on the nature of the problem ('depression' or 'insanity'), who is approached for help (e.g. 'experts' may be accorded certain rights over 'someone else's' problem, providing that the problem is 'see-able' as being of a particular kind). In short, with a notion such as 'personal problems', we have a phenomena which can be treated as 'possessable' rather in the way that W. W. Sharrock (1974) states that knowledge may be treated as 'owned' — after all, when talking of personal problems we are, in a variety of ways, talking of knowledge of personal problems. An instance of this use of the concept 'problem' as a 'possessable' (though not using a possessive pronoun per se) can be found in transcribed conversation A. 11: 12:

Co. CI.

'Why. have you a problem?' 'I'm afraid so yes . . . .'

After a construction in terms of 'having' a problem, it would be quite routine to refer to 'your problem', and an example of the notion 'problem' as a 'possessable', this time involving two possessive pronouns, is to be found in conversation I, lines 358-59, a telephone conversation between the counsellor and a teenage girl. 358: 359: 360: 361: 362:

Co. 'yes but, er, you know if money's not all that plentiful and your father's got his problems and' CI. '(yeah) that's really what's upset me' Co. 'yes, and as you say you can't be asking for the money and new clothes when you know there isn't the money available'.

Here, the possessive pronoun 'his' is used quite straightforwardly in constituting the possession (and therefore the 'possessability') of problems. However, the first possessive pronoun in this sequence is the 'your' in '. . . your father's got his problems . . .'. The 'your' in this statement provides for combinatorial reference, i.e. a reference to two persons or parties, or, more accurately, incumbents of relationally-paired categories (See Sacks, Lecture 9, November 2, 1967). As manifested in this instance, the possessive pronoun 'your' preserves the address function of the personal pronoun 'you', addressing in this case the girl who is calling, and also provides for the appending of a term such as 'father'. One systematic way of referring to two persons, then, is to use a possessive pronoun plus another term (probably a term invoking a category) such as 'father'. In this case, we



see that as a result, the relational pair invoked is that of 'daughter-father', and the 'making available' of this duplicatively-organised relational pair serves as a locus for the distribution, or imputation, of rights and obligations — or for negotiations and formulations regarding such distributions. Lines 361-62 of conversation I invoke the establishing of just such a set of rights and (particularly manifestly) obligations — i.e. what is permissible for a daughter to claim from her father given the fact that her father 'possesses' a problem or problems, in this case 'money problems' (which in turn serves as a focus for how, specifically, the daughter's obligations are to be dispensed). Consequently, we can see the relation of the daughter (and her problem) to the problem 'possessed' by her father, and can make cause-and-effect imputations, etc. (see below). The affiliative use of a possessive pronoun, as in 'your church' (which is itself conceivable as a form of combinatorial reference), may also provide for the introduction of combinational references invoking relational pairs. Thus, if, as in extract D of conversation L the caller complains against her church, the counsellor can use the possessive pronoun 'your' (involving, as it does, an address function) to invoke a relevant category such as 'vicar': 253: 254:

Co. do you - have you spoken to your vicar or anyone at church? do - do they know that/you're infirm?

This might be seen as invoking the relational pair category 'member of the congregation' — 'vicar', and, indeed, the invoking of such relational pair categories is frequently done through the use of possessive pronouns, whereby the possessive is 'repaired' by the invoking of a pair — position category. Thus, 'your' in the above instance can be seen to be repairable as invoking the category 'member of the congregation' (of a unit of the device 'church') after the category 'vicar' has been introduced. This gives us a good example of the 'prospective repair' of a pro-term, whereby the redressing operation is addressed to what transpires later in the conversation — even if the 'later' is the very next word. Combinatorial references such as 'your father', particularly when closely accompanied by references such as . . got his problem', can do highly delicate and extremely compressed interactional work concerning the discussion of a problem. Some of that work has been indicated above, in terms of the locating of rights and obligations within a 'relational pair' locus. It may also be said, however, that the uses of personal and (perhaps particularly) possessive pronouns may be said to constitute the multi-faceted relationship of a 'possessable' such as 'a problem' to a category such as 'father'. Unlike a category such as 'kleptomanic', a category such as 'father'



does not, at least in any routine or immediately observable way, involve 'category-bound' or 'built-in' personal problems. Nor does a category such as 'daughter'; and though the relational pair 'daughter-father' may imply the possible relevance of a given realm of problems, the pairing is not readable as predictive in specific terms of any particular problem such as 'money troubles'. Rather, when the problem 'money troubles' is raised, the fatherdaughter category-pairing may be used as, for instance, a warrant for introducing the problem, for talking about it, for 'being concerned', for showing how 'someone else's' problem affects oneself, and for providing ways of talking about the money problem (as, e.g. a 'family problem', 'parental problem', etc.). T h e use of possessive pronouns such as 'your', as in '. . . your father's got his problems . . . ' i n conversation I, is, then, intricately bound up with the way that the daughter can help to demonstrate her precise 'relationship' to her father's problem, i.e. that she is reporting her father's problem (and has, via the 'daughter'-'father' relationship pairing, a right to report it, or have an interest in it), but that her father's problem also gives her a problem. T h e use of possessive pronouns may also help to provide for an implicit temporal formulation of the following kind: 'My father has got problems and so I've got problems', so that the father's problems may be seen to have 'come first' with the daughter's problems eventuating from them as derived matters. Such temporal constructions of 'generic' and 'derived' can be used for such interactional work as blame-allocation ('it's basically not my fault/ my responsibility'), for indications as to how the daughter's problems might be eradicated (e.g. by solving her father's problems), etc. or simply the work of accounting for why the caller's problems are what they are. For instance, such temporal formulations allow of 'cause-and-effect' imputations. Using possessive pronouns, then, may assist in constituting the caller's 'relationship' to the 'possession' of a problem, where the possible range and variety of such relationships is quite considerable. Moreover, the selection of a given possessive pronoun, is, as we have said, subject to membership and circumstance analysis. Nor is this the end of our considerations. The commonsensically-known fact that problems comprise possessables is part and parcel of a generic matter with regard to the 'constituents' of problems and their disclosure. Particularly problems which involve or entail personal or mental (especially emotional) states are problems which any competent society member is expected to know best on his or her own behalf. 'Knowing best', here, involves such matters as certain conventional rights over the avowal and specification of the state/problem and over whether or not to disclose the state/problem, to whom to disclose it, and to whom to disclose it first.



Jeff Coulter uses the scheme of personal pronouns to illuminate the nature of these conventions." To declare depression in the first person 'I'm so depressed today' is normally to have that avowal accorded some form of privilege, e.g. to be 'taken at one's word', whereas second and third-person ascriptions of depression normally involve the 'building in' of some form of guardedness, as in 'you seem depressed today', 'He looked depressed today' (unless, in the latter case, the speaker was reporting a third party's prior self-avowal that s/he was depressed). These 'guardedness' particles display interlocutor's orientations to the 'known fact' that persons normally have communicative rights to 'know better' on their own behalf, and to disclose or withhold a problem or problematic personal state. The notion of 'possessing' problems is deeply embedded in this broad range of conventions regarding the avowal and ascription of mental predicates, personal states, some kinds of 'self-knowledge' and the like (Coulter, 1975). Whilst working the elementary 'scheme' of personal pronouns gives us one analytic angle on such matters, it must be said that the 'scheme' itself does not generically 'organise' these matters; rather, the pronouns operate within, and they are used in the service of this broad and complex mosaic of communicative conventions in our culture concerning these matters. As we have already noted, one management strategy in the domains of counselling and psychotherapy — domains characteristically involving 'personal/ emotional states' and the like — may use 'we', rather than 'you' or 'anyone', as a warm or supportive way of talking to persons about their problems. Again the selection and use of personal problems does not work simply — i.e. without special pleading — according to the elementary scheme of mutually-exclusive alternatives, since a whole array of 'complicating' conventions constantly and awkwardly arise. ('Complicating' and 'awkward' for analysts committed to the elementary scheme not, of course, for members or interlocutors.)

Partitioning We may now move on to a brief consideration of what I shall call the 'partitioning' uses of personal (and, for that matter, possessive) pronouns. Interactional selections and uses of personal pronouns have what Speier (1969) calls 'strong effects on the constitutive properties of group membership'. In this regard, we may see that personal pronouns possess two properties: the 'property of inclusion' and the 'property of exclusion'. A pronoun such as 'we' is, initially and 'in itself' (so to speak), an 'includer' of others through which the issue of which 'others' is a matter for a 'redressing



operation' — involving, inter alia, members' orientation to what Sacks calls 'the principle of directiveness', concerning who is being addressed. As Speier says, the principle of directiveness addresses par excellence the fact that speech is an interactional process in which a speaker communicates with others who stand in a hearer relationship to him/her. And a pronoun such as 'we' also involves a 'property of exclusion', i.e. the distinguishing of included parties from 'everyone else' (though who is included in the 'we' is itself in principle open to ambiguities). In this respect again we have manifested one of Sacks' observations concerning pronominals, mentioned earlier — namely that 'we' and 'they' may stand in systematic alternation to each other, and that therefore the joint use of these terms may be a strong inclusion/exclusion technique.'They do x but we never do x' may constitute racial, sexual, class, family etc., alignments through tying operations to membership categories, etc. Again, who is excluded has to be redressed by the co-conversationalists, although this, as Speier points out, may occasionally involve what J. L. Austin calls 'misfiring'. Such uses of pronouns are, then, part and parcel of our procedures for social alignment. Indeed, one might even suggest that the use of a pronoun such as 'we' may signal an activity of both exclusion and inclusion, a constituent feature of this activity being an analysis as to who is included and excluded. Often guidelines for such an analysis are explicitly provided by one of the speakers, as in 'They, the Catholics, get help, but we, the Protestants, don't'. As Speier points out, 'we' + category may quite satisfactorily provide for referential adequacy, although frequently the use of a pronoun counts upon an 'understood-as-relevant' (or perhaps previously introduced) category for its 'relevance'. For example, the utterance 'we had a statistics lecture to-day' may be read as tacitly addressed to the category 'student'. Similarly, the directiveness of 'you' must be established; 'you' can, interactionally, stand for 'me', 'someone else', 'they', etc. An instance of such partitioning using involving pronouns may be found in this excerpt from our data, where the caller says: 415:


'no beggars do that, I haven't become one of them yet'

where the term 'them' is used as being exclusive of 'I' (and; for that matter, 'you' or 'we') by the caller. However, it must be reiterated that such partitioning work does not simply or always involve straightforward grammatical divisions between first, second and third-person singular and plural. An observation we can use in order to further exemplify this point is the way that Richard M. Nixon, particularly when addressing such controversial issues as Watergate, some-



times used the term 'your President' — which may be pernominalised as 'he' — rather than saying 'I'. By using 'your (or 'the') President', or the pro-term 'he', Nixon could claim entitlements such as being accorded trust, the 'benefit of the doubt', the belief that his motives were 'lofty' not 'selfinterested', the right to secrecy — all by virtue of invoking the category 'President (of the U . S . A . ) ' , which he might not be able to straightforwardly or credibly claim by using the pronoun 'I'. 12 The pronoun 'I' might be seen as indicating such readings as 'this particular person, presently the incumbent of the category " P r e s i d e n t " ' , or even 'speaking in one's personal rather than presidential capacity'. H o w e v e r , it is to be noted that such 'third-person speech' is tacitly to be taken as referring to President Nixon, the man who is making the speech. M a k i n g third-person speech stand for 'I' may establish a sense of what N. Elias calls a 'distanciation' of the speaker from his category-incumbency, which in turn may be crucial when, for example, claiming a point of principle concerning category-incumbency — even when the speaker is a, or the, k n o w n category-incumbent. Indeed, the fact that third-person speech can be 'designed' to focus on the category rather than on any particular incumbent (when, e.g. the person speaking is, and is known by others to be, an incumbent of that category) constitutes the 'power of ellipsis' of this form of speech — and it is just interactional 'power' which renders untenable any simple analytical reference to partitioning in terms of person and number. This power (i.e. the way third-person speech can be made to stand also for 'I') cannot simply be analysed in terms of such basic grammatical rules simply because such uses of speech. What these uses can achieve, is fundamentally a conventional matter, in the sense of involving complex interactional rules and relevances in regard to the 'reading' of person, number and a vast variety of other phenomena. Whether the grammatical model for some version of it and what features of the model are used is a contextually occasioned matter. It is worthwhile examining Sacks' analyses of third-person speech to show what kinds of interactional work can be done in terms of such speech. Sacks examines Truman's words to the then President Johnson: 'I think I ought to report to the President. He might want me to do something'. Here, 'he' substitutes for 'you', and permits of orienting to Johnson as the President rather than simply as a private person. Partitioning practices, then, may count on person and number as one, but only one, facet of the resources mobilised for such work; and, as the Truman-Johnson example illustrates, such practices are not exempt from the redressing operations outlined a b o v e . T h e interesting feature of this example, and also the treating of s o m e o n e as a 'non-person' (see note 8), and a number of the other examples



mentioned earlier is that the use of pro-terms, and particularly personal pronouns, may assist in selecting identifications for persons that highlight some of the capacities in terms of which those persons are treatable, whilst cutting out other capacities — including, possibly, the capacity of bona-fide participant or fully ratified co-conversationalist. Hopefully, this recapitulation of interactional relevance in the selection and use of personal pronouns has served to highlight that although personal pronouns may be seen as doing 'partitioning work', such partitioning work is a complex and interactionally delicate phenomenon that cannot be analysed or explicated in terms of any approach exclusively based on a simple grammatical analysis of personal pronouns in terms of person and number. From the above considerations, we might add a few observations on the kind of analytic operations which some linguists (mainly) seem to be performing. These operations — broadly involving what I have termed 'decontextualisation' — are not without a certain oddness. These operations are found par excellence in linguistic analyses of personal pronouns such as those I have outlined above. They seem to operate by extracting ad hoc elements of society-members' uses (in this case their uses of personal pronouns) from a broader, more intricate and complex range of uses, formalising and indeed reifying the selected elements into an analytic 'model' whereby they are combined into a 'set' or 'scheme'. They then attempt to return them to the arena(s) of action purportedly similar to that from whence they came, using the scheme to characterise and account for (in a variety of ways involving differing degrees of abstraction and formalism) those arenas. Of course, the complex texture of relevances internal to that setting — features the analytic exercise by its very nature leaves out — frequently compromises the applicability of such an overarching scheme, so the ensuing analysis tends to be shored up with, hedged around by mutatis mutandis riders, invented examples and other conveniences. The advent of computers has meant we can now get a machine to go through this odd operation with us. The apparent authority lent by computational representation of formal schemes and 'cases' of their application should not, however, be allowed to disguise the oddness of the analytic exercise. Instead of formulating a deracinated model and then having to conjure up more or less real-world instances where it might more or less apply, how much more straightforward and fruitful it would have been to account for the selection and use of pronouns in terms of the texture of relevances to which interlocutors orient in particular settings; from this, we may begin to examine members' practices and conventions concerning pronoun selection and use in various types of setting (e.g. counselling or formal-organisational settings) and — cru-



d a i l y — types of sequences. Such a progression should be conducted modestly a n d with d u e caution, particularly in view of the manifestly in situ nature of p r o n o u n selection and use. Having said that depending on the instance the schema of personal p r o n o u n s doesn't hold, or holds in debilitatingly limited ways, or holds some times m o r e than others, how does this leave us with regard to the possibility of cross- or inter-disciplinary work between (in this case) sociology and linguistics? Not in as bad a shape as the comments above might seem to indicate; for cross- and inter-disciplinary considerations can be very fruitful i n d e e d , providing researchers on both 'sides' are prepared to forego some of their treasured assumptions concerning such work. T h e value of interdisciplinary considerations inheres in large part in the ways in which they allow us to sharpen up our intra-disciplinary conceptual distinctions. T h e y also help us identify, bring into visibility and critically assess those tendencies and predispositions in one discipline which are also i n c o r p o r a t e d , in other ways, into others, as the example of Elias' study indicated. O f t e n o n e finds that the assumptions tacitly incorporated into one discipline are m a d e explicit by another discipline, or used in other ways which r e n d e r t h e m m o r e noticeable. In the case of Elias' paper, for instance, a linguist or, say, a Wittgensteinian philosopher might wish to point out that the notion of a single (and unitary) perspective 'attached' to a personal p r o n o u n ' s 'role' smacks in some respects of the 'Sapir-Whorfe' hypothesis or a sociological variant of 'linguistic determinism (or perception)' in its 'strong' form. This kind of approach, then, involves 'bouncing' the work of one discipline against that of another, and this is only possible if analysts are p r e p a r e d to loosen their grip on their assumption of interdisciplinary comp l e m e n t a r i t y . This assumption is often only sustainable on the thinnest of w a r r a n t s , namely an apparent substantive or thematic warrant, such that studies f r o m various disciplines are deemed complementary if they address w h a t seems to be 'the same theme', travelling together under the broad p r o t e c t i o n of 'communication studies', 'interactionst analysis', 'discourse analysis' or similar flags of convenience. This claim can only be allowed to prevail if the various analysts diplomatically suspend the usual ethic of a c a d e m i c criticism or scepticism regarding the differing and on occasion o p p o s i n g conceptual foundations of the disciplines involved. T h e assumption of interdisciplinary complementarity seems often to o p e r a t e amongst analysts in largely phatic ways but valuable though this may b e it detracts attention from the real benefit of interdisciplinary work, that of t h e heightening of practitioners' critical awareness of the differing concep-



tual bases, approaches and options within their own discipline. In the present case, the way in which sociologists of the ethnomethodological and conversation-analytic persuasions address what is apparently 'the same t h e m e ' may cast light not only upon how some linguists proceed but upon basic analytic procedures too. Certainly a reply from linguists might be expected to do this. This assumption is linked with another self-imposed constraint, namely the assumption of ¿«fra-disciplinary consistency and consensus. In fact, most disciplines are, of course, internally rather highly diversified, being composed of several quite distinct conceptual domains. T o be sure, it may be the case that there is, in general terms, a greater analytic affinity between two conceptual domains from different disciplines than between each domain and others within its 'own' respective discipline. However, the viability of cross- or inter-disciplinary work is even more o f a contingent and pragmatic matter than even the occasioned nature of these affinities would suggest. Much depends, for instance, on the nature of the data involved, particularly since what counts as data tends to be 'perspective-specific'. What we often end up with — and it is a not inconsiderable o u t c o m e — is temporary, practical and limited associations between some set of analysts with another set from, perhaps, another discipline in relation to a given field or corpus of data. For example, two spheres where this seems possible are those of conversation analysis in sociology and pragmatics in linguistics — hardly an 'all-or-nothing' alliance. F o r the rest — and this has its undisputable analytic benefits too — it seems reasonable to anticipate the conceptual and methodological elucidation whch often derives from differing analytic standpoints; without wishing to advance the Balkanisation of interdisciplinary work, I hope to have demonstrated in at least a prefatory way that this elucidation may be the o u t c o m e of interdisciplinary disputation as much as consensus. In this sense, I regard this Chapter as an example of interdisciplinary work. At least I hope to have made a first move in the process of horse-trading over particular issues that interdisciplinary work at its best entails, and which helps ensure that such work does not take the form of static agreements but instead constitues a continuing, dynamic and, above all, other than naivelyconducted process.



Acknowledgements I should like to thank the following for valuable assistance with this Chapter: G r a h a m Button, Peter Halfpenny, John Heritage, John Lee and Wes Sharrock. If after all this I haven't got it right and still claim exclusive authorship it has to be my own fault. Some segments of this analysis were circulated in microfiche form: (Pragmatics Microfiche Vol. 1, fiche 2, December 1975). However the analytic argument in this present paper differs significantly both in nature and extension.

Notes to Chapter 11 1. The data I use is largely drawn from transcriptions of audio-recorded two-party telephone calls to suicide prevention/crisis intervention centre. The data was analysed extensively in my published Ph.D. thesis: Calls for Help: A Sociological Analysis of Telephone Calls to a 'Crisis Intervention Centre', Sociology Department, Warwick University, Coventry, England, CV4 7AL, July 1974 (including two chapters, Chapters 7 and 8, on the use of pro-terms in this data corpus). 2. By 'organisational references', here, I preserve Harvey Sacks' term for references made by and about those persons holding office in bureaucracies/formal organisations — references to the bureaucracy itself or for the officers' incumbency within it. Empirical analysis of these uses of pronouns follows. See also H. Sacks, Lecture 1, May 1971, Department of Social Science, University of California at Irvine, p. 2ff. 3. It is notable that a proper name may be about the only item which is constant about people over a period of a lifetime; surnames particularly are 'consistent identifiers' in a way that pronouns cannot be. An interesting example of the problem in using pronominals for such purposes can be found in one of the old G o o n Show scripts. 'Who's that at the door?' 'It's a man who calls himself "me".' 'Ask him to prove it.' Of course, the 'him' in the third utterance does work, through a tying operation, to provide for consistency of reference to the person in question. 4. On the in-principle ambiguity of the personal pronoun 'you', see again H. Sacks, Lecture 2, Fall 1966, pp. 2ff. Sacks notes particularly the use of sequential procedures or 'operations' to redress the term 'you' as being a singular or plural use. Sacks also points to 'you' as the prototypical 'everyone', 'someone', 'people' term. Again, we see that interactional relevances are crucial in the doing of such a 'redress'. The present paper owes a very great deal to Sacks' consideration of such issues; transcriptions of his lectures are obtainable from the School of Social Science at the University of California at Irvine. As an aside, we might add that the ambiguity noted in his model is, similarly, a theoretically-given ambiguity (Lecture, April 26,1971). 5. When 'we' is observably being used to stand for 'I', one rule for doing tying is 'look for an organisational reference or organisational identification' (e.g. the identification in the counsellor's opening utterance: 'Hello, crisis intervention centre, . . .'). In this regard we can see that the use of tying terms involves built-in instructions for establishing the term (and therefore the utterance) to which the tying term ties. Of course, a basic example is: if the pronoun 'he' is

TALK A N D SOCIAL ORGANISATION used, look for an incumbent of the category 'male' from the membership categorisation device 'sex'. It need hardly be said that such instructions do not solve all the contextual problems of the establishing of ties — e.g. there may be several males mentioned in the conversation. This 'sexual categorisation' reference shows that members of the so-called 'set' of personal pronouns are not all of similar status in terms of the interactional work they can be made to do. Moreover, as is well known, the pronoun 'he' has often been used 'generically', to stand, in effect, for 'she', or to tie to some sex-neutral or inclusive noun, e.g. 'the author', or to 'one'. Recent changes in these conventions show a dynamic element which is not easily built into the rather static formalism of the grammatical scheme of personal pronouns. It is also interesting that although (so far as I know) linguists have not been able to adequately handle the interactional conventions in pronoun selection and use, they are of course de facto quite aware of the conventional possibilities in the substitution of plural pronominals for singular pronominals. Consider, for example, the following linguist's observations on how 'sexual partitioning' can be avoided in the making of sexually-neutral social reference to a given individual. This is a letter from The Guardian, April 21,1975, p. 12: A plural answer to the singular pronoun problem: Sir, — Mary Stott (Guardian Miscellany, April 17) is looking for a third person singular sexless pronoun to infiltrate into English and I suggest that she needs to look no further than the pronoun 'they'. 'They' is already used quite extensively with a singular referent in cases where the use of 'he' or 'she' would be too specific as in: 'If anyone wants to do that they are welcome to try'; 'Nobody here likes chocolate, do they?'; 'When you find someone who knows, tell them . . .' (American usage is clearly discriminatory in preferring 'he' in these examples.) Admittedly it would produce an odd effect if there was a specific singular referent: 'The victim wrote a letter in which they complained . . .' But it is more likely to catch on than some concocted pronoun. If your contributor wants to try it out, I wish them the best of luck — Yours sincerely David Lee Department of Linguistics University College of North Wales Bangor. For an example of the use of the personal pronoun 'their' in ordinary talk, see Joan Sims' comments of filming on a mountainside (from the same issue of The Guardian): That was a treat, says someone, ironically, but Joan Sims demurs: 'Not if you suffer from vertigo. I was the only person who had to have their chair facing up the mountain.' As Sacks (Lecture 11) says, the arguments concerning organisational uses of 'we' and 'they' do not, of course, preclude other, category-based, uses of these terms. For instance, a father who is more strict with his daughter than with his son, can be referred to as 'They are always like that', where 'they' indicates all (or typical) incumbents of the category 'father'.



7. See Austin, 1962; and H. Sacks' comments, Lecture 19, Fall 1966, p. 1 for a discussion of these issues with extensions to which I shall refer below. See also Sacks' Lecture 11, November 9,1967. 8. Parenthetically, we have another example of communicative conventions concerning 'normals' and those they are normal against. This example is immortalised in the title of a B.B.C. Radio 4 programme for the disabled, namely 'Does H e Take Sugar?' (where a 'normal' is putatively addressed to speak on behalf of a co-present disabled person). Whilst this example might appear at least to exemplify the partitioning aspects of the simple grammatical scheme of linguistics, there is certainly no way in which this simple grammatical scheme models the conventions for pronoun selection in this instance. On the contrary it would seem far more fruitful to examine the issue of selection in terms of the eminently social organisational conventions and participation rights used in the articulation of social (including, of course, communicative) relations and interactions between 'normals' and those who might not be fully ratified co-participants (children, the mentally handicapped and the senile aged would seem to be other cases in point). We might begin by looking at asymmetries in participation (including speaking) rights as between these categories of person. Sacks has m a d e several observations concerning the social organisation of conversational interaction in adult-child interaction. For a general approach to such phenomena, see Speier, 1969, andMacKay, 1974. 9. O n affiliative uses and relational uses of personal pronouns, and on other issues pertaining to 'possessability', see H. Sacks, Lecture 17, Fall 1966. Speier, 1969, also cites Schneider's point that possessive pronouns help multiply the referential possibilities in respect of the available categorisations for persons — e.g. the categorisation 'mother' can be made to do a great variety of work when the possessive pronoun 'my' (as opposed to 'your', 'his', 'their', etc.) mother is 'attached'. 10. For classic analysis of membership and circumstance, see Schegloff, 1972. 11. This is a highly abbreviated and bowdlerised version of issues which have been insightfully and elegantly analysed by Jeff Coulter (1975) and Sacks (1975). 12. For similar examples, and for more detailed statements of such uses, see Sacks, Lecture 9, November 2, 1967, p. 9 and Speier, 1969: 134-35, 153, note 8. Sacks' lecture involves an examination of the problem of fixing an address pronoun onto a previous identification or noun, and on how we can see some uses of 'he' as substituting for 'you'. By extension, we may see how a person may select the pronoun 'he' instead of T after invoking one of his category-incumbencies.

12 Epilogue: The definition of alternatives: Some sources of confusion in interdisciplinary discussion WES SHARROCK University of Manchester and BOB ANDERSON Manchester Polytechnic

In refuelling ships at sea the actual transfer of fuel from one vessel to another can be a minor and relatively simple part of the whole operation. The tricky part can come in getting the ships aligned and connected, then keeping them there whilst the tranfer is made. This seems a good metaphor for the relations between the various approaches within the human studies and, particularly, for the problems involved in reciprocal criticism amongst them. T h e actual criticisms are, of course, the nub of the matter, but their relevance and effect depend very heavily upon the ways in which the approaches are aligned for comparison, the way the context for criticism is set up. In this epilogue we will try to discount much criticism of Conversation Analysis, that sort which is intended not just to discredit or revise any of the findings of its particular conversational inquiries, but which is designed to show that the whole strategy of investigation is entirely wrong headed or, if not quite that, then fundamentally flawed. 290



Reciprocal criticism within the human studies is very difficult to make adequate and effective, and not simply because the various standpoints involved are rather diffuse and thus insulated against criticism in the way of which Popperians so fiercely disapprove. The main obstacle to serious interchange (at this stage of things at least) is the problem of mutual understanding amongst different points of view, since the arguments are not between rival theories or hypotheses so much as between alternate 'modes of analysis', and the understanding of these really involves getting the hang of how these ways of thinking, investigating and interpreting, work. Getting a grip on one mode of analysis is hard enough, but really forceful criticism often requires that one be able to get inside more than one. Not only is it difficult enough to get a grasp on one way of thinking, it is also often the case that the ability to get inside a second one is inhibited by mastery of the first. True entry into that second standpoint may require the abandonment of the entire apparatus of conceptions making up the first. Criticism of Conversation Analysis is clear testimony to the fact that many people do not see its point. Much of the criticism of Conversation Analysis is of a kind common in the human studies, which perhaps results from the kind of difficulties to which we have been pointing. It is the kind of criticism which does not suggest a better way of doing the things that have been attempted, but which casts doubt on the value of attempting such things at all, calling for a different kind of study. Thus, criticism of Conversation Analysis (hereafter CA) is rarely of the kind which suggests different or superior ways of examining conversation's organisation and more typically of the sort that complains that CA has not solved the problems that the

critic has in mind. W e do not aim to provide a counterblast to such critics, arguing that their own kinds of inquiry leave much to be desired, but propose, instead, to engage in some exposition of CA, though in a manner rather different to that which its practitioners usually do, which is that which has been used in the previous chapters in this collection. CA makes something of a principle of presenting itself through its work, showing the kind of inquiries that it makes and the kind of conclusions that emerge from them, a strategy which has on certain assumptions much to recommend it but which in our experience often does very little to increase the comprehension of the true nature of C A ' s efforts amongst those with doubts. As often as not, doing that just confirms, or even amplifies, their reservations. They can see from CA's studies, certainly, the kinds of materials that CA examines, the kind of things that it says about them and the way in which different studies under the auspices of CA perhaps complement and elaborate each other but they cannot see, from those studies, what underlies and motivates them. They



cannot, most basically, see what the 'problematic' is and, of course, without a recognition of that, many of the moves appear quite arbitrary. We shall be mainly concerned with the kind of criticisms of CA which are made from points of view dominated by 'linguistic' concerns and can note that the fundamental objection which emanates from these sources is that CA is unsystematic. Our basic aim is to show just how wide of the mark that complaint is, and that it is unlikely one can find a more systematic exercise anywhere in the human studies than CA, but it is, we think, likely to be in the forefront of objections because the critics are unable to recognise the character and extent of CA's systematicity. We will, therefore, be concentrating attention on those things which underlie CA inquiries and which hold them together, and our exposition will not be of the particular results of CA investigations nor of their cumulative structure, but of the elementary study policies which motivate and direct these studies. Our hope is that some clarification here will remove some crucial misunderstandings and obviate much irrelevant criticism. It is our conviction that the misunderstandings amongst the approaches to the human studies are more important than the disagreements (at the present time). There are disagreements aplenty, we have no desire to minimise that fact, but parties to controversy often seem to be mistaken about the nature of the differences dividing them. Further, their misunderstandings are often over basic and elementary questions rather than over developed and sophisticated issues, and it is for this reason that we think that attention to some of the most simple and primary considerations may be more useful than attempts to build on the sophisticated and complex arguments and analyses that have gone before. We start from some issues relating to the place of CA in sociology and then turn to discussion of its relations with linguistics, specifically that area designated as 'discourse analysis'.1 The issues pertaining to CA in sociology are germane in a double fashion: there are many misunderstandings about CA in sociology and we might be able to contribute to clearing up some of those. Second, insofar as the argument is directed to people whose main background is in linguistics, then they may find a much greater awareness of what CA is 'up to' if they have some sense of the ways in which it bears the marks of its origins in sociology.

The Basis of Social Order One problem which unites a good many sociologists is that of 'social order'. They seek to understand how activities in society possess such



orderliness as they do. They disagree about the answer that is to be given to this problem. H o w e v e r , they often agree about the form that a solution to it should take. T h e y agree, that is, that what is sought is some general principle which provides for social order. They disagree on what the principle is. T h u s , there are those who think that the general principle which assures order in society is that of harmony or agreement. There are others who could not disagree more about the nature of the general principle, and think that it is ' p o w e r ' or 'control' which should be identified as this. The back and forth b e t w e e n these two broad points of view is long standing and continues. The purpose of much sociological investigation is, then, to show the relevance of the preferred general principle, to show that the principle is general and that it permeates social life, that this or that activity exemplifies the working of harmony or, alternatively, control. It is not our intention to question the legitimacy of that conception of what sociologists do, but only to doubt the supposition that inquiries are definitive, identifying the only possible conception of how sociologists might think of the problem of social order. They are not. T h e r e is at least one other conception. 2 This is one which seeks to understand the practical production of social order, which seeks to understand how activities fit together into stable or changing patterns, with how activities make up patterns of activity and how, through their interrelation, they produce and reproduce the activities they compose. W e appreciate that presenting deep, dense arguments with this degree of condensation does not help to elucidate them but will in the articulation of the discussion do this in a w a y which should make much clearer just what we are saying here. A t this point, suffice it to say that the task of seeing how activities relate under the rubric of some general principle is replaced by that of seeing how the activities 'dovetail' with one another. Thus, where the usual concern with social order is primarily about the relations between groups, organisations, institutions and persons, the interest, in this 'alternative' conception is primarily on the relationship between activities. Rather than thinking of society as a system of groups, institutions, positions etc. it thinks of it as a system of activities (insofar as it is a system at all). Perhaps it begins to be apparent why we have stressed the problem of 'getting the hang' of the various approaches in the human studies. If the kind of outline w e have given of the options is at all accurate, then one can see that the move from the one to the other is a shift in the way in which p h e n o m e n a are to be viewed: one is being invited to look at social organisations as a system of activities rather than a set of interpersonal relations. T h a t , surely, is a significant change which must ramify through all the things one does. H o w it does so is not, of course, at this point, at all clear: it is far f r o m clear what is involved in looking at something as a system of activities.



A n investigation conducted under the auspices of the practical management of social order does not and cannot provide an answer to the question 'what general principle provides for social order?' If examined f r o m the point of view of an interest in that question then such an investigation will appear to lack an answer to it. However, such a study does not fail to p r o d u c e an answer, since it does not try to answer it. It has withdrawn that question and substituted another one. W h a t is involved here is a perfectly legitimate step in the work of theorising, namely that of varying the givens. A n y approach to inquiry has its boundaries, the matters which stand outside the reach of its inquiries, the things which must be taken on board without examination. Sociology, thus, typically takes it for granted that there is a 'world of daily life', that the c o m m o n p l a c e affairs of life-in-society are regularly carried on and are possessed of an orderly and (relatively) stable arrangement, the object being to identify the conditions which engender and assure the occurrence of such arrangements. From sociology's point of view it is simply a given that the m e m b e r s of society somehow organise their activities to comprise the affairs of daily life, that they somehow co-ordinate their various doings to make up such commonplace matters as the provision of meals, the delivery of mail, the election of rulers, the holding of sporting events etc. etc. Insofar as this is considered problematical, then it will be treated as (characteristically) posing a problem of 'general principle': do people stick at the affairs of life b e c a u s e of - in general - an attachment of shared values and common rules or because of the dull grind of economic compulsion? It is not, we repeat, a complaint that sociologists take for granted the fact that people stick at everyday affairs sufficiently to make happen, to put together, the round of diversified activities: given their problematic there is no reason why they should open this up to inquiry. However, if someone wants to ask about that which sociology usually takes for granted, namely that there is an orderly round of everyday affairs available in the first place and asks how do those ordinary activities put themselves together then they are free to do so. Note, though, that the question concerns not origins but constitution: what do the affairs of daily life consist in? T h e r e is an issue between these two approaches to social order. It is that this latter question cannot be raised within the framework provided by the m o r e usual conception of the problem. There is an issue of access to phenomena. T h e complaint is: one cannot, from within the framework of the traditional conception of order, organise inquiries into the ways in which e v e r y d a y affairs are constituted, how they make themselves up as the e v e r y d a y affairs that they are.



Notice, the claim is not that one must abandon the received frame of reference and take up the new problem, but only that if one wants to take up the problem then one will be compelled to step outside the received framework. The claim is, first, that there is a bona fide and unresolved problem for sociological inquiry which is that of the organisation of everyday affairs as such, and second, that this problem cannot be examined from within, or by simple modification of, the established framework of investigation. In order to open up to investigation the organisation of commonplace activities one needs a new framework, for the conduct of inquiries into social order conventionally depends upon taking these things as given, the accepted form of inquiry requires that these issues be treated as givens. Within sociology proposals such as those we have been sketching are met with objections, many of which are motivated by assumptions about the way in which all sociological investigations must fit within some 'master scheme' and which complain that the proposed studies do not fit comfortably within some envisaged comprehensive framework. Thus, for example, proposals to open up the world of daily life for examination in its own right, as an organised arrangement of activities, is characteristically treated as the basis of a complete and general conception of what sociology is and can do. Hence, it is objected that we cannot take 'everyday life' at face value and that we cannot allow that to be the ultimate locus of sociological inquiry, that we must 'go outside of it' or 'behind it' to really understand it. Thus, the conflict is made to appear as one between those who envisage the world of daily life as the ultimate end of all inquiries and those who (with philosophical and scientific precedents on their side) want to insist that it cannot be. However, this is to treat a proposal to open up a problem for examination as though it comprised a master plan for the future of all inquiry, as though a commitment to consider an (allegedly) neglected topic was an attempt to circumscribe, finally, the whole range of sociological possibilities. Someone proposing to examine the availability of the world of daily life and to shift the way that the problem of social order is viewed might hope, even expect, that this move would have ramifications throughout sociological thought, but it would be a remarkably prescient person who could see if it would have far reaching consequences and what these would be. Criticism, thus, is characteristically of and between projects. We have before us a battery of proposals as to how we might go about sociological inquiry. Sociology is a heavily programmatic pursuit, its main and dominating elements being plans for investigation, actual investigations often being 'toy' versions of projected investigations, exercises to show what we might do if we began to follow a programme through fully seriously. However, the conversion of sociological programmes into a successfully accumulating



collection of investigations often proves more than a trivial and technical problem and argument is, consequently, more characteristically about the promised than the proven success of plans for research. Much criticism, therefore, consists in allegations of constitutional limitation, claims that one approach will be constitutionally incapable of doing what another approach can do, without too much concern to ask whether it is relevant to a given strategy that it be able to do the things that (at least allegedly) it cannot do. Thus, the kind of criticism to which this alternative proposal is subjected are prevailingly to the effect that it will not explain this issue or be able to take account of that one, criticisms issued on the assumption that the proposal outlines and envisages a comprehensive scheme for the treatment of all the problems which might conceivably fall within the domain of sociology. However, as we have outlined it, the proposal does nothing of the sort, and the question of whether it can explain this matter or take account of that one can only come after it has been decided whether it needs to provide an explanation or an account for these, whether its proposal is faulted by offering less than a complete scheme for sociology (or even, on less immodest scale, for the whole of the social sciences), / / i t were to be established that such an approach did need to attempt explanation of this or accounting of that, it would then need to be established what it would take and how such limitations as there are to the proposed strategy are to be determined: how are we to be able to tell, with any confidence, whether a proposed strategy will eventually be able to tackle this, that or the other problem? What a strategy might achieve is really only something to be tested out in practice by applying it and seeing how far it will goThere i s a d e e p divergence of attitude, here, and it is a divergence which makes a difference to one's judgement as to whether certain matters are worth arguing about. Not, that is, because one does not think them important, but just because argument about them — at this point in time — is not going to make any useful progress. An attitude which can be taken is that which we can call 'living in the investigative present'. This is not, we think, the prevailing attitude of the human studies, where one of 'living into the envisaged future' is vastly more common. Much work in the human studies is governed primarily by the need for a sense of direction, by the conviction that we must be going somewhere and that the need is to take us to our destination as quickly as possible. Thus, and in the most classic instance, it is supposed that we are moving towards becoming like one of the natural sciences, and we must therefore do that which, in the present, promises to m a k e us most expeditiously like the natural sciences. Current moves are, therefore, judged first and foremost in terms of where they are likely to let us end up. Thus, for example, CA will be criticised because it does not seem to



enable us to adopt that sort of 'systematic' (i.e. sometimes quantitative) method that we shall need to employ if we are to become more like natural (even genuine) sciences. 'Living in the investigative present' is an attitude which may be adopted in response to the suspicion that judging present moves in terms of an envisaged future requires the kind of prescience which is simply not available. If, after all, a persistent problem for all of us is that of moving from outlining a programme of inquiries to actualising it, then this means that we are all very far from being able to see if things will work out, let alone how they might do so. No one is in any good position to judge which strategies of inquiries are likely to pay off best in the longer term, and we might as well then, consider what we can do now without prevailing concern for where we might be when and if we manage actually to do the things we are now attempting. From the point of view of 'living in the investigative present' it looks as though there is a great deal of 'writing off of problems and strategies as insignificant, worthless, pointless, ineffective etc. where there can be no real understanding of what it is that is being written off. A great many judgements just look very ill founded and vastly premature. For example, one of the things which is subject to recurrent debate amongst sociologists is the possibility of an 'interpretive' sociology. 3 People argue about whether sociology is destined to become a genuine 'hard' science or whether it must always remain an interpretive (and therefore 'soft') one? A r e interpretive methods sufficient to comprise adequate inquiries or must they be supplemented by (or even be supplements to) much more 'objective' forms of inquiry? These debates continue ad nauseam but one thing which is clear about them is that those who join them have little clear idea of what an interpretive sociology might actually be. An opposition to 'programmatic' discussion can result. One need not oppose programmatic discussion as such, but one can object when it acquires the character that it takes in sociology, that of endless programmatics. There is little point in continuing the discussion in that way. If an 'interpretive sociology' is an arguable possibility for sociology, then the best way to find out if it is a viable one is to get on with contriving one, with trying to work out what one might be accepting or rejecting if one goes for an interpretive sociology. Of course, one is not promising a quick solution to the arguments. It cannot be supposed, if one is attempting to move beyond the realm of programmatics, that one will rapidly be able to say whether an interpretive sociology will be, first, a viable strategy and then whether it might be the optimal one. Seriously attempting to put together such a thing as an 'inter-



pretive sociology' cannot be undertaken in the expectation that such a thing can be produced at a stroke, that one's initial efforts will indeed result in anything that is clearly indicative of the final shape of the whole. Living in the investigative present means that the construction will have to be a step-by-step matter, with only the most limited capacity to see where present steps will take us and, very often, leaving us without any capacity to say right now whether we shall, sooner or later, be able to answer a given question or tackle a problem. Developing an interpretive sociology is not like following out a prepared blueprint to construct something whose structure and properties can be anticipated, it is rather a matter of discovering what such a thing might be as one goes along trying to turn a vague idea into something specific. We have been pointing to some things which make relations between the kind of strategy we have been outlining and that from which it deviates and showing it is not that these lead to disagreements on specific points, but that it produces divergent judgements as to how far some matters can be discussed at all (in present circumstances). CA has certainly taken an attitude of living in the investigative present, which means it must be simply unresponsive to much criticism. Much argument is about whether it is worthwhile' following out CA's line of inquiry but such argument invites claims and counter claims as to what CA might come to be, how its inquiries might finally turn out, but CA is in no position itself to say this and hence less than likely to be impressed by critics who seem, somehow, to know this.

The Organisation of Social Actions We have subsumed CA under the heading of sociological approaches but have tried to indicate that its relation to other approaches is likely to be an uneasy one, not because of direct rivalry, but just because they are, so to speak, at a tangent to one another, in some way related and marked by common concern, but in other respects the vehicles of very different policies. We should expect that the relations of CA to some approaches in linguistics may be no less uneasy and for much the same reasons. Sociology, as we have seen, pervasively makes the assumption that the world of daily life is an orderly, predictable place. It assumes that the world of daily life 'makes sense'. When we say this, we are not proposing anything more drastic than that (say) when we see someone in a store handing over money and being given a loaf in exchange that we are seeing someone buying a loaf of bread, or when we hear someone say to someone else 'Take out the



garbage' that we are hearing someone being told to take out the garbage. Ordinary, everyday activities are readily recognisable for what they are, visible as the commonplace occurrences that they are. For the kind of alternative we have outlined above, the fact that these ordinary events are recognisably commonplace occurrences is not something which is detachable f r o m their commonplace character. Being readily recognisable is an essential feature of their commonplaceness and so, naturally, the issue of how ordinary activities can be recognised for what they can become is a central question. How do those who inhabit it make sense of their everyday world? Thus, one way in which an 'interpretive sociology' might be initiated is by examining the ways in which people 'make sense' of their everyday environment, how they 'see the sense' of the most ordinary activities. Seeing the sense of ordinary activities means being able to see what people are doing and saying, and therefore one place in which one might begin to see how making sense is done is in terms of the understanding of everyday talk. It should not be imagined that one is going to set the understanding of people's sayings against the understanding of people's doings, such that we shall then have to face problems as to how sayings relate to doings. O n e focusses on how people make sense of talk as a way of getting access to the examination of the way people make sense of each other's activities, and one sees that making sense of each other's talk is integral to, and often identical with, making sense of each other's doings. It is not as if we have (say) social activities on the one hand and linguistic matters on the other, but that we are undertaking to look at talk as an organised social activity. O n e of the ways in which CA goes about this engenders objections. CA characteristically works with tape and transcript of verbal exchanges, showing little acquaintance — if any — with the circumstances under which the recordings were made or the way of life from which they were extracted. Thus, C A meets with the 'enough data' question right from the start. Does it have enough data to be able to say anything? The answer to such questions must, of course, depend on at least two things: What the data is, and what it is being used to license talk about? From CA's point of view we cannot say what the data is until we have examined it. Can we say anything about social organisation from the contents of tape recordings and transcripts of same? Well, what is there on a tape recording: How can we say unless we listen to it and characterise what is recorded on it (and not just characterise it in some gross and general way, but in terms of its detailed and specific features). If we do that, then we are examining and analysing the talk. From CA's point of view, it is not as if we can determine what data is independently of the analysis of it and thus CA's investigative exercise is to determine the charac-



ter of the data, to discover what the contents of tape and transcript are, and coterminously, to discover what kinds of things can be said about and on the basis of that data. The point is certainly not to suppose that any and all questions can be answered on the basis of such materials but to try to work out, through inspection of them, just how many and what sort of questions can be solved. After all, the value that data can have is not a function of its intrinsic character but of the questions that are put to it, and the ways in which they are designed to extract maximum legitimate value from the materials available. In many ways, then, C A adopts a reverse strategy to that followed by most forms of sociological inquiry. It takes a 'data driven' approach, making its question not 'what data do we need to answer this question?' but, by contrast, the very different question 'what questions can this data answer?' It may, of course, be that work in conversatonal analysis does (as work in any area can do) exceeds its legitimate brief and claims things which it cannot legitimately claim, asserts things that could only legitimately be said if, say, one knew a great deal about the immediate circumstances or general milieu of the data collection but this would be faulty C A , not evidence of the intrinsic inadequacy of its materials or methods. At the very least should not audio recordings be complemented with video recordings? Are not the paralinguistic and kinesic environments of verbal interaction absolutely essential to the interpretation of the verbal component? Is it not even likely that the verbal component is much the less important part of the communicative process. Once again we are in the position in which people are making judgements as to the relative importance of things which they have not examined. How important is talk to 'the communicative process', is it more or less important than the non-verbal component? Who is to say? No one really knows what 'the communicative process' is nor what parts the verbal and non-verbal components play in it since neither have been analysed and described in any systematic way. It was, thus, partly and importantly, in reaction against what can only be called prejudicial judgements about the nature of the communicative process that C A formed its character: it was not prepared to accept that the verbal was much less consequential than the non-verbal channel when no one could say what went on through the verbal channel. Consequently, it set itself the task of taking a good look at the 'verbal component' to see in what that consisted, considered as a socially organised matter. It is probably important to stress that CA's programme is not designed to reverse the judgement on the relative importance of verbal and non-



verbal communication, to claim that the verbal is what matters at the expense of the non-verbal. It is designed to identify and describe the organisation of the verbal exchange as such, to see how the talk is organised as talk, a task which can be conducted quite independently of the examination of non-verbal communication. At least, it is possible to undertake the examination of the organisation of the verbal exchange and to identify some of its organisational features without having also to take systematic account of non-verbal phenomena. It is, of course, tempting to think that one must examine verbal and non-verbal phenomena in conjunction because they are clearly related p h e n o m e n a , but the issues pertaining to the way in which phenomena are to be examined, whether simultaneously or independently, must depend crucially upon the conception of the sequenced character of operations making up the programme of analysis, the kinds of relationships one is looking for. O n e may, for example, be looking for 'correlational' connections between verbal and non-verbal behaviour, seeking to find if the two 'move together'. Thus there is the predominant concern with the relationship between speech and gaze direction, one which is designed to show that there are predictable points in the course of a speech exchange at which parties to it will direct their gaze at each other. Such a method of investigation is not designed to identify the structure of either the verbal or the non-verbal elements of behaviour. It makes no attempt to work out in a thorough or systematic way of what either channel of communication consists, how they are made up and how their respective structures might be integrated. From CA's point of view the examination of the relationship between the verbal and non-verbal aspects of communication is not necessarily to be sought by looking for some correlations between particular features of the two 'components' but, instead, of looking for interconnection of their respective structures: only if one has specified what the structure of the verbal and non-verbal channels respectively are can one start to work out how those structures are interconnected. Thus, though there might well be points at which the analysis of the organisation of the verbal exchange cannot proceed without consideration of the relevance of some 'non-verbal' occurrence there is no reason to suppose that one cannot begin to examine the organisation of the talk without simultaneously examining non-verbal activities. What is it to determine the structure of the organisation of talk? Such a question cannot be given some general answer, but must always be relativised to a point of view. From CA, to determine the structure of the verbal



exchange is to identifhy its properties as an organised system of activities, which means, at its most rudimentary level, to specify what conversation consists in as a succession of actions, and to see how those actions are related one to another, how they build up the conversational sequence. This, however, transforms the character of the problem, away from that of the connection between dimensions of a communicative process and into that of saying how the connections between actions are generated. The primary characteristic of the utterances that CA deals with is often less that they are verbal actions, but that they are actions. There is, thus, no need for C A to insist that verbal actions can only relate to other verbal actions for they may relate, as well, to non-verbal ones. One feature of the interrelation of actions to which C A pays great attention is that of pairing, the relation of 'first' and 'next' action. It is entirely possible for first action to be a verbal action and for 'next' to be a non-verbal one, as for example a first, spoken greeting can be returned with a smile or other gesture, as a spoken offer 'help yourself can be responded to by doing just this. CA is not, then, precluded from looking at 'non-verbal behaviour' except by its own 'one thing at a time' strategy, though if it is to extend its inquiries from talk as such to other aspects of the 'communication process' it will be constrained to do this by looking at such behaviour to see what kinds of actions constitute it or how it relates to the organisation of verbal actions. It will have the first task of doing, for non-verbal behaviour, something analogous to that which it has d o n e for verbal behaviour, namely trying to identify the actions constituting it (which is not, for example, what gaze directions studies try to do). Those who propose to examine verbal and non-verbal communication conjointly may do so in the supposition that they have a superior strategy to that pursued by C A but from the latter's point of view it will not appear so, only that the very phenomenon to which CA addresses its inquiries will be lost. T h e very things that CA wants to examine, the socially organised structure of the verbal exchange, is not something that the proposed mode of analysis is likely to be intended to capture and the things that CA needs to identify are ones it will not discriminate. For example, one may examine Beattie's Talk (1984) as a reiteration of just the kinds of claims about the indispensibility of analysing talk and non-verbal behaviour conjunctly, and which seem to make these claims in an unsuitably generalised manner, without any indication of whether for some purposes it might be the case that one indeed cannot isolate the analysis of the verbal from the nonverbal and in yet other cases that this might be, strategically just the move to m a k e . Beattie's investigations are into such matters as turn taking, interruptions and (that eternal favourite) gaze direction and these are m a t t e r s which are germane to the examination of conversation and to CA



but there is no interest shown by Beattie in, nor evidence that his strategy would enable him to address, those issues which are central to CA's inquiries, those involved in describing the character and organisational interrelaton of utterances, in seeing how the talk constitutes and compounds conversational structures. Claims as to the relative superiority of this or that approach are made with respect to their adequacy in respect of such large and vague topics as 'language', 'interaction', 'communication', 'discourse' etc. with little attention being paid to specific things that a particular strategy may (and usefully) be seeking. Thus 'Discourse Analysis' (hereafter D A ) and Conversation Analysis are set up as being in more or less direct competition, something which is perhaps natural if one thinks of them both as 'approaches to discourse'. If one thinks of them in that way, then it must seem that we cannot have two different and comprehensive approaches to discourse and therefore we must choose one of the two but if one sees that to treat them as candidate approaches to 'discourse' (as if each sought to give comparably general and comprehensive accounts of the same thing) is to fail to recognise that they are actually interested in discourse in very different ways, seek to determine very different things about discourse and have very different ambitions with respect to the treatment of 'discourse as a whole' then one might also see that they do not provide an occasion for choice of this sort. Choice is, of course, inevitable. There are numerous ways in which one may set out to investigate matters in human social life, and there is a vast multiplicity of things to which one might attend. The choices which the human studies face are presented as though they were the kind that scientists in disciplines with some mature theories might face, those of choosing which of two rival theories is best but, in our submission, this is seldom actually the case. There is a need for choice, but it is more of the kind that the economist identified as 'opportunity cost': In order to do one thing one must forego the opportunity of doing another. Few of the strategies available to the human studies are very developed or of much complexity, and they are not capable of accommodating the diversity of interests and problems that different researchers might conceivably have. It is not, then, that an approach can claim to surpass another by incorporating its interests into its own, more comprehensive system, and tending to them there. They tend to claim superiority, rather, by denigrating and seeking to rule out those interests which they cannot accommodate and, as often as not, this means that these interests must be diminished on methodological grounds: one defines legitimate topics on the grounds that they can be treated by a preferred method, rather than identifying preferable methods on the grounds that they provide access to a richer range of topics.



Thus, there must be a choice between 'dicourse analysis' and 'conversation analysis' but, if we are right, it is, initially at least, simply because one cannot do everything at once nor go at things in very different ways simultaneously. One cannot look at talk both from the point of view of CA and that of D A at once: one can look at it from that of either, but not from that of both. There are some overlaps between them, but these do not presage a close convergence or easy integration of strategies. Discourse analysis is concerned to provide a corrective to a tradition of linguistic analysis which has, in its view, been too prone to treat the sentence as an isolated object of analysis without regard for the location of sentences within a sequence of sentences (its discourse aspect) or in terms of its relation to the uses to which language can be put (its pragmatic character). What is being disregarded within the main traditions of language study is, to put it crudely, the fact that language is a social institution. The main tradition of language study knows perfectly well that language is a social institution but the fact that it is does not figure very largely in the form of analysis that is there developed and employed. It has, at best, a tacit presence. However, the attempt to develop that main tradition runs into difficulties, some of which are manifestly because it pays insufficient attention to the fact that language has a part to play in social life and verbal exchanges. Consequently it is felt necessary to give the fact that utterances occur in social contexts a much greater prominence than it has been given hitherto, but such a step is designed as a modification, rather than a transformation, of the frame of analysis developed by the main tradition, the exercise is still primarily a linguistic exercise into which 'sociological considerations' enter in a supplementary role and which, when they do, are handled in a largely ad hoc fashion. As an indication of this, consider the following remarks from Brown & Yule's (1983) Discourse Analysis, the first of two useful and simple illustration of the extent of the problems that we have been discussing. They propose that 'from the discourse analyst's point of view' a promising approach to the study of social meaning 'is offered by a consideration of that area of conversation analysis which investigates turn taking.' They point to the work of Sacks, Schegloff, Jefferson and others in this area and, particularly, to the role of the notion of 'adjacency pair' in this work. They initially grant that the notion of the 'turn' as a unit of analysis is reasonable, but immediately qualify: 'However, most conversational data consists of more substantial "turns" in which several utterances can occur, or in which the basic adjacency pair organisation is difficult to determine.' They provide a data extract, shortly to be reproduced, to make these difficulties visible, and on its basis are able to suggest that some of the interrogative forms function



as both answers and questions, and that the final declarative form is not, in fact, an answer to any of the questions. This provides them with sufficient ground for overall judgement on the capacity of CA: 'the immediate question which springs to mind is how does the analyst determine when an interrogative form counts as a question in an adjacency pair, or as part of an insertion sequence, or even as an answer? This type of question is never really raised by those undertaking the analysis of conversational interaction, largely because little attempt is made to discuss the relationship between linguistic form and the interactive functions proposed.' (Brown & Yule, 1983: 230) Such objections may, in terms of D A ' s own project, be telling ones, if what is sought is a systematic relationship between 'linguistic form' and interactive function. But from CA's angle the making of them amounts to no more than begging the question, for what CA has taken as its most elementary methodological presupposition is that the identification of the interactive function of an utterance cannot be determined in this formulaic way, that it is a local and circumstantial matter, involving the examination of the way the utterance is embedded in the interactional sequence to determine just what interactional role a given utterance has. There have been various arguments designed to show that from the point of view of CA many characterisations of 'linguistic form' are superficial, and do not capture the interactional character of the utterances they are designed to describe. CA does think about the relationship between linguistic form and interactive function, but it does not reach the same conclusions that D A does, that it would be practicable to provide 'the analyst' with a mechanical procedure for identifying the interractive function of a given utterance type. Further, Brown & Yule's objections betray an apparent failure to appreciate that the priorities of CA are substantially different from those of D A at a methodological level. Brown & Yule, and D A more generally, show a concern which is characteristic of many in the human sciences, which is to have an explicit and mechanical method for processing data, where the objective of inquiry is to provide the analyst with a set of worked out categories and explicit procedures for characterising data and where it is a shortcome of the method if it leaves it uncertain or equivocal as to how an instance should be described. However, it is notable that CA does not give primacy to these policies, that its objective is not to equip the sociologist/ psychologist/linguist with methods to describe events in conversation but that it is concerned to identify and describe the methods which the participants in conversation themselves use to order and describe conversational



events. This does not, we should stress, bespeak an indifference to the provision of rigorous control over one's inquiries, but it does mean that the considerations of what would provide the most rigorous treatment of a particular matter will be at variance with those which others in sociology, psychology and linguistics would prefer to use. There is, thus, a policy of systematic transformation of the objects of inquiries into conversational matters. For C A the relevant question is not 'can we, the conversational analyst, provide a surefire way of telling this or that' but, invariably, 'is there a conversational way of telling this or that?' Far from being an evasion of the difficulties or of the standards of scholarship that is, we submit, a matter of pursuing, entirely consistently, one of the suppositions of the programme, namely that the observable character of social phenomena, their visibility for what they are, is not something contingently but essentially connected to the nature of them. It is not just a casual and occasional concern of conversationalists that they should have their co-participants recognise what they are doing for what it is, that they should see the projected sense of remarks and recognise the action implications of an utterance. Having others see what you are saying and responding appropriately is part of the very business of conversation and it is therefore-always relevant to ask, in CA's terms of reference, how are the parties organising their talk to make these things recognisable amongst themselves, what conversational ways are there of seeing or showing this or that? Such questions do not suppose that there must be resolutions to ambiguities and uncertainties, for 'clearing up' obscurities, confusions, misunderstandings etc. is one conversational task amongst others and not always the highest priority item: What some interactional item did is something that may remain wholly unresolved. This point applies, also, to problems in the identification of turns just as much as to the identification of utterance types. T h e problem is not to provide us, conversational analysts, with a way of deciding whether a bit of talk comprises a distinct turn or not, but to see how parties to talk decide this. For just this reason that CA has always placed greatest emphasis on the fact that the turn is itself an interactional^ defined unit, that it is for the participants themselves to figure out whether or not a turn was complete, no stronger solution to such questions being available to t h e m — very often — than that it was possibly complete but this, it transpires, is often good enough for the organisation of conversation's business. Brown & Yule point out that 'most conversational data consist of more substantial "turns" in which several utterances can occur' as though this were something unknown to CA and a basic fact of which it is incapable of taking account, but CA's account of turn taking has precisely been designed to take account of this fact and to enable it, therefore, to treat as 'a



conversationalist's problem' that of getting an extended opportunity to talk in an environment regulated by a turn taking 'machinery' which favours short utterances (to put it crudely). Thus, for example, there has been a t t e n t i o n paid (to give just one example) to the need to anticipate and notify t h e likelihood of a long stretch of talk coming up as in the prefacing of stories: T h e basic observation is that stories characteristically take more t h a n o n e utterance to do (Sacks, no date, 1974). T h e crucial notion, quite a central one to C A , which Brown & Yule f e a t u r e in their discussion, is that of 'adjacency pair' but their judgement on its analytic utility does not begin from a concern with the role that it plays in C A . It is assumed that the notions of 'turn' and 'adjacency pair' are being o f f e r e d as analytic units out of which we could potentially compound a systematic analysis of all the utterances in a conversation, but this is not the way they are perhaps best understood. A d j a c e n c y pairs are singled out for attention in C A because they are widespread p h e n o m e n a in conversation and are (so to speak) very useful t h e r e in resolving some of the problems of co-ordination which confront conversationalists (cf. Schegloff, 1968, 1972). There is no suggestion that every utterance must be part of an adjacency pair or that every sequence must be a composite of such pairs. A d j a c e n c y pairs provide part of the answer to C A ' s main problem, which is 'how is it that parties to conversation are able to co-ordinate turns at talk in such ways that they provide appropriate steps in conversational s e q u e n c e ? ' It just is not to be supposed that such problems are always solved in t h e same way, that something which provides a solution provides the only solution. Adjacency pairs are singled out because they provide one elementary and f r e q u e n t solution to the problem of what to do next in conversation. If a party to conversation can identify a previous utterance as one of a normatively linked pair, such that its occurrence makes the production of a second, corresponding part of the pair appropriate, then that person knows (at least) what kind of utterance is appropriately produced next. Thus, if 'question' and 'answer' are paired in this way, the production of a question m a k e s the provision of an answer appropriate. Of course, since the connection between the parts of the pair is normative, it is not assured that what will h a p p e n is that the appropriate second part of the pair will appear on cue, t h o u g h the fact that the first part has been issued can be most relevant for organising the ensuing talk (as Brown & Yule's example shows well). The e x a m p l e is: George: Zee:

Did you want an ice lolly or not? W h a t kind have they got?


TALK AND SOCIAL ORGANISATION George: How about orange? Zee: Don't they have bazookas? George: Well, here's twenty pence + you ask him.

The fact that one question gets 'answered' with another question is hardly news to CA, and the fact that an interactional sequence does not produce something identifiable as an answer to a question is hardly likely to surprise it either. That a question is asked does not ensure that it will be answered, it does not even guarantee that it will be addressed: questions can go unanswered. CA has, to make just one point, been careful often to speak of 'candidate answers' in order to respond to some of the complexities of the situations questions create: after all, whether something is an answer to a question (sometimes) depends on its informational status, whether it gives the right information, rather than on its linguistic form. The point about the notion of adjacency pairs is not that it predicts, given a first, there will be a next. It explains, rather, what we might call an 'orientational fact', namely that given a first, parties will be looking for a next and hence may find that such an appropriate next did not occur. Brown & Yule's own discussion displays this orientational fact: seeing a question, they start examining the following utterances to see if they can find an answer amongst them? That there is no identifiable answer is, for them, a 'noticeable absence': a non occurrence is a remarkable thing (cf. Sacks, 1972a). If DA complains that CA does not look at the relation between linguistic form and interactive function, then the reciprocal complaint that DA does not pay much attention to the character of the actions that utterances perform is also in order. Brown & Yule do not seem to appreciate that CA's first methodological rule is to look to see what kind of action an utterance performs and from this point of view, George's question is of primary interest not because it asks a question but because it makes an offer. George is not asking Zee about her psychological state, whether she wants a lolly or not, he is offering to buy her one, even offering to get it for her. An 'offer' is also the first part of an adjacency pair, making relevant the next action of taking up or declining the offer. Now, nobody supposes that anyone who is made an offer will immediately accept or reject it, because there can be quite long gaps between making an offer and finally accepting it or rejecting it (e.g. ringing back can be involved, consulting others first etc.) However, if an offer is made then we shall be disposed (as ordinary speakers of the language and ordinary members of the society) to see what the recipient of the offer does next as a response to the offer, and that is, of course, how we understand Zee's first remark, as a conditional acceptance of the offer: she is



not rejecting it outright, but making it depend on what kind of lolly she can have. Thus, we find a quite routine follow up to an offer, namely questions designed to clarify its character, elucidate its conditions and so forth. In this case it turns out that George goes through with the offer to buy the lolly for Z e e but withdraws on the implication that he will get it for her, giving her the money to get her own. Thus, the offer is 'dealt with' though not in terms of a simple accept/reject response, we could not see the sense in the sequence without following it in terms of the relevance of the 'accept/reject' alternative. It is likely that Brown & Yule, again like many of their colleagues in the human studies, have at the back of their mind an idea of these studies as putatively predictive in character such that we should be developing methods which will enable us to say what will happen, to predict what the next action will be if a first one occurs. Their strategy of analysis seems, like many others, to rest upon the assumption that its task must be 'determination of outcomes', i.e. that we must seek to see what happens as a result of, and therefore as predictable from, a set of specifiable conditions. From such a point of view, a notion of adjacency pairs such as we have outlined will probably seem 'slack' since it does not predict, at all, what will happen. Given a first, an appropriate next may occur but it need not. Surely what we want, the demand often goes, is a way of saying what will happen next? There are other possible notions of what our tasks may be, however, and the kind which C A may be seen to be engaged in is rather more in fashion (given the 'new realist' philosophy of science's rising stock, with its emphasis on investigating the nature and constitution of things) than the view of the objective of inquiry as prediction (though, of course, it takes a long time for many forms of inquiry in the human studies to catch up with fashions in the philosophy of science): CA is concerned with the nature and structure of conversation, and as such one thing it is concerned to do is to capture and preserve what conversation is likely to do and it takes it that, for those involved in conversation, it does not have a definitely predictable character. Conversation is a risky business, such that one can seek to predict and control what will happen next, but one is not assured that what one projects will happen. The adjacency pair allows for just that fact, for the production of a first part of such a pair makes relevant, but does not ensure, the occurrence of a next. W e have indicated that we are more concerned to map divergences than to argue the rights and wrongs of them, and we have therefore been mainly aiming to show that Brown & Yule's criticism of CA seems premature and deriving from a lack of appreciation of the very different kind of exercise that C A is from that which Brown & Yule envisage for themselves. The



criticism has been of Brown and Yule's understanding of CA and, therefore, of misconceptions about the basis on which D A might claim superiority. Our arguments do not show that D A is not superior to CA nor does it show that there is something constitutionally wrong with D A : after all, given the nature of our complaint against Brown & Yule it would be premature to do this. If successful, we have shown that (so far) the efforts of CA are neither invalidated nor rendered superfluous by those of DA.

Forms of Systematic Inquiry However, we have not yet touched upon what is offered as the most telling objection against CA by DA, that the former is an unsystematic exercise, and that the latter, putatively more systematic, is therefore the superior one. We shall argue, as before, not that the situation is reversed and that it is D A which is the unsystematic exercise (though we think this is probably so) but only that there is no basis for D A to allege superior systematicity. There are two different conceptions (at least) of what systematicity involves and there is no prima facie reason why one should be vaunted above the other. Consider the second example from Brown & Yule, which involves citing a fragment of (invented) data by Widdowson, A: B: A:

That's the telephone. I'm in the bath. O.K.

and the summary of his argument that 'it is only by recognising the action performed by each of these utterances within the conventional sequencing of such actions that we can accept this sequence as coherent discourse' (quoted in Brown & Yule, 1983). Such an instance shows clearly enough that f r o m D A ' s point of view, the fact that utterances perform actions is a residual one, and hence one which is given no systematic attention. Where some utterances cannot be readily interpreted as a coherent sequence then one may note that they are connected as actions, with one party suggesting the other answer the phone and the second indicating why they cannot do this, the first then accepting this. It is entirely legitimate for D A to treat the fact that utterances perform actions as a matter of only occasional note, since they have quite a specific concern with relations between utterances, which is to find 'coherence' between them, something which can sometimes be done by linguistic features, sometimes by reference to the kind of conventions that a Gricean analysis employs and sometimes by taking note of the kind of



connections that actions may have to one another. However, let us note that from the point of view of a concern with the organisation of social actions such a study policy is quite unsystematic, making no response to the fact that any utterances can be examined as performing a social action and that one can raise, as a problem to be systematically pursued, that of the kind of connections which there may be between one action and another. Such a problem, naturally enough, changes the focus of investigation entirely, replacing the search for linguistic coherence with an investigation of the organisational relations between utterances in a conversation and the way in which utterances are organised into conversations. Such an inquiry requires pervasive (and exclusive) attention to the action-performing role of utterances and to the investigation of their organisational role. A further alteration ensues. D A sees itself as engaging, primarily, in the description of 'cognitive' processes, describing the understandings which enable people to interpret expressions, whilst CA sees itself is treating primarily of interactive processes. From its point of view, D A stands in much the same relation to its elected phenomena as sociology conventionally takes to its own, namely that it takes the givenness of the activities (or utterances) it seeks to examine for granted. People do, in orderly ways, produce mutually intelligible utterances and the issue is to determine how, once those utterances are produced, they are interpreted. F o r C A , however, it is legitimate to ask how the phenomena comes to be available in the first place, how persons are able to organise their activities in such a way as to produce mutually intelligible exchanges of utterances — how is discourse made to happen? Thus, the concern is with the production and management of a socially organised occasion, the production of a flow of activities — verbal ones, as it happens — in co-ordinated sequence. Thus, the utterances which comprise a conversation are to be inspected for the ways in which they generate talk, for their potential in bringing about further talk, and for their capacity to shape the course of subsequent talk. Thus, to revert to the topic of adjacency pairs, one of the things which makes it of considerable interest is its capacity to project further talk: the production of a first part calls for the production (by another speaker) of a next. CA is, then, overridingly concerned with talk as a collaborative matter and with how parties to an occasion can jointly produce an organised sequence of talk, which means that it cannot treat problems of interpretation and understanding independently of those of production. For conversationalists, seeing what an utterance says is not a retrospective, reflective, academic, theoretical matter, but one of direct practical implication. For them, the issue is, abidingly, 'what to say/do now?'. Seeing what kind of action an utterance performs and what kind of next action it implies, invites, demands etc. is the primary issue,



integral to the generation and management of the exchange within which the utterance occurs. CA is concerned, then, with the analysability of conversation in media res. It is not a matter of looking for regularities which will only reveal themselves through the application of sophisticated methods of investigations but of seeking to identify those features of conversational organisation to which conversationalists themselves attend and of seeking to see how they respond to those features. The process of interpretation with which CA is primarily concerned, then, is that which is situated within the conversation itself, which is involved in following and developing the course of the talk, and the investigative strategy is, therefore, directed toward describing the way in which interpretation and understanding are articulated with conversation's organisation. Rather than looking for 'cognitive processes' involved in interpretation, CA — consistently with its policy of looking for conversational solutions to problems — is engaged in searching out conversational practices for achieving and exhibiting understanding. Note that understanding is regarded as an achievement. That is, it is not something which is automatic or assured and parties must therefore reciprocally design their respective remarks in such ways that the projected recipient of them will see what they are saying. It is partly because of this that CA gives prominence to 'recipient design' (Sacks & Schegloff, 1979) as a feature of the organisation of conversational utterances. Utterances in conversation are not directed towards anonymous 'speakers of the language' but toward specific others, and conversationalists therefore pay pervasive attention to the issues of to whom they are talking, what such persons may be expected to know, what they will know without having to be told, what they will be able to read into what has been said without it being put into so many words, what they will be interested in and so forth. 'Recipient design' points the investigation toward the ways in which utterances are constructed specifically so that they will be understood by this recipient. Since understanding is not guaranteed and since such practices as conversationalists have for designing recipient-intelligible utterances are not failsafe, then misunderstandings are a possibility in conversation and, therefore, they provide a source of conversational trouble, there is a need, then, to describe the ways in which such troubles reveal themselves, may be detected, diagnosed and resolved. All in all, CA is examining conversation — to use a term from ethnomethodology — as a self-explicating system,4 a policy which can be pursued with respect to social activities and settings of all kinds. The policy involves seeing how the setting makes its own organisation visible to participants,



how its arrangements can be examined from within so that people can see 'what is happening here' and determine 'what we're supposed to do now'. Thus, when applied to conversation it means examining how the talk making up the conversation is organised so that parties to it can determine 'what has been said', 'what we are talking about', 'where we are in this conversation', 'what further course this conversation might take' and so on. The answers to such questions are to be sought by looking at how the parties are talking, at whether they are talking in such ways as to be (say) 'opening up a conversation' or 'preparing to bring the talk to a close', whether they are talking in such ways as to show that 'one party follows what the other is telling', that 'one party is checking out that the other has not misunderstood', that the recipient is 'seeking clarification of the teller's remarks', whether the parties are 'in the midst of talking about some particular topic', are 'searching around for something to talk about' or are 'competing to decide which of two or more preferred topics they are going to talk about'. Such things as these are 'audible phenomena' in conversation. They are not ones which can only be discovered by statistical analysis or by close timing of utterances, the pauses between them etc. (there are, of course, some features of conversation which would be discoverable only by such means). That is, if we listen to a tape recording of a conversation we can simply hear that (say) everyone involved in talking about the same topic or (alternatively) we might hear the talk as involving two parties trying to get different topics started at once, and CA's problem is: How are these phenomena made audible to us, how can we hear them in the utterances produced? What features of and relations between utterances enable us to say that they are addressed to the same topic or, alternatively, that they prefigure two quite different topics. The point is, of course, to work out how those phenomena are available to the persons in the conversation such that those hearing a particular remark can see what topic it could introduce such that they can then say something which will also be 'about the same topic' or which might cut that topic off before it develops? Such inquiries are wholly consistent with the maintainance of CA's focus on its elected phenomena, which is the organisation of conversation as a sequence of turns at talk, and with its task of describing the relationship between turns at talk as making up organised overall sequences. All such inquiries are dominated by and organised around the topic of 'turn taking'. In the demarcation of academic territory, takeover bids are sometimes made. D A seems prepared to make one for CA. If one is interested in 'discourse', then conversation is one form of discourse and one thing which obviously occurs in conversation and some other form of discourse is 'turn taking'. Conversational analysis has developed a fairly elaborate account of



turn taking and it seems, therefore, only reasonable to suppose that one might include the phenomenon of turn taking within the range of topics covered by a 'discourse analysis'. However, such assumptions are reasonable enough if it is thought that turn taking is CA's topic but this is not, we think, the best way to consider the matter. It is perhaps more apposite to see the examination of turn taking as the method rather than the topic of CA. T h a t does not involve looking at turn taking as one aspect of conversation but, instead, the examination of conversational activities wholly from the point of view of the necessity for turn taking. The policy is to examine anything and everything in conversation to see in what ways it is affected by/responsive to the basic organisational fact that conversation is a turn taking pursuit. Conversation is, virtually by definition, the alternation of turns at talk and, therefore, whatever happens in or is done through conversation, must 'fit in' to the environment of alternations at talk. CA's work, then, involves considering conversational materials as comprised of 'sequential objects', as p h e n o m e n a which are environed by, constituted of and distributed over turns at talk, and CA aims to describe the ways in which the occurrences of conversation are organised as sequential objects. CA's strategy is, then, to see just how far the description of events in conversation relative to the organisation of turn taking can be taken, perhaps finding that matters which may appear, on first inspection or from the point of view of some other kind of analysis, entirely independent of turn taking are significantly shaped by turn taking requirements. Thus, there can be no question (from CA's point of view) of making an a priori demarcation of topics between those which can be considered in terms of turn taking and those which cannot: the only way to determine how pervasive and in what ways the requirement for turn taking makes itself felt (organisationally speaking) is to examine phenomena relative to their placement in and constitution by turns at talk. Far from providing a treatment of one subordinate topic within 'discourse', then, C A provides a method which could be applied in the reconsideration of a very broad range of topics within that area, of looking at the p h e n o m e n a that D A is intended to cover as turn taking phenomena (insofar as they occur in conversation or are located in other turn taking systems). Looking at the matters treated by D A in terms of turn taking considerations could have far reaching ramifications for the strategies and topic structure of 'discourse analysis'. T h e point of this section has been to counter claims that CA is unsystematic. It seeks to be thoroughly systematic in its single minded application of a point of view in the examination of a central problem. It seeks to



view conversation as a system of activities and to see how such a system can be organised 'from within', how participants in talk can collaboratively construct orderly sequences of turns at talk and thereby accomplish routine activities of their life in society. It is persistent in its examination of conversation as an organisation of interrelated social actions and equally persistent in its examination of how those actions can be embedded in and concerted through the turn taking sequence. It is equally persistent in its exclusion of phenomena and problems which cannot be treated in such terms and in maintaining its focus on the organisation of talk as such. It is this last self-imposition which is apt to be regarded as CA's most gross, obvious and grievous error. How can one possibly know what is going on in a social situation (and in the talk which comprises it) without knowing a very great deal about the situation, the characters involved in the talk, the structure of their relations, the history of their personal connections, the business they have in hand etc.? Such a question is probably best looked upon as rhetorical, meant to need no answer since possessing the implication that one cannot possibly know what is going on in a social situation without knowing a great deal about it. Simply having a tape recording of some talk cannot, surely, be enough? 'Enough for what?' is the only sensible response. Any method can be abused, of course, and it is entirely possible that people will seek to use tape recordings of talk as the basis for claims which they cannot, on the basis of such data, make (but making claims which are ostensibly licensed by those actually unsupported by one's data is hardly a failing unique to conversation analysts). The fact that a method can be abused does not mean that it is intrinsically deficient: whether or not one can rely on tape recordings alone depends, very much, upon what one is doing with them. Materials do not intrinsically possess or lack value (one might, for example, despise archaeology on the grounds that it involves grubbing about in the middens of lost civilisations). How much use can be made of materials will relate to the problems that you have and the techniques that you develop to treat them. Let it be remembered, then, that CA's focal concern is with the talk and therefore a tape recording is a recording of the very phenomenon that CA intends to inspect. It does not commence upon the inspection of those materials and that phenomena on the basis of suppositions that it will be able to do this or that, but initially undertakes the examination just because it does not know what such data might be good for. Until it has examined them and tried to analyse them CA does not know what it will find on tape recordings, what kind of phenomena they will make available to it and what kinds of problems it will be able to pose and resolve through the investiga-



tion of such materials. The nature of tape recordings as data is something to be discovered by inquiry: What they are data of is not to be determined in advance. Such determinations should be cautiously made. It is not a matter of making such materials answer any and every question in sociology, but of seeking to see which questions can be put to them in such ways that they can be answered through these materials. Clearly there must be issues which cannot be answered on the basis of tape recordings of talk alone and without familiarity with many 'background' matters involved in the situation and the relations between those involved there are many things one cannot definitely determine about some sequence of talk. However, the response of CA to that fact is to try to develop a mode of analysis which is largely independent of such knowledge. It is not trying to devise a form of inquiry which can substitute for the possession of local, specialised or expert knowledge in understanding some things about social situations and organisations of activities but in trying to identify those things which can be studied without reliance on such knowledge. Reliance on the talk/transcript alone is a device for enabling discriminating analysis and specification of relations between organisational features. It is not enough to make programmatic pronouncements to the effect that there are (must be) relations between talk and social context or that talk itself is a form a social action. The need is to be able to say just which features of talk relate to just what kinds of features of the social context, to say just what kinds of social actions make up sequences of talk and how they do so. Thus, C A has sought, as its prevailing objective, to show how the social context of the occasion 'the conversation' is relevant to the production and interrelationship of utterances, to show just what the activity 'talking together' consists in, to achieve the analytic isolation of those features of the verbal exchange which are shaped by the fact that they are being done through conversational talk. Such a method does not involve denying that things which happen in talk may be decisively shaped by such matters as the respective social standings of the parties involved, their organisational capacities, their personal connections, the history of their relationship etc. for, of course, it does involve (often) disregarding the ways in which this might be so in order to isolate and examine pure turn taking and conversational phenomena. CA is quite able to take notice of the fact that persons in conversation have social positions and affiliations without modification of its basic strategy. Thus, it can assume (under the principle of recipient design) that conversationalists will be attentive to just such matters in interpreting others' utterances and in



designing their own. However, it is the kind of interest that is to be taken in such matters that is at issue and CA's is, as we have already indicated, in the audible character of conversational occurrences, with determining how they can be heard in the talk. Let it be clear that CA is not aiming to provide methods which will enable us to say with greater definiteness than a conversationalist might just what is happening on any particular occasion of talk. CA's aim is not to uncover the facts about the particular social situations that it examines, to establish for sure (say) that these conversationalists were a conjugal pair, old college chums or employer and employee. It is not indifferent to these facts either, however, for it can legitimately, and within its elected frame of reference, concern itself with the organisational consequences that they do/could have, with their possible import for the structure of the conversational sequence. The investigations are, then, only properly understood if they are recognised to be into organisational possibilities, and the issues which arise from this pose questions about what possible differences such facts would make and just how the memberships and affiliations of participants matter for the course and character of talk. Thus, the issue is not (say) to be certain that this conversing couple are man and wife but to be clear about what difference it would make to the interpretation of what they are saying to each other if they were man and wife? Such a question is not to be answered by establishing that they are man and wife but by examining the talk again to see which and in what ways its features are linked to social membership. The analysis seeks to identify formal possibilities rather than to pin down instances. The common supposition is, we suppose, that it is the organisation of relationships in the social setting and social relationships which determine the course of talk and, thus (should), govern how we hear (i.e. understand) the talk. C A makes a different supposition, namely that the character of the social situation and the nature of the social relationships between the participants are audible in the talk. Give someone a transcript and they can, very often, get quite a definite sense of who the parties to the talk are, in what capacities they are relating to one another, what kind of personal relationships they have and a great deal more beside. Thus, from listening to a tape one can soon tell, say, that it is a recording of a classroom activity, what kind of class it is, which person is the teacher, which pupils are 'teacher's favourite' etc. Similarly, listening to the opening of a telephone call in which the parties play 'identifying each other by sound of voice' one can see that these are familiars, that they know each other well enough to anticipate mutual recognition by voice alone and so forth. Thus, one can entirely legitimately and without need for resort to more or different kinds of data,



raise questions about the ways in which talk can embody and constitute social relations and not by making unjustified assumptions about facts to which one does not have access. One can do this by examining the way in which social relations are 'audibly present' in verbal exchanges, seeking to determine just what it is about a sequence of talk which makes it quite audibly (say) a conversation between old friends, a student ringing a teacher at h o m e , a member of the public calling an organisation in search of help or a service? There has been, in sum, a shift from considering how social relationships determine the course of talk to asking what social relationships consist in, considered as exchanges of talk.

Conclusion W e have been looking at some of the problems involved in the relations of Conversation Analysis and discourse analysis and have been trying to suggest that these are often matters of misunderstanding rather than of direct disagreement. We have, or course, looked at this from the side of CA, suggesting that its critics in D A show, by their objections, that they do not really see what it is about. We do not thereby imply that those from CA who criticise D A are necessarily any more perceptive in their assessment of what that is about and what might be wrong with it. Whether they are or not would require another investigation. Nor do we intend to suggest that were the misunderstandings cleared up that the prospect of disagreement would evaporate with them. No, we have no doubt that, as carried on, CA and D A are quite incongruous and cannot be fitted together (not, at least, without considerable change in one or the other). In its crudest terms, we can say that one of the difficulties is this: that, by and large, D A is motivated to idealise out the very things that Conversation Analysis wants to examine. T o say this is not to make a criticism but to point to a rigidity. D A cannot readily adapt to take into account those things (appertaining to the interactional coproduction of an ordinarily orderly exchange of talk) which are the very stuff of CA's preoccupations. It cannot do so because its investigative method depends upon, consists in, abstracting out those things. For it to take notice of CA's problem, issues and phenomena would, thus, require an extensive re-orientation of its whole mode of analysis. It cannot just take the topics CA treats and add them to the list of things with which it deals. Neither, of course, can CA just encompass the stuff of DA's inquiries since the former's methodological ideals are just such as deny to C A the capacity to adopt the kinds of idealisation and abstractions which are



the stuff of the latter. For CA to take up those would be for it to deviate from its own programme of inquiry. T h e relations between things viewed from the standpoints of D A and C A , we are suggesting, is rather like that between the two components of a gestalt switch. What one is looking at is, in one sense, the same thing but what is seen is very different and the transition between one picture and another is abrupt, discontinuous. All too often one's remarks on matter such as these (i.e. on the interrelation of disciplines or approaches) are listened to for counsels of hope or despair. Either one is saying that there is hope for much closer and more effective co-operation between different kinds of inquiry or one is saying that there can be no hope of this, that they cannot even talk to each other. O u r discussion has been conducted under the attitude of 'living in the investigative present' which we identified earlier and thus directed toward saying what the situation presently is. What that might portend for the future is something we cannot foresee but toward which we would take neither an 'optimistic' nor a 'pessimistic' stance. At present there are severe difficulties of understanding between alternative strategies in the human studies (the kinds of difficulties between D A and CA being the sort also found in many other areas). O n e of the key difficulties which creates misunderstanding is, we think, that too little concern is given to identifying the level at which problems arise. T o o often the disputes are focussed upon specific issues when, as we have tried to show in this case, the divergence is at the level of frameworks. Divergence between frameworks is very different from disagreement within frameworks. Consequently much criticism is quite ineffective because it is made as though between parties who share the same frameworks when it is precisely on those frameworks, on the whole shape of modes of analysis, that they differ. We have, thus, tried to show how DA's criticism of C A is made as though it were of an enterprise directed to the tasks and sharing the assumptions of D A when it is not this. Within a shared framework, it is the case that one cannot have two parties saying very different things, without raising the question, which one of us is right? In a divergence of frameworks, however, the situation is not at all the same. The relationship between frameworks is often best cast in terms of relative power. If m o d e of analysis A can adequately handle its own problem and phenomena and can also encompass the range of problems with which framework B is concerned, then framework B becomes redundant. Situations like this require the development of modes of analysis of relatively great power, but this is what we do not (at present) have in the human studies. Mode of



analysis A does not usually have the capacity to take up and solve the problems faced by mode of analysis B so in an attempt to claim superiority it is more apt to try to 'write off the latter's problems as unimportant, irrelevant and so on, aiming to monopolise the field by restricting the range of problems that can be acknowledged to those with which it is competent to deal. This, perhaps, explains why method assumes the peculiar, dictating position that it does in the human studies: One finds, that is, not the insistence that we ought to develop methods that will enable us to tackle problems but, much more often, the contention that we should restrict our problems to those our preferred methods can handle. The fact is (at the present time) that the modes of analysis available in the human studies are of limited power relative to each other, such that attempts to claim the superiority of one over another (except on some very specific point) are, at best, premature: D A seeks comparative superiority over C A but does so without any clear conception of what it is that C A is trying to do and what it (CA) would regard itself as having achieved and, hence, without any conception of what it (DA) would have to do to match or surpass that. A n understanding of that would require a much more thorough understanding of the whole mode of analysis within which C A operates. There is (we think) much more heterogeneity than direct disagreement in the human studies for there are many very different kinds of problems that can be raised that will require very different kinds of solutions. In setting up strategies it is largely the case that one can only get one off the ground by restricting it, by excluding many questions and disregarding many difficulties. This is not a criticism of any one approach but a suggestion that, in the primitive and partial character of our efforts, we are all very much in the same boat. There are plenty of controversies, of course, but these are (we suggest) because, guided by a preoccupation with developing sociology, or linguistics, or psychology, we treat primitive attempts to get a grip on some problems as a candidate solution to the question of what general strategy the discipline (or even the human studies) as a whole should adopt. Of course, these heterogeneous strategies cannot all provide the right general approach, only one could do that. However, the evidence of experience is that no one of them provides anything like a general approach. This approach opens up some phenomena and allows one to get at certain problems and provides a reasonable solution to them, but (usually) at the price of having to close one's eyes to various difficulties and of having to entirely disregard a range of problems: that approach, by contrast, can deal with some of those neglected problems but only by setting aside the problems that the other gives primary attention to etc. etc.



W e have been making our points largely in terms of the conflict of frameworks, but they apply also to assumptions about the complementarity of strategies. The possibility of interdisciplinary collaboration is not something to be raised and pursued just because strategies are concerned with what are (superficially) the same phenomena. These relations, too, need to be considered at the level of frameworks. It is for this reason that we have emphasised CA's sociological character so persistently, not because we hold that it is better to be sociological than linguistic or anything like that but simply to show that what is identifying and distinctive about CA, what gives sense and force to its moves and findings, depends upon recognising its primary disposition to treat issues as 'social organisational' ones. Putting CA together with more linguistically motivated forms of inquiry in a systematic way which would not just divest the former of its raison d'etre would not be a simple matter of taking more notice of the fact of turn taking in conversation, but of integrating what are, at this time, quite incongruous frames of reference: some radical revision of one, or both, would be required but that (which is where we came in) can only be done if one has a strong understanding of the respective modes of analysis involved, something which has — so far — rarely been in evidence.

Notes to Chapter 12 1. 'Discourse' is currently a very popular word and there are numerous quite different strategies which project themselves as some form of 'discourse analysis'. In this discussion we have in mind that kind of discourse analysis which draws its inspiration from linguistic modes of inquiry and which finds expression in books like Brown & Yule (1983), Coulthard (1977) and Labov & Fanshel (1977). These sources are themselves fairly heterogeneous in their views of what discourse analysis actually consists of but somewhat more uniform in their conviction that, whatever that is, it is an improvement over Conversation Analysis. 2. W e refer to 'ethnomethodology'. The canonical, but difficult, account of these matters is given in Garfinkel (1967). Simpler introductions are provided in Sharrock & Anderson (1986) and Heritage (1984). 3. 'Interpretive sociology' is, broadly speaking, that kind which seeks to give 'the actor's point of view' a central part in schemes of sociological analysis, and which, therefore, requires a (more or less) significant feature of its operations the description of circumstances as seen by those who must act in them. 4. Pollner (no date) gives the best account of this but a more easily obtainable statement is Pollner (1979).


R. ANDERSON, Senior Lecturer, Department of Sociology, Polytechnic, Manchester, U.K.


G. BUTTON, Senior Lecturer, Department of Social and Political Studies, Plymouth Polytechnic, Plymouth PL4 8AA, U.K. C. GOODWIN, Associate Professor of Anthropology, University of South Carolina, Columbia, SC 29208, U.S.A. G. JEFFERSON, Research Associate, Department of Sociology, University of York, York, U.K. J. R. E. LEE, Lecturer, Department of Sociology, University of Manchester, Manchester M13 9PL, U.K. A. POMERANTZ, Assistant Professor Department of Speech, Temple University, Philadelphia, Pennsylvania, U.S.A. H. SACKS, Late Professor of Social Science, University of California, Irvine, California, U.S.A. E. SCHEGLOFF, Professor of Sociology, University of California, Los Angeles, CA 90024, U.S.A. W. W. SHARROCK, Reader, Department of Sociology, Manchester University, Manchester M13 9PL, U.K. D. R. WATSON, Senior Lecturer, Department of Sociology, Manchester University, Manchester, M13 9PL, U.K.



A N D E R S O N , R . J . , HUGHES, J . A . & SHARROCK, W . W . 1985, The Sociology


London: Longman. ATKINSON, J.M. & DREW, P. 1979, Order in Court: The Organization of Verbal Interaction in Judicial Settings. London: Macmillan. ATKINSON, J. M. & HERITAGE, J. (eds) 1984, Structures of Social Action: Studies In Conversation Analysis. Cambridge: Cambridge University Press. AUSTIN, J. L. 1962, How To Do Things With Words. Oxford: Oxford University Press. B A U M A N , R . & SHERZER, S. ( e d s ) 1974, Explorations

In The



Speaking. Cambridge: Cambridge University Press. BEATTIE, J. 1984, Talk. Milton Keynes: Open University Press. BENSON, D . J . & HUGHES, J . A . 1983, The




London: Longmans. BLOUNT, B. & SANCHES, M. (eds) 1975, Sociocultural Dimensions Of Language Use. New York: Academic Press. BROWN, G. & YULE, G. 1983, Discourse Analysis. Cambridge: Cambridge University Press. BUTTON, G. 1985, End Of Award Report: The Social Organisation Of Topic Closure In Naturally Occurring Conversation. G00230092. London: Economic and Social Research Council. — 1986, Conversaton-In-A-Series. Paper presented to the Talk and Social Structure Conference, University of California Santa Barbara. — forthcoming, Varieties of Closings. In G. PSATHAS (ed.), International Competence. Norwood, N.J.: Ablex. BUTTON, G. & CASEY, N. J. 1984, Generating the Topic: The Use of Topic Initial Elicitors. In J. M. ATKINSON & J. C. HERITAGE (eds), Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press, 167-90. B U T T O N , G . , D R E W , P . & HERITAGE, J . ( e d s ) 1986, I n t e r a c t i o n a n d L a n g u a g e U s e .

Special Double Edition, Human Studies, Vol. 19, Nos. 2-3. BYRNE, P.S. & LONG, B . E . 1976, Doctors Talking To Patients. London: H.M.S.O. ClCOUREL, A. 1964, Method And Measurement In Sociology. New York: Free Press. CORSARO, W. A. 1979, We're Friends, Right?: Children's Use Of Access Rituals In A Nursery School. Language In Society, 8:315-36. 323



COULTER, J. 1975, The Operations Of Mental Welfare Personnel. Unpub. Ph.D. University of Manchester. — 1979, The Social Construction Of Mind: Studies In Ethnomethodology And Linguistic Philosophy. Totowa, N.J.: Rowan and Littlefield. — 1982, Rethinking Cognitive Theory. London: Macmillan. COULTHARD, R. M. 1977, An Introduction To Discourse Analysis. London: Longmans. DOUGLAS, J. D. (ed.) 1970, Understanding Every-Day Life: Towards The Reconstruction Of The Sociology Of Knowledge. Chicago: Aldine. ELIAS, N. 1966, Reflections on personal pronouns. Unpublished paper, University of Leicester (Sociology Dept). ERVIN-TRIPP, S. 1976, Is Sybil There: The Structure of American English Directives. Language In Society, 5:25-66. EVANS-PRITCHARD, E. E. 1937, Witchcraft, Oracles And Magic Among TheAzande. Oxford: The Clarendon Press. FISHER, S. & TODD, S. D. (eds) 1986, Discourse and Institutional Authority. Alex Publishing Corp. Norwood. New Jersey. 07648. GARFINKEL, H. 1967, Studies In Ethnomethodology. Englewood Cliffs, N.J.: Prentice Hall. GARFINKEL, H . , LYNCH, M . & LIVINGSTON, E . 1981, T h e W o r k O f A D i s c o v e r y

Science Construed With Materials From The Optically Discovered Pulsar. Philosophy Of The Social Sciences, 11:138-58. GARFINKEL, H. & SACKS, H. 1970, On Formal Structures Of Practical Actions. In C. MCKINNEY & E. A. TIRYAKIAN (eds), Theoretical Sociology. New York: Appleton-Century-Crofts, 338-66. GOFFMAN, E. 1978, Response Cries, Language, 54: 787-815. GOODMAN, N. (ed.) 1953, Fact, Fiction And Forecast. Cambridge, Mass.: Harvard University Press. GOODWIN, C. 1981, Conversational Organisation: Interaction Between Speakers And Hearers. New York: Academic Press. GOODWIN, M. H. 1980, Process Of Mutual Monitoring Implicated InThe Production Of Description Sequences. Sociological Inquiry, 50, 3-4: 303-17. In D. H. ZIMMERMAN & C. WEST (eds), Special Double Issue on Language and Social Interaction. GUMPERZ, J.J. & HYMES, D. (eds) 1972, Directions in Sociolinguistics. New York: Holt, Rinehart and Winston. HARRE, R. 1972, The Philosophies Of Science. Oxford: Oxford University Press. HEATH, C. 1979, In Breaking Co-Presence. Unpublished manuscript. University of Surrey. HEIL, J. 1981, Does Cognitive Psychology Rest On A Mistake. Mind. Vol. XC No. 359: 321-42.

HERITAGE, J. C. 1984, A Change-Of-State Token And Aspects Of Its Sequential Placement. In J. M. ATKINSON & J. C. HERITAGE (eds), Structures of Social Action: Studies in Conversation Analysis, 299-345. HEYDON, J. D. 1975, Cases and Materials on Evidence, London: Butterworths. JEFFERSON, G. 1972, Side Sequences. In D. SUDNOW (ed.), Studies in Social Interaction, 294-338. — 1973, A Case Of Precision Timing In Ordinary Conversation: Over-Lapped Tag-Position Address Terms In Closing Sequences. Semiotica, 9:47-96. — 1978, Sequential Aspects Of Story Telling In Conversation. In J. N. SCHENKIEN



(ed.), Studies in the Organization of Conversational Interaction, 219-48. — 1979, A Technique For Inviting Laughter A n d Its Subsequent Acceptance/ Declination. In G . PSATHAS (ed.), Everyday Language: Studies in Ethnomethodology, 79-96. — 1980a, E n d Of G r a n t Report O n Conversations In Which 'Troubles' or 'Anxieties' A r e Expressed ( H R 4805/2). London: Social Science Research Council. Mimeo. — 1980b, O n "Trouble-Premonitory" response T o Inquiry. Sociological Inquiry ,50: 153-84. — forthcoming, T w o Explorations Of The Organisation Of Overlapping Talk In Conversation. Tilburg, Netherlands: Tilburg Papers In Language And Literature, N o . 28. JEFFERSON, G . & LEE, J. R . E . 1981, The Rejection Of Advice: Managing The Problematic Convergence Of A 'Troubles-Telling' and A 'Service Encounter'. Journal Of Pragmatics, 5:399-422. JEFFERSON, G . SACKS, H . & SCHEGLOFF, E . A . 1984, On Laughter

in Pursuit


Intimacy. C e n t r o Internationale di Semiotica e di Linguistica, Working Papers, Università di U r b i n o . Italia, No. 135. JEFFERSON, G . & SCHENKIEN, J. N. 1977, Some Sequential Negotiations In Conversation: U n e x p a n d e d and Expanded Versions Of Projected Action Sequences. Sociology, 11: 87-103. Reprinted in J. N. SCHENKIEN (ed.), Studies in Organization of Conversational Interaction, 155-72. KATZ, B. 1972, A n Examination Of T h e Methodological Determination Of A Reason For A Visit In A Medical Setting. Mimeo, University Of British Columbia. LABOV, W . & FANSHELL, D . 1977, Therapeutic Discourse. New York: Academic Press. LEMPERT, R . O , . & SALTZBURG, S. A . 1982, A Modern Approach To Evidence (2nd e d . ) American Casebook Series: West. LEVINSON, S. C. 1983, Pragmatics. Cambridge: Cambridge University Press. LOWE, I. 1969, A n algebraic theory of English pronominal reference (Part I), Semiotica, 1, 4, pp. 397-421. LYNCH, M . , LIVINGSTON, E . & GARFINKEL, H . 1983, T e m p o r a l O r d e r in l a b o r a t o r y w o r k . I n KNORR-CETINA & MULKAY, p p . 205-38.

MACKAY, R . W . , 1974, Conceptions Of Children And Models Of Socialisation. In R . TURNER ( e d . ) , Ethnomethodology,


McKlNNEY, C. & TIRYAKIAN, E. A . (eds) 1970, Theoretical Sociology. New York: Appleton-Century-Crofts. MAYNARD, D . 1975, Topic Initiation In Conversation. U n p u b . Master's Thesis, University of California Santa Barbara. MEHAN, H . 1978, Structuring School Structure. Harvard Educational Review, 45: 311-38. PATERSON, A . 1982, The Law Lords. London: Macmillan. PIKE, K . L . 1966, Modelling pronominal reference, Language, 41, pp. 192-215. PIKE, R . L. 1973, Sociolinguistic Evaluation Of Alternative Mathematical Models: English Pronouns. Language 49:121-60. POLLNER, M. n o date, 'Notes O n Self-Explicating Settings'. U n p u b . Man. University Of California Los Angeles. — 1979, Explicative Transactions: Making A n d Managing Meaning In Traffic C o u r t . In G . PSATHAS (ed.), Everyday Language: Studies in Ethnomethodology, 227-53. POMERANTZ, A . M. 1978, Compliment Responses: Notes O n The Co-operation Of



Multiple Constraints. In J. N. SCHENKIEN (ed.), Studies in the Organization of Conversational Interaction. New York: Academic Press, 79-112. — 1980, Telling My Side: 'Limited Access' As A 'Fishing' Device. Sociological Inquiry, 50:186-96. — 1981, Speaker's Claims As A Feature Of Describing: A Study Of 'Presenting The Evidence For'. Paper presented to the 76th Annual Meeting of the American Sociological Association, Toronto, Canada. PSATHAS, G. (ed.) 1979, Everyday Language: Studies In Ethnomethodology. New York: Irvington. — (ed.) 1983, Interactional Competence. Norwood, N.J.: Ablex. SACKS, H . 1963, Sociological Descriptions. Berkeley Journal Of Sociology, 8:1-19. — 1964-72, Unpublished transcribed lectures, University of California, Irvine. (Transcribed and indexed by G. Jefferson.) — 1971, Lecture, University of California, Irvine, April 26. — 1972a, An Initial Investigation Of The Usability Of Conversational Data For Doing Sociology. In D. SUDNOW (ed.), Studies in Social Interaction. New York: Free Press, 31-74. — 1972b, On The Analyzability Of Stories by Children. In J. J. GUMPERZ & D. HYMES (eds), Directions in Sociolinguistics. New York: Holt, Reinhart and Winston, 325-45. Reprinted in R. TURNER (ed.), Ethnomethodology, 216-32. — 1974, An Analysis Of The Course Of A Joke's Telling In Conversation. In R. BAUMAN & J. SHERZER (eds), Explorations in the Ethnography of Speaking. Cambridge: Cambridge University Press, 337-53. — 1975, Everyone Has To Lie. In B. BLOUT & M. SANCHES (eds), Sociocultural Dimensions of Language Use. New York: Academic Press, 337-53. — 1978, Some Technical Considerations Of A Dirty Joke. In J. N. SCHENKIEN (ed.), Studies in the Organization of Conversational Interaction, 249-70. (Edited by G. JEFFERSON from unpublished lectures: fall 1971, lectures 9-12.) — 1983, On members' measurement systems. In G. PSATHAS (ed.), Interactional Competence. Norwood, N.J.: Ablex. — 1984, Notes On Methodology. In J. M. ATKINSON & J. C. HERITAGE (eds), Structures of Social Action: Studies in Conversation Analysis, 21-27. — no date, Aspects Of The Sequential Organisation Of Conversation. Unpublished Mimeo. SACKS, H . & SCHEGLOFF, E . A . 1979, Two Preferences In The Organisation Of

Reference To Persons In Conversation And Their Interaction. In G. PSATHAS (ed.), Everyday Language: Studies in Ethnomethodology. New York: Irvington, 15-21.

SACKS, H . , SCHEGLOFF, E . A . & JEFFERSON, G . 1974, A Simplest Systematics For

The Organisation Of Turn-Taking For Conversation. Language, 50: 696-735. (A version in J. N. SCHENKIEN (ed.), Studies in the Organization of Conversational Interaction, 7-55.) SCHEGLOFF, E. A. 1968, Sequencing In Conversational Openings. American Anthropologist,

70: 1075-95. ( R e p r i n t e d in J. J. GUMPERZ & D . HYMES (eds),

Directions in Sociolinguistics, 346-80.) — 1972, Notes On Conversational Practice: Formulating Place. In D. SUDNOW (ed.), Studies in Social Interaction. New York: Free Press, 75-119. (Reprinted in P. P.GIGLIOLI (ed.).)

— 1984, On Some Questions And Ambiguities In Conversation. In J. M. ATKINSON & J. C. HERITAGE (eds), Structures of Social Action. Studies in Conversation



Analysis, 28-52. Reprinted from Pragmatics Microfiche, 1976,2.2 D8-G1. — forthcoming, The Routine As Achievement. Human Studies, 9, 2-3. Special D o u b l e Edition on Interaction And Language Use, edited by G. BUTTON, P . D R E W & J . C . HERITAGE. SCHEGLOFF, E . A . , JEFFERSON, G . , & SACKS H . 1977, T h e P r e f e r e n c e F o r Self-

Correction In The Organisation Of Repair In Conversation. Language,



SCHEGLOFF, E . A . & SACKS, H. 1973, Opening U p Closings. Semiotica, 7:289-327. (Reprinted in R. TURNER (ed.), Ethnomethodology, 233-64.) SCHENKIEN, J. N. 1971, Some Methodological And Substantive Issues In The Analysis Of Conversation. Unpub. Ph.D. dissertation, University Of California, Irvine. — 1978, Studies In the Organization Of Conversational Interaction. New York: Academic Press. SCHUTZ, A . 1962, Collected Papers: Vol. I, The Problem Of Social Reality, Vol. II, Studies in Social Theory, Vol. Ill, Studies in Phenomonological Philosophy. The Hague: M. Nijhoff. SEARLE J. R. 1984, Minds, Brains And Science. Cambridge, Mass.: Harvard University Press. SHARROCK, W. W. 1974, On Owning Knowledge, In R. TURNER (ed.), Ethnomethodology. Harmondsworth: Penguin Books Ltd., 45-53. — 1982, The Analysis Of Social Interaction Through The Use Of Audio-Visual Records. End Of Grant Report. Economic And Social Research Council: No. H R 5362. SHARROCK, W. W. & ANDERSON, R. 1986, The Ethnomethodologists. Chichester: Ellis Horwood. SOROKIN, P. 1955, Fads and Foibles in Modern Sociology. Chicago: Chicago University Press. SPEIER, M. 1969, The Organization of Talk and Socialization Practices in Family Household Interaction. Unpublished Ph.D. dissertation, Sociology Department, University of California, Berkeley. — 1973, How to Observe Face-to-Face Communication: A Sociological Introduction. Pacific Palisades, California: The Goodyear Publishing Co. SUDNOW, D . (ed.) 1972, Studies In Social Interaction. New York: Free Press. TURNER, R. (ed.) 1974, Ethnomethodology. Harmondsworth: Penguin Books Ltd. WATSON, D . R . 1974, Calls For Help: a Sociological Analysis OfTelephone Calls To A 'Crises Intervention Centre'. Unpub. Ph.D. thesis, Warwick University. — 1986, Doing the Organizations Work: An examination of a crisis intervention centre. In S. FISHER & A. DÜNAS TODD (eds), Discourse and Institutional Authority: Medicine, Education and Law. Norwood, N.J.: Ablex Publishing Corp. pp. 91-120. ZIMMERMAN, D. & POLLNER, M. 1970, The Everyday World As A Phenomenon. In J. DOUGLAS (ed.), Understanding Every-Day Life: Towards the Construction of the Sociology of Knowledge, 80-103.


Appreciation - and intimacy 163, 166, 169-70, 173, 176, 179-80, 182-4, 187, 191 - see also closings, sequence types Arrangements, see closings, sequence types Aspiration, in transcription 14 Assessment 176-7, 179-80, 186, 189-90 - access 182 - no-access 182-3 Atkinson, J. M. & Drew, P. 1 Atkinson, J. M.&Heritage, G. 1, 19n., 206n., 245n„ 264 Austin, J. L. 276n., 282 Azande 219-21

Accountability, natural 36, 37-8,258 Accountings 88, 90, 95-7, 99; see also corrections Acknowledgement token 126, 178, 180, 183n„ 190 Actions, relations between 32-5, 302, 308,310,315 Address terms 73,77, 102n„ 266-7, 269, 283n. Adjacency pairs 35, 36-7, 55-6, 61-2, 101, 104n„ 134n. - and critique of conversation analysis 304-5,307-9,311 Affiliation and intimacy 159-70, 173, 176, 180, 183, 186-91 Agreement 253, 257-8 - as general principle of social order 293 - preference for 5, 57-9, 61-6 Alternative, in consecutive reference 91-5, 98 Amplitude - increase in 155, 173-4 - reduction in 182 Analysis, constructive 266 Anderson, R. J. 6,7 Anderson, R. J., Hughes, J. A. & Sharrock, W. W. 244 Answers 35, 37, 56 & n„ 56-67, 307-8 - selection 57 - see also summons Apparatus 56n. Appositional, pre-placed 74, 80

Back-references, see closings, sequence types Beattie, J. 302-3 Beginnings - conversational 36-8, 245 - earliest possible 73-4 - recycled as repair technique 75-6, 78-80, 83-5 and turn transfer 46, 74-5 - sentence 76-7 - turn recycling 5, 70-85 and turn projection 71-4 Benson, D. J. & Hughes, J. A. 19n„ 20n. Blame-allocation, and possessive pronouns 280 328

INDEX Brown, G. & Yule, G. 292n„ 304-5, 306-10 Button, G. 5-6, 101-51 Button, G. & Casey, N. J. 38n., 112n. Button, G., Drew, P. & Heritage, S. 1 Byrne, P. S. & Long, B. E. 249n. Calendars, private 222-4 Calls, reason-for, see closings, sequence types Categorisation in sociology 28-9 Category organisation 256 & n. Category usage 246, 276n., 277, 279-80, 283 Characterisation 7, 29-31; see also descriptions Cicourel, A. 20n. Closing extension 104n., 126, 147 Closings 5-6, 101-49, 245 -archetype 102, 123, 126-7, 128, 132-5, 137, 140-1, 143-4, 148 - closing components 102, 125-7, 134, 147 floor passing 124, 126; see also movement in closings - components 102-4, 107-8, 110, 114-26, 128, 133 - e x t e n d e d 134, 137-41 -foreshortened 134, 135-7, 139-41 -initiation 101-2, 105-6, 115, 119, 121, 128-9, 135-6, 139, 145-7 and laughter 169 - optional insert 126 -reinitiation 107-12, 114, 116, 118-20, 121, 126-8, 139, 143-4 - sequence types 104-28 appreciations 109n., 122-8, 146-7 arrangements 104-9, 110-12, 114, 118, 121-2, 128, 139-40, 141, 143-5 back-references 109-12, 114, 118, 126, 128, 141-3 'in-conversation' objects 116-18, 128, 140, 142 reason-for-calls 109n„ 121-2, 128, 145-6 solicitudes 118-20, 121-2, 128, 144-5 topic initial elicitors 112-16, 117-18, 128, 142-3

329 - terminal components 102, 107 - see also movement Co-ordination 39-40, 54, 58, 307 - in closings 101, 134n. - of prior turn endings 75 - in turn taking 40-2, 71-2, 84 Coherence, conversational 30, 36, 230-1,276,310-11 Collaboration - in closings 117, 142-3 - in communicative event 22-3, 37-8,58, 99, 172, 254, 256-9,261,311,315 - in doctor-patient interaction 254, 256-9 Commonsense, and conversational organisation 34-5, 29-30, 38 Communication - non-verbal 300-2 - as occasioned and situated 261 - as orderly, see order - see also talk Compliments 56 Components, repetition of 5 Compromise 66 Conditionals, counterfactual 222 Confirmation request 83-4 Consequence clause 71 Constraint, see organisation Context - of conversation 4, 41-2, 247-9, 251, 304,312,316-17 - independence 42, 48, 248-9, 259 - organisational 4 - of pronoun use 269, 27 In., 283 - as self-explicating 22-3, 249, 252 - sensitivity 42, 248-9, 259 - of social phenomena 23-5, 33 - and turn taking 42 Contiguity - preference for 5, 58-60, 65 - in sequences 57-9 Contingency clause 71 Continuation, conversation 5-6, 102, 107, 115-18, 134, 142-4, 148 - after impropriety 171-2 Control, social 39, 42, 293 Conversation - as locally produced 247-8 - as natural activity 9, 32-3 - as orderly 245-9, 299, 301,3, 315, 317

330 - as self-explicating system 312-13, 315 - telephone 37, 76-7, 136, 170-90, 317 Conversation analysis - critiques of 47-51, 290-2, 296-7, 299-300, 304-5, 308-10, 315, 319 - as generative 46, 47 - limitations of 245-6 - as linguistics 2-4, 6, 19, 50-1, 292, 298, 304-5,321 - methodology 3, 47-52, 244-7, 249, 298-303,305-9,314-16,320 - and multilingual studies 2, 4, 51-2 - as sociology 1-5, 22-6, 31, 34, 42, 46-7, 49-51,321 critique of 292-3, 298, 300 discontinuity with 19-20 in everyday interaction 244-8, 255-6 - as system atic 292, 296, 301, 310-18 - see also discourse analysis; interpretation Correcting, and correction 95-7, 99 Correction 5, 87-8 - acceptance 88-9, 94-5, 97-8 - e m b e d d e d 86, 95-9 - exposed 86, 97-9 - rejection 88-90, 94-5, 97-9 - s e l f 86, 96, 97n. - solicitors 255-6 - see also repair Corsaro, W. A. 21 In. Coulter, Jeff 281 C o u l t h a r d , R. M. 292n. Counter-questions 56n. Court room, culture of 7, 227-42 Courtesy phrases 73, 77 Data - in conversation analysis 299-300, 305, 318 - typicalities as 264-5 Data collection, and sociology 3, 22, 24-5,247,315-16 Decontextualisation 284 Departure - as multi-party event 6, 212-15 - as single-party event 211-12 Description - discrepant 227, 228-41 - everyday 28-30

INDEX - as f u n d a m e n t a l to understanding 26, 27 - problems of 26-31 - s t y l e s o f 7, 227 - in transcription 16 Designators, pronouns as 267 Determinism, linguistic 285 Deviance in conversation 46 Directiveness, principle of 282 Disaffiliation and intimacy 160-1, 163, 170, 174-8, 182, 186, 190-1 Disagreements 58-9, 62-3, 64-7, 72 Disattention and intimacy 161, 163, 165-6, 168-79 Discourse analysis 6, 51, 292 & n., 303-5, 308,310-11,313-14 - critique of 318-20; see also conversation analysis Doctor-patient interaction 7, 249-59 Drew, P. 209 Due-point and overlapping talk 172-3, 176, 183 Economy, principle of 252, 255 Elias, Norbert 262-3, 265-6, 283, 285 Ervin-Tripp, S. 214n. Escalation of intimacy 162-3, 165, 169-70, 173-4, 187-91 Ethnomethodology - and conversation analysis 1-2, 22-6, 50, 245, 248 - and sociology 20n., 24, 293-4 Evans-Pritchard, E. E. 219-20 Exchanges, remedial, see repair Explanation, importance of in sociology 27 Expressions, indexical 23 Fact and opinion 7, 226-7, 229, 234-5, 240-1 G a p 45, 75, 80 Garfinkel, H. 2, 20n„ 22-6, 27-9,40, 248, 293n. Garfinkel, H., Lynch, M . & Livingston, E. 247n„ 248n. G a z e 213-14, 301,302 Gesture 302 G o f f m a n , E. 206, 210, 276 G o o d m a n , Nelson 222



G o o d w i n , C. 1,6,49, 206-16 G o o d w i n , M. 206n. Greetings 33, 36, 55-6, 73, 247, 302

Intradisciplinary studies 286 Introductions 73 Invitations 36

Harre, R. 27 'he', 'she', use of 264, 267, 276 Heath, C. 206n., 211 Heritage, J. 20n., 293n. Heydon, J. D. 233n.

Jefferson, G . 5, 7, 10, 38n„ 86-100, 206n„ 304 Jefferson, G . & Lee, J. R. E. 264 Jefferson, G „ Sacks, H. & Schegloff, E. A. 152-205 Jefferson, G „ Schegloff, E. A. & Sacks, H. 5-6, 43

'I' 282-3 - as excluding 269 - and excuse/apology 272 Identification - complete official 228, 229-32, 235, 239-40 - numbers/characterisations 228-9, 233-41,284 - relational 228, 230-2, 235, 241 Implicativeness, sequential 36, 37-8 Impropriety as mark of intimacy 160-91 Improvisation 35, 38 Inbreaths, as laugh-terminal objects 155, 156n„ 157-8, 187-8 Inhalation, in transcription 14 Insults 56n. Interaction - in communicative event 261, 264 use of p r o n o u n s 266-70, 277, 279-80, 282-4 - moral dimension 276 - non-verbal c o m p o n e n t 300 - organisation of 2-3, 32-5, 38, 54-5, 206, 249-50, 255 Interdisciplinary studies 4, 261, 285-6, 319-21 Interpretation and conversation analysis 43-4, 49, 2 8 5 - 6 , 3 1 2 , 3 1 7 Interruptions 74n„ 78-80 Intersubjectivity 31, 38 Intervals, t i m e d / u n t i m e d 12 Intimacy 6, 160, 170, 187 - invitation to 160-2, 165-7, 169-70, 173-4, 179-80, 183-90 - r e c i p i e n t ' s response to 160-3, 165-9, 171-6, 179, 183-91 Intonation - and laughter 155, 158 - in transcription 13 - and turn extension 77-8

Katz, B. 254n. Kickers 1 7 5 & n „ 182 Knowledge - and opinion 218 - ownership of 278 Labov, W . & Fanshell, D. 214, 292n. Language - natural, in use 2-4, 50-1 - as social institution 304 Laughter 6, 43, 120, 152-205 - a n t i c i p a t o r y 172-3, 182, 187 - in expanded affiliative sequence 159, 161-70, 172-6, 179-90 - extension of 158-9, 167-9, 190 - as pre-affiliative 168 - as socially organised activity 152, 154-9, 167, 170 - termination of 168-9, 169, 187-8 - in transcription 14-15, 152 Lee, David 17In. Lee, J. R. E. 3, 5, 6, 19-53 Lempert, R. O. & Saltzburg, S. A. 226, 23In., 233n. Levinson, S. C. 50n. Linguistics - and critique of conversation analysis 292 - grammatical models in 51, 261, 263 - and interdisciplinary studies 285-6 - see also conversation analysis Lowe, Ivan 263-4, 265-6 Lynch, et al 2 MacKay, R. W. 276n. Markers - disjunct 258-9 - excusing 126-7 - interruption 72, 81, 82-3

332 - misplacement 72-3, 81 - uncertainty 233 Maynard, D. 175n. Mead, G . H. 263 Medicine, bureaucratic 251-6 Mehan, H. 264 M e m b e r s h i p and circumstance, analysis of 277n., 280 Misunderstandings 66, 312 M o v e m e n t in closings - d r a s t i c 111-16, 118, 128, 142-3, 147-8 - minimal 109-11, 114-16, 119-20, 122, 126-8, 143-4, 147-8 - opportunity space for 128-41 after first possible close c o m p o n e n t 128-30, 133, 135, 141-5, 147-8 after first terminal 131-2, 133-4, 138-9, 147-8 after second close c o m p o n e n t 130-1, 133, 135, 147-8 Multi-disciplinary approach 4, 6; see also interdisciplinary studies Names, proper, and pronouns 267 Negotiation 35, 38, 148 - in laughter sequence 168, 180 Nixon, Richard M. 282-3 'non-person', concept of 276n., 284 'normals', and possessive pronouns 276 &n. Objects, 'in-conversation', see closings, sequence types Obscenity as mark of intimacy 160, 163-5, 185, 187, 190 Occupations, and organisational environment 249-52, 259 Offers 56, 308-9 Openings, see beginnings Operationalism 24, 28 Opinion - and facts 226-7, 234-5 - and knowledge 218 O p p o r t u n i s m , principle of 252, 255, 259 Opportunity cost, and choice of approach 303 Opportunity space, see movement in closings

INDEX Order - of everyday life 298-9 - as external rule-based 22-3 - and organisation 6-7, 20-1, 24-5, 39, 245-8, 258-9,311 - as self-explicating 22-3, 256 - social and conversation 38, 39-43, 261, 264, 311 general principles of 293-6 - see also organisation Organisation - constraints of 249, 252, 254, 271 - local/turn-by-turn 36-7, 39-42, 45, 55-67 - oriented-to 34-5, 36, 38, 44-6, 74-5, 245-9, 256, 258-9, 305-6,312 in non-talk interactions 246-7, 250 - overall 36-7 - in paediatric clinic 254-8 - second-order 46 - social, of language in use 2 - 3 - of topic 36, 38 Overlap 5, 10-11 - absorption techniques 80-3 - as deviant 46 - see also utterances, overlapping Overspill in closings 105-8, 118 Paediatric clinic - as everyday environment 249-58 - transcripts of 249-50, 253 Pair parts 33, 55-67, 302; see also adjacency pairs Participant analysis 33, 43-4, 46, 305-6, 315 Partition - by sexual status 266, 27 In. - of person and n u m b e r reference 262, 264, 266, 270, 276n„ 281-4 Pike, K. L. 6 , 5 1 , 2 6 3 , 265-6 Pitch - increase in 155, 173 - in transcription 15 Pollner, M. 312n. Pomerantz, A. 7, 49, 186n., 213n., 226-43 Power as general principle of social order 293 Preference system in closings 115-16

INDEX Pro-verb 275-6 Problem as a possessable 276-81 Pronouns 6, 175, 183 - choice of 171, 261-2, 264-7, 267-8, 277, 281,284-5 - as c o m p l e m e n t a r y 263, 266 - and distinct perspectives 262, 263 - as excluders 281-2 - grammatical model 261-83 as heuristic device 261, 264 interdisciplinary work 261, 265, 285-6 and omnirelevance 265-6 - as includers 281-2 - and intra-disciplinary studies 276 - linguistic analysis 284-5 - as mutually exclusive alternatives 262-3, 267 - and organisational reference 266, 270-5, 284 - possessive affiliative use 276-7, 279 combinatorial reference 278-9 and ownership, 276, 277-9 and relational pairing 278-80 - and prospective circumstantiality 274-5 - and provisional relevant 274-5 Proposal 253, 257 Proterms - in consecutive reference 90-1, 93 - as interactional objects 6, 91, 275, 277, 283-4 - prospective repair 91 Psathas, G . 1 Psychology, and conversational organisation 50 Question projection 72 Questions 36, 37, 56-67, 308-9 - alternatives in 63 - multiple 59-60 - see also answers, counter questions Quotation formats 72 Recipient design 312, 316-17 Recognition-placement 173n. Recycling, of conversational decline 116; see also beginnings, conversational

333 Reference, consecutive, to same object 90-3, 94-5 Rejection of proposal 257 Relations, social, and analysis of talk 317-18 Relevance, transition 4 0 - 1 , 4 5 , 101, 131, 133-5 Repair - conversational 35, 86, 90, 94-6, 245, 247, 255 - recycling as 75; see also beginnings, recycled, as repair technique - systematic mechanisms 75 Repetition - in consecutive reference 91-2, 93 - and overlap 76, 84; see also utterances, overlapping Request 214 Response selection 56-8 Role allocation 35, 37-8 Rudeness as mark of intimacy 160, 164-5

Sacks, H. 5, 6, 7, 19n„ 30-1, 33, 54-69, 86n„ 175n„ 177n„ 178 - on adjacency pairs 104n., 304 - on characterisations 233n. - on conversation analysis 264-5, 302 - on conversational organisation 25-7, 33, 36n„ 38, 49, 206n., 213n„ 276n„ 307 - on description 29-31 - on interpretation 43 - on organisational reference 266n. - on personal pronoun use 267, 269n., 270, 274n„ 276, 282, 283n. - on personal relevance 7, 217-25 - on possessability 276n. - on second story 45, 181n. - on third-person speech 283-4 Sacks, H., Schegloff, E. A. & Jefferson, G . 20, 39-40, 45, 206n„ 248 Sapir-Whorfe hypothesis 285 Schegloff, E. A. 5-7, 37 & n., 46, 70-85, 175n„ 269, 304 - on m e m b e r s h i p and circumstance 277n. Schegloff, E. A. & Sacks, H. 31, 36n. - on conversational closings 101, 211

334 Schegloff, E. A., Sacks, H. & Jefferson, G. 13 In. Schenkein, J. N. 1, 8, 254n. Schneider, 276n. Schultz, 28-9 Self talk - embeddedness of 6,208-10, 213-14 - as interactive 6, 206, 212-15 - as non-interactive 206 - orientation to 210-15 - understanding of 208-9 Self-quotation 276 Sequence, basic 55-6 Sequence types - in closings 104-48 - in opportunity space 141-8 Sequencing in conversation 47, 206, 307-8,310-11,313-15,318 Sequential analysis 5-6, 49, 54-67 - in doctor-patient interaction 7, 249-59 Setting, see context Sharrock, W. W. 3, 6, 7, 249n„ 278 Sharrock, W. W.& Anderson, R. 19n„ 20n., 49, 244-60, 290-321 Silence 64, 171-2, 186, 190,211 Sims, Joan 47In. Skip-connecting 177, 187n. Sociology - and commonsense 24-5, 29-30 - and general principle of social order 293-5 - and interdisciplinary studies 285-6 - and interpretation 43-4 - interpretive, possibility of 297-8, 299 - and methodology 20, 22, 24, 27-9,46, 244 - and nature of social phenomena 22-3, 26-7,31,247, 259, 295-6, 298-9 - and personal pronouns 262-3, 265, 285-6; see also data collection Solicitudes, see closings, sequence types Sorokin, P. 49n. Sound, in transciption 13-15 Sounds, non-speech, in conversation 6, 152-6; see also laughter Speaker - change of 39-40, 71-3, 75-6, 81-3

INDEX - numbers 41-2 - selection of 45, 267 - status 41-2 Speech, third person 283 Speech exchange systems, see conversation; talk Speed of utterance - in transcription 14 - and turn extension 77-8 Speier, 264, 276n„ 281-2 Stipulation, decontextualised 266 Stories, second 45 Story, second 45, 181n„ 186, 189 Story telling 30, 117, 245, 247, 307 Stress, in transcription 15 Stretchings, word 78-9, 173, 176-7 Sudnow, D. 1 Suicide, reasons for 7, 217, 18, 221, 224 Summons 35-6; see also answers Systematics, simplest 20, 39-40, 42, 45-6 Talk - institutional nexus of 244-5 - non-interactive, see self talk - and non-verbal components of interaction 300-1 - as orderly 2-5, 25, 299, 303, 315-17 - triggered off 171 - see also communication; conversation; interaction Tape-recordings, use of 9,25, 31, 299-300,315-16 Termination, conversation, 102, 104, 106, 135-40, 144-8; see also closings 'they' - as excluder 282 - organisational reference 270-5 Topic - bounding 101 - as constraint on conversation 36, 42 - development in closings 11-12 - initial elicitors, see closings, sequence types - initiation 178 - introducing 35, 38 - last 104-5 - organisation 245 - in question and answer pairs 61 - s h i f t in 174, 184, 189


INDEX Transcript - symbols 9-17 - use of 25-6, 31, 244, 246-8, 299-300, 316-17 Transformation, pronominal 264, 266, 267-9 Truth, and description 29-30 Turn - beginnings, see beginnings, turn - extending 77-9, 145 - identification of 306 - turn-constructional c o m p o n e n t s 40, 59-63 Turn taking 5, 35, 37-42, 71, 206, 304-5, 313-15, 321 - completion point 77-9 - constraints on 36, 104n., 134, 147 - rules for 40-1, 43, 45-6, 75, 177, 307 - transition point 77 Turn types, projection of 71-2 Turner, Roy 252n., 276 Understanding - achievement of in conversation 22, 23,34, 4 3 - 4 , 3 1 2 - and recycling of beginning 82-3 Utterances - as contextually e m b e d d e d 33-4, 305 - continuous 11-12 - monitoring of by participants 44 - overlapping 10-11, 42-3, 45-6, 73-84, 172, 182-3 m a n a g e m e n t of 176, 177, 1 8 0 & n . and self talk 207

- performative, conventional construction 272, 274-6, 281, 283 - sequential placement 33-5 - simultaneous 10, 75-6 - as social actions 310-12

Validity, problem of, and sociology 24-5, 27-8 Video recordings, in analysis of talk 249, 300 Volume, and self talk 213

Watson, D. R. 6, 7 , 5 1 , 2 6 1 - 8 9 'we' - as excluder 264, 281-2 - organisational reference 267-72, 274n„ 281-2 Weber, Max 27 Work tasks 7, 249-52, 259 - constraints on 250 - as locally managed 250, 252-3, 254, 256-7

'you' - as address term 266-8, 278 - as 'anyone' 268-70, 272 'your', address function 278-9 Z i m m e r m a n , D. & Pollner, M. 20n.

Lightning Source UK Ltd Milton Keynes UK UKHW022044251122 412793UK00013B/782