Reporting talk cách nói về chủ đề báo cáo trong tiếng anh

Ngày nay trong các công sở, yêu cầu đối với các nhân viên là phải biết cách soạn một báo cáo chuyên nghiệp, để làm việc đó cần thiết mỗi người cần trang bị cho mình các kiến thức liên quan, phải thảo luận và xin ý kiến đóng góp cho nội dung báo cáo hoàn thiện hơn. Đây là cuốn sách sẽ mang lại tự tin trong quá trình soạn các báo cáo bằng tiếng anh cho các bạn.

Reported speech, whereby we quote the words of others, is used inmany different types of interaction In this revealing study, a team

of leading experts explores how reported speech is designed, theactions it is used to perform and how it fits into the environments inwhich it is used Using the most recent techniques of conversationanalysis, the authors show how speech is reported in a wide range

of contexts – including ordinary conversation, story-telling, newsinterviews, courtroom trials and medium–sitter interactions Pro-viding detailed analyses of reported speech in naturally occurringtalk, the authors examine existing linguistic and sociological studies,and offer some pioneering insights into the phenomenon Bringingtogether work from the most recent investigations in conversationanalysis, this book will be invaluable to all those interested in thestudy of interaction, in particular how we report the speech ofothers, and the different forms this can take

E L I Z A B E T H H O L T is Senior Lecturer in English at HuddersfieldUniversity She has contributed to the journals Research on Lan-guage and Social Interaction, Text, Social Problems and Language

in Society

R E B E C C A C L I F T is Lecturer in Linguistics at the University ofEssex She has contributed to the journals Language, Language inSociety, Journal of Sociolinguistics and Lingua

E D I T O R S

Paul Drew, Marjorie Harness Goodwin, John J Gumperz, DeborahSchiffrin

1 Discourse strategies John J Gumperz

2 Language and social identity John J Gumperz

3 The social construction of literacy edited by Jenny

Cook-Gumperz

4 Politeness: some universals in language usage PenelopeBrown and Stephen C Levinson

5 Discourse markers Deborah Schiffrin

6 Talking voices: repetition, dialogue, and imagery in

conversational discourse Deborah Tannen

7 Conducting interaction: patterns of behaviour in focusedencounters Adam Kendon

8 Talk at work: interaction in institutional settings Paul Drewand John Heritage

9 Grammar in interaction: adverbial clauses in AmericanEnglish conversations Cecilia E Ford

10 Crosstalk and culture in Sino-American communicationLinda W L Young

11 AIDS counselling: institutional interaction and clinicalpractice Anssi Pera¨kyla¨

12 Prosody in conversation: interactional studies edited byElizabeth Couper-Kuhlen and Margret Selting

13 Interaction and grammar edited by Elinor Ochs, Emanuel

A Schegloff and Sandra A Thompson

14 Credibility in court: communicative practices in the CamorraTrials Marco Jacquemet

15 Interaction and the development of mind A J Wootton

16 The news interview: journalists and public figures on the airSteven Clayman and John Heritage

17 Gender and politeness Sara Mills

18 Laughter in interaction Philip Glenn

19 Matters of opinion: talking about public issues Greg Myers

care physicians and patients edited by John Heritage andDouglas Maynard

21 In other words: variation in reference and narrative DeborahSchiffrin

22 Language in late modernity: interaction in an urban school BenRampton

23 Discourse and identity edited by Anna De Fina, DeborahSchiffrin and Michael Bamberg

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

First published in print format

ISBN-13 978-0-521-82483-5

ISBN-13 978-0-511-33382-8

© Elizabeth Holt and Rebecca Clift 2007

2006

Information on this title: www.cambridge.org/9780521824835

This publication is in copyright Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.

ISBN-10 0-511-33382-X

ISBN-10 0-521-82483-4

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org

hardback

eBook (EBL) eBook (EBL) hardback

List of contributors pageix

1 Introd uction

2 Interact ive Footing

3 ‘I’m eyeing you r cho p up mind’: report ing and

7 Designin g con texts for report ing tact ical talk

8 Active voicing in court

9 Sp eaking on beh alf of the public in bro adcast

PR O F E S S O R ST E V E N

CL AY M A N

Department of Sociology, University ofCalifornia, Los Angeles, USA

DRRE B E C C A CL I F T Department of Language and Linguistics,

University of Essex, Colchester, UK

JO A N N E KE R B Y Formerly School of Human and Life

Sciences, Roehampton University, London

DRJO H N RA E School of Human and Life Sciences,

Roehampton University, London

DRRO B I N

WO O F F I T T

Department of Sociology, University ofYork, UK

We are deeply indebted to Paul Drew for encouragement at all stages

of the production of this volume and for helpful comments on the finaldraft Makoto Hayashi also provided generous intellectual supportwith his wide-ranging knowledge of work on reported speech Wealso thank Andrew Winnard, Helen Barton, Jayne Aldhouse and allthose at Cambridge University Press for their careful shepherding ofthe manuscript through the presses

The method of transcription used within this volume was veloped by Gail Jefferson (but see the Appendix to Chapter 6 forsome additional notations used within that chapter.) The systemattempts to capture some of the features of the interaction relevant

de-to its organisation, including turns, overlapping talk, pauses, andintonational features such as emphasis and marked rises and falls inintonation

This summary of transcription notations relies heavily onAtkinson and Heritage (1984) and ten Have and Psathas (1995).Simultaneous turns

Where turns are begun simultaneously they are marked by a singleleft-hand bracket at the start of the turns:

Overlapping turns

When turns do not start simultaneously, the point where the newturn begins is marked by a left-hand bracket within the first turnand at the start of the new turn:

Where overlapping talk stops, it is marked by right-handbrackets:

] Tom: I used to smoke [a lot] more than this

Contiguous turns

When turns are latched (i.e there is no interval between the end ofone turn and the beginning of a next) equals signs are used at theend of the first turn and the beginning of the second:

Equals signs are also used where a transcriber has put on to anew line elements of a turn that form part of the continuous flow ofspeech often due to intervening overlapping talk:

When overlapping turns end simultaneously and a subsequent turn

is begun without an interval, the overlapping turns are followed byright-hand brackets and equal signs The new turn is marked by aleft-hand bracket at the beginning:

Intervals within and between turns

Intervals in talk are timed to the tenth of a second and marked bynumbers in parentheses either within a turn at talk:

(0.0) Lil: When I was (0.6) oh nine or ten

An interval of less than, or around, one tenth of a second is marked

by a period within parentheses:

Characteristics of speech delivery

Punctuation is used to convey characteristics of speech deliveryrather than grammatical units

Sound stretch

A colon indicates a stretched sound:

Longer stretches of sound are indicated by multiple colons:

::: Tim: I’m so::: sorry re:::ally I am

A comma indicates a continuing intonation:

A question mark indicates a rise in intonation:

? V: A do:g? enna cat is different

P: Yih ever take’er out again?

Marked rises or falls in intonation are marked by upward- ordownward-pointing arrows immediately before the shift:

"# L: AND uh "we were looking rou-nd the #sta:lls ’npoking about

Less marked rises or falls in intonation along with some stretchingcan be marked by underlining immediately preceding a colon for

a fall:

a: L: we (.) really didn’t have a lot’v cha:nge

and underlining of a colon for a rise:

a: P: I’m (h) eyeing your cho:p up mi:nd

An exclamation mark indicates an animated tone:

Emphasis

Underscoring indicates emphasis:

Still L: he said oh: hhello Lesley (.) "still trying to buy

Increased volume is marked by the use of upper case letters:

Decreases in volume are marked with a degree symbol at thestart and end of the quiet talk:

J: I’ll go ahea:d, and, hh pay for it when it comes

and
he’ll never kno:w
,

Aspiration

A series of ‘h’s preceded by a dot indicates an in-breath:

Out-breath is indicated by a series of ‘h’s without a preceding dot

Laughter and smile voice

Laughter is indicated by an attempt to convey the sound using ‘h’sand vowels, while laughter particles within words are indicated by

an ‘h’ or ‘h’s in parenthesis:

eh D: "UH you fin:nished with that ch"o::p p#et eehh he he

he [he he he

(h) P: [I’m (h) eyeing your cho:p up [mi:nd

Talk which is noticeably quicker than surrounding talk is marked

by ‘more than’ and ‘less than’ symbols either side of the fast talk:

> < P: >In fact d’yuh think they will< enjo:y co:mpany.Indecipherable sounds

Where a transcriber is unable to make out a sound or a series ofsounds, spaces within parenthesis are used to convey the approxi-mate extent of the missing talk:

D: huh hee it’d [bust

Where the transcriber is able to guess at the words or soundsused, these are included within parentheses:

(in) J: "Santa Claus brou:ght it (in his sle::d)

Where the transcriber is unsure of who made a sound, emptyparenthesis may be used instead of a name or initial:

Verbal descriptions

Double parenthesis is used to enclose a description of the talk orsome other phenomenon present during the interaction that thetranscriber does not want to convey by attempting to representthe sound For example, in the following extract an ‘f’ in doubleparenthesis marks falsetto intonation:

! Vera:! Well I said tuh Jean how abou:t it

Horizontal ellipses

These indicate that parts of the same turn are omitted:

D: as I say yu- yuh go down steep

Vertical ellipses

These indicate that intervening turns have been omitted:

Mum: "oh: dea:r Thursday

.Mum: Oh: they make you pay f’r putting it on again too:

Numbering of lines or turns

This is done for convenience of reference Intervals within talk arealso numbered

Rebecca Clift and Elizabeth Holt

(I)n real life people talk most of all about what others talk about – they transmit, recall, weigh and pass judgement on other people’s words, opinions, assertions, information; people are upset by others’ words, or agree with them, contest them, refer to them and so forth.

(Bakhtin, 1981 : 338)

1.1 Introduction

This volume is an investigation of reported speech in naturally ring spoken interaction We recurrently use talk to report talk,whether we are reporting the compliment someone gave us orconveying how we made a complaint or told a joke In the followingextract, for example, the speaker uses reported speech as part of astory relating how she was the victim of a nasty put-down (arrowed):1

occur-(1) [Holt: C85: 4: 2–3] (Lesley has been looking around the stalls at achurch fair)

Lesley’s animation of the man’s words is the culmination ofher reporting of a series of actions It is this phenomenon – thereproduction of prior talk in a current interaction – that the studies

1 For a key to transcription symbols, see pages xi–xvii

in this volume are concerned with Together they bear witness tothe use of reported speech and its variant forms across the range ofinteractional contexts from ordinary conversation to so-called insti-tutional talk such as political interviews and debates While en-gaging with material as diverse as story-telling, witness testimony

in court, interaction between spiritual mediums and their sittersand video data of an aphasic man, the chapters have a central focus:the design and placement of reported speech – and thought – insequences of interaction Aspects of design include its lexical andprosodic construction; issues of placement relate to how turns inreported speech are built to follow particular others, and the re-sponses that they in turn generate In the extract above, for exam-ple, Lesley introduces reported speech as the climax of the storyshe has been telling; story climaxes, as we shall see, are one of therecurrent interactional sites for reported speech The design andsequential placement of reported speech thus display systemati-cities which are only available by close analytic attention to severalinstances of the same phenomenon; the chapters in this volumeare characterised by a commitment to such analytic attention

A more detailed survey of the contents follows in due course, butfirst we sketch the background to existing work on reported speechand the main theoretical issues to have emerged from it As we shallsee, the relatively recent advent of interactionally grounded studies

of reported speech has promised to illuminate many of the theoreticalissues formerly regarded as intractable The rationale for adoptingthe rigorously empirical approach of conversation analysis is dulyset out here, followed by some of the earlier findings from conver-sation analytic work on reported speech; it is in this work that thecurrent contributions have their origins

1.2 Background and main themes

Work on reported speech in recent years has emerged from a variety

of disciplines, most prominently literary theory, philosophy, tics and sociology.2 The proposal of the Bakhtin/Volosinov circle

linguis-2 For a comprehensive bibliography of work on reported speech, seeGu¨ldemann et al (2002)

that much of what we say is permeated with the voices of others hasproven highly influential beyond the domain of literary theory;much subsequent empirical work has pursued Bakhtin’s notion of

‘polyphony’ and his claim that any utterance contains ‘the concealed or completely concealed words of others’ (1981: 92).Within philosophy, reported speech has been of interest in itsreflexive capacity (D Davidson,1968–9,1984; Quine,1960) and

half-in this respect converges with work on metapragmatics withhalf-inlinguistics (see, for example, the collection in Lucy,1993), whichhas its origins in Jakobson’s concern with reported speech as ‘aspeech within speech, a message within a message ’ (1971: 130) It

is the work in linguistics that has produced the most diverse range

of perspectives Across this diversity it is nonetheless possible toidentify three central concerns in the literature: that with forms ofreported speech; with its authenticity, and with what it does Whileall three, as we shall see, continue to be the focus of ongoingresearch, it is evident that the concern with forms of reportedspeech generally predated work on its authenticity, and it is only

in relatively recent years that research has focused on what reportedspeech does in interaction This latter focus marks the increasinginfluence on linguistic research of work in sociology, and it is at theintersection of these two domains that much conversation analyticwork on reported speech has emerged and where the current studyhas its starting point To chart the route to this point, we nowbriefly sketch the three main preoccupations of previous work inreported speech

1.2.1 Forms of reported speech

Of structural linguistic studies, a major focus has been the tion between so-called direct reported speech (DRS) and so-calledindirect reported speech (IRS) Jespersen proposed that:

distinc-When one wishes to report what someone else says or has said (thinks orhas thought) – or what one has said or thought oneself on some previousoccasion – two ways are open to one Either one gives, or purports to give,the exact words of the speaker (or writer): direct speech Or else one adaptsthe words according to the circumstances in which they are now quoted:indirect speech (oratio obliqua) (1924: 290)

On Jespersen’s account, extract (1) – cited above – shows an tiation of the former; the extract below, in which a speaker

instan-is summarinstan-ising or conveying the ginstan-ist of a previous thought orlocution, is an example of the latter:

Jenny here conveys what Ivan purportedly said without claimingfidelity to his original utterance, the presence of the pronoun ‘I’clearly indicating that Jenny is speaking from her perspective.Besides this proposed distinction in the linguistics literature be-tween DRS and IRS, more recent work has focused on what hascome to be known as either ‘free indirect’ or ‘quasi-direct’ speech(Coulmas,1986; Banfield, 1973,1982; for a survey, see McHale,

1978), an amalgam of direct and indirect reported speech:

(3) [NB: II: 2: 10]

The majority of Nancy’s report here is indirect: the pronouns arefrom the point of view of the current speaker, not the originalspeaker However, ‘en eVOIder’ (line 6) appears to be directlyreported Elements of the last part of the reported speech – ‘willc’ntinue t’remember th’class en gro: w from it’ (lines 8–9) – appearalso to be directly reported

Much linguistic research has been grounded in this proposedthree-way distinction between direct, indirect and quasi-directspeech Thus Li (1986) provides a detailed characterisation of thedifferences between direct and indirect reported speech in lexico-syntactic and prosodic terms; Banfield (1973), Partee (1973),

Mayes (1990) and Longacre (1985) have also compared direct andindirect reported speech with respect to their distinctive structuralcharacteristics Of more functionally oriented research, Coulmasclaims that, while IRS is related from the current speaker’s point ofview (see also Leech and Short,1981), DRS:

is not the reporter’s speech, but remains the reported speaker’s speechwhose role is played by the reporter (1986: 2)

And according to Li (1986), DRS is used to convey both the formand content of the reported utterance, including gestures and facialexpressions In IRS, however, the speaker has the option of com-municating a comment on the utterance as it is uttered Thus, if theutterance is reported in an angry voice, in direct form the anger will

be heard as the reported speaker’s, and in indirect form it will beheard as the current speaker’s comment on the utterance

The concern with different forms of reported speech has led tolively interest in its introductory components, sometimes called

‘quotatives’ (Mathis and Yule,1994), most commonly in English –

as in extract (1) – a pronoun and a verbum dicendi such as ‘say’.Such quotatives may be present in what is identifiably both DRSand IRS, although in English one common characteristic of indir-ect reports is that the quotative is followed by the complementiser

‘that’ (Li,1986).3However, while variants of pronounþ say may

be considered the paradigmatic introductory component of ported speech, research has identified a number of alternatives

re-So Tannen’s (1989) survey of quotatives includes ‘tell’, ‘go’ and

‘like’ The apparently increasing use of beþ like as an tory component has been the focus of recent attention by Blyth

introduc-et al (1990), Romaine and Lange (1991), Ferrara and Bell (1995),Tagliamonte and Hudson (1999), Macaulay (2001) and Cukor-Avila (2002) The claim by Romaine and Lange that ‘like’ blursthe boundary between DRS, IRS and reported thought, claimingless commitment to the original than ‘say’ does, touches on thesecond of the three main concerns in the linguistics literature in thisdomain: the authenticity of reported speech

3 See Haakana (this volume) for Finnish as a contrast case

1.2.2 The authenticity of reported speech

Research into reported speech began with the assumption (derivedfrom the lay assumption (see Mayes, 1990: 330–31)) that directspeech is more accurate than indirect speech Thus, Bally (1914)viewed DRS as ‘a phonographic reproduction of the thoughts andwords’ of the original speaker (quoted in Clark and Gerrig,1990:795) But more recent work has shown how DRS is, in fact, rarely

an accurate rendition of a former locution Volosinov (1971) wasthe first to criticise the assumption that reported speech is anauthentic rendition of the original, proposing that the meaning ofthe original utterance is inevitably altered in the reporting context(see Dubois (1989) on what she calls ‘pseudoquotation’, and Stern-berg (1982) on claims regarding the reframing of reported speech).This claim has been supported by psycholinguistic research ThusLehrer (1989) shows that, in experiments to test the memory ofprose, subjects tend to remember the meaning of utterances ratherthan the form, and that verbatim recall is unusual Mayes (1990:331) investigated the authenticity of the reported speech in hercorpus and claimed that at least 50 per cent were inventions bythe current speaker Included in her collection, along with ‘plausiblequotes’ and ‘improbable quotes’ (for example, a speaker reporting

an utterance made twenty years earlier), were ‘highly improbablequotes’ (such as a ‘Greek chorus’ where a quote is attributed tomore than one person) and ‘impossible quotes’ (including hypo-thetical quotations) Thus it would seem that the term ‘reportedspeech’ is somewhat of a misnomer;4 as we shall see, one of theconcerns of this volume will be to engage with the reasons for this.1.2.3 What does reported speech do?

While early linguistic studies of reported speech were ingly concerned with structural questions for which the use of con-structed exemplars or literary texts was perceived to be adequate,the past twenty years have seen an increasing number of empiricalstudies of reported speech In part this is due to a convergence of

overwhelm-4 Tannen (1989) goes so far as to adopt the term ‘constructed dialogue’ forthese reasons

structural and comparative linguistic concerns: many languagesgrammaticalise quotative constructions (see, for example, Cohen

et al (2002) on a range of East African languages, and the tion in Aikhenvald and Dixon (2003), and there has been keeninterest amongst typologists in this grammatical encoding ofreported speech (see the collections in Lucy (1993) and Gu¨ldemannand von Roncador (2002)) This move away from literary andtextual materials towards naturalistic speech data in a variety oflanguages has also engendered an increasing interest in functionaland pragmatic aspects of reported speech So comparative linguisticstudies, grounded in the ethnographic tradition, have investigatedaspects of reported speech in the languages of North America (see,e.g Collins, 1987; Moore, 1993; Urban, 1993), Austronesia(see, e.g Besnier,1993; Parmentier,1993; McGregor,1994), Southand Central America (see, e.g Adelaar,1990; Basso,1986; Shoaps,

collec-2004) and Africa (see, e.g Aaron,1992; Clements,1975)

Of linguistic studies concerned with the generic properties ofreported speech, many have remarked on the dramaturgical quality

of DRS in particular (see Li, 1986; Tannen, 1989; Wierzbicka,

1974) It has been proposed that reported speech is used in storiesnot only to replay an interaction but also to enable the speaker tosimultaneously convey his or her attitude towards the reportedutterance Labov (1972) distinguishes between ‘external evalu-ation’, where the point of a story is explicitly explained, and

‘internal evaluation’ where it is conveyed through the story itself.DRS is, he argues, a means of internally evaluating the story and istherefore more effective because it allows the recipient to draw his

or her own conclusions about the characters and events recounted.Mayes (1990) notes how reported speech is often used at the climax

of stories and proposes this as an effective way of conveying thepoint of a narrative.5

Much research in recent years, aiming to pursue the interactionalmotivations for the use of reported speech, has shown the influence

of the sociologist Erving Goffman’s observations on social action Goffman noted that:

inter-5 The association between reported speech and the climax or punchline ofstories is not restricted to English (see, for example, Polanyi,1982; Li,

1986; Larson,1987)

In daily life the individual ordinarily speaks for himself, speaks, as it were,

in his ‘own’ character However, when one examines speech, especially theinformal variety, this traditional view proves inadequate When a speakeremploys conventional brackets to warn us that what he is saying is meant

to be taken in jest, or as mere repeating of words by someone else, then it isclear that he means to stand in a relation of reduced personal responsibilityfor what he is saying He splits himself off from the content of the words byexpressing that their speaker is not he himself or not he himself in a seriousway (1974/1986: 512)

In observing that reported speech is an intrinsic feature of the way

we interact, Goffman echoes Bakhtin; but Goffman subsequentlyproposed that reported speech is a natural upshot of a more generalphenomenon in interaction: shifts of ‘footing’, defined as ‘the align-ment of an individual to a particular utterance ’ (1981: 227).Goffman is concerned to break down the roles of speaker andhearer into their constituent parts The speaker subsumes the roles

of ‘animator’ – ‘the sounding box’, the ‘author’ – ‘the agent whoscripts the lines’ and the ‘principal’ – ‘the party to whose positionthe words attest’ (1981: 226) All three roles may be played by aspeaker at the same time, but often they are not For instance, thevice-president reading out the speech on behalf of the president isonly the animator The author may be the president in conjunctionwith a scriptwriter The principal is the president, as well as therepresented political party she represents In reporting the speech ofanother person the speaker is the animator but not the author orprincipal Thus, our ability to use reported speech stems from thefact that we can adopt different roles within the ‘productionformat’, and it is one of the many ways in which we constantlychange footing as we interact (see Levinson (1988) for an elabor-ation of Goffman’s proposal)

The ‘reduced personal responsibility’ that Goffman claims forreported speech therefore appears to account for much of thelicence that speakers seem to take in using it; thus, Goffman( 1981 ) notes how curses and taboo utte rances may be used withgreater freedom than if speakers are speaking ‘in their own voice’.Goffman’s work has proven foundational in the investigation ofreported speech in interaction because it recognises that as much is

to be learned from examining the context of reported speech – andthe switch from non-reported to reported speech – as examining (as

many structural studies had) just the reported speech itself WhileGoffman is not in his own work concerned with the analysis ofactual instances of interaction (for a critique, see Schegloff,1988),

it provides a framework for researchers concerned with ing reported speech in its most basic environment of occurrence:ordinary conversation Before examining some of the products ofthis research, we provide a brief sketch of some of the basic tenets

investigat-of conversation analysis

1.3 Conversation Analysis: a brief sketch

Conversation analysis (CA) – the adopted name for what is perhapsmore accurately termed the study of talk-in-interaction – takes as abasic tenet the fact that social interaction is not haphazard butorderly, and that the methodical, organised nature of our social lifecan be studied by close attention to naturally occurring materials(for more detailed explication of the methods of CA, see Atkinsonand Heritage, 1984; Heritage, 1984a, Chapter 8; and Psathas,

1995) The transcription of these audio- or video-recorded ials according to the system devised by Gail Jefferson (see ‘Tran-scription conventions’ on pages xi–xvii) involves registering features

mater-of the production and articulation mater-of talk – and its absence – whichcapture the temporal unfurling of turns-at-talk So features such asoverlapping talk, in-breaths, the infiltration of laughter into talk,aspects of pace and prosody – all elusive to memory or intuition –are captured in the transcript and so accessible for their possibleinteractional import These transcriptions then make the data avail-able for repeated inspection and analysis This has two importantconsequences: it allows for methodological transparency, such thatthe presence of the data makes any analysis accountable to it, anddisputable because of it; and it also enables the collection of mul-tiple examples of the same phenomenon, which reveals the system-aticities underlying the apparent disorder and fragmentation ofinteraction It is in establishing these systematicities that interpret-ation becomes analysis And, because the analysis focuses on pat-terns observable in the data, analysts are able to avoid speculatingabout participants’ intentions and understandings, or externalconstraints and influences that might impact on their conduct.Schegloff and Sacks note of their pioneering work in this field:

We have proceeded under the assumption (an assumption borne out by ourresearch) that in so far as the materials we worked with exhibited orderli-ness, they did so not only to us, indeed not in the first place for us, but forthe co-participants who had produced them If the materials (records ofnatural conversation) were orderly, they were so because they had beenmethodically produced by members of the society for one another, and itwas a feature of the conversations we treated as data that they were pro-duced so as to allow the display by the co-participants to each other oftheir orderliness, and to allow the participants to display to each other theiranalysis, appreciation and use of that orderliness Accordingly, our analysishas sought to explicate the ways in which the materials are produced bymembers in orderly ways that exhibit their orderliness and have theirorderliness appreciated and used, and have that appreciation displayedand treated as the basis for subsequent action (1973: 290)

Turns are, in the first instance, built to contribute to the sequence

of actions in which they occur; thus to analyse them in isolation is

to ignore the way they are built to display analysis of, and pation in, the actions embodied by prior turns Every turn-at-talktherefore displays the participant’s definition of the situation; itdisplays an understanding of the activity sequence to which itcontributes, and of what is an appropriate contribution to thatsequence This has an important methodological upshot: the ana-lyst can use the sequential nature of turns at talk as a resource foraccessing the participants’ analysis of the nature of the actionsengaged in.6 From this perspective we can see how Goffman’sobservations on footing and the relationship between reportedand non-reported speech have been an important influence onconversation analytic research into reported speech It is to thiswork – the foundation for the current volume – that we now turn.1.4 CA studies of reported speech

partici-In some respects, detailed analysis of reported speech in context hashighlighted differences between claims by linguists and sociolin-guists and conversation analytic ones, while in others CA researchhas supported and extended previous findings We begin by con-sidering some of the discrepancies illuminated by existing CA work

6 For a more detailed consideration of CA method and its contribution tolinguistics, see Clift (2005)

In section 2we ident ified three main concer ns in the linguis ticsliterature regarding reported speech The first of these was anoverwhelming concern with a proposed distinction between directand indirect forms of reported speech; different structural andfunctional features were attributed to each form What CA is able

to investigate, grounded as it is in participants’ own orientations, iswhether this purported distinction is an interactionally salient one.And indeed it would seem that it is not: recent empirical work hasdemonstrated that in practice these distinctions are often less clear-cut than authors have suggested, and functions attributed to oneform are demonstrated by other forms For example, above we sawthat Li (1986) claims that prosodic features may be used to distin-guish between direct and indirect forms Given that DRS is thought

to be a replaying of a former utterance conveying both form andcontent, while in indirect speech the current speaker can comment

on the reported utterance, it seems reasonable to assume (as manyauthors have) that they will recurrently have different prosodiccontours So this would suggest that DRS may be accompanied bydramatic shifts in prosody or voice quality to distinguish it from thecurrent speaker’s unfooted utterances, highlighting that these arenot the speaker’s own words, and conveying the way in which theutterance was ‘originally’ said In indirect speech, on the otherhand, it is claimed that prosody or voice quality may be used toconvey the current speaker’s attitude towards the reported utter-ance However, in a study of reported speech in German, Gu¨nthnernotes that prosody and voice quality played a crucial role in stagingdialogues in her data She claims that:

simple dichotomies of direct versus indirect speech unduly reduce thecomplexities of reporting past dialogue: direct speech may also incorporateand contextualise the reporter’s interpretation and evaluation of thereported dialogue (1997b: 250)

Indirect as well as direct speech, she finds, is used to present bothwhat was said and how it was said Gu¨nthner concludes thatalternating between different forms of reported speech may con-tribute to a range of activities, such as distinguishing betweendifferent speakers, or to help differentiate background informa-tion from the climax of the story However, these functions cannot

be straightforwardly attributed to the different forms; rather ‘we

seem to be dealing with a complex web of factors in the dynamicinterrelationship of reported and reporting discourse’ (1997b:268; see also Klewitz and Couper-Kuhlen (1999) for an account

of the prosodic features marking out reported speech sequences).Thus, analysis of reported speech in interaction has revealed thatdistinctions between DRS and IRS are not always clear-cut Forexample, a speaker may begin reporting an utterance with whatappears to be DRS, but then switch to IRS (Holt,1996) Indeed

as Bolden (2004) shows, the boundaries of reported speech andother non-reported talk may not be clear-cut at all; in herstudy of conversational Russian, she shows how what she calls

‘fading out’ (2004: 1106) may be deployed to a number of actional ends; in particular, potential problems of alignment andevidentiality

inter-In spite of the fact that CA research has shown that a clear-cutdistinction between IRS and DRS is not always warrantable, andthat uses or actions attributed to one form may also be demon-strated by the other, research focusing on DRS in interaction hassubstantiated and extended linguistic and sociolinguistic claims.For example, the fact that DRS purports to be a replaying of aprior locution has proved to be central to many aspects of the usemade of the device in interaction According to Holt (1996: 229),DRS gives the recipient ‘access’ to the reported utterance, enablingthe recipient ‘to assess it for himself or herself’ Wooffitt (1992), in

an analysis of accounts of the paranormal, finds that speakers useDRS to make their claims more robust Direct speech has long beenseen as a way of not simply recalling a locution but also givingevidence about its form and content Holt demonstrates thatDRS can be used in interaction to give evidence of a former locu-tion: the reported speaker appears to be ‘allowed to speak forhimself or herself’ (1996: 230) However, shifts in prosody or voicequality, as well as other components in the sequence such as storyprefaces, can implicitly convey the speakers’ evaluation Thus,recipients are able to be the first to explicitly evaluate the reportedincident, with tellers sometimes concurrently joining in with theevaluation (Holt,2000)

This characteristic of letting the recipient interpret the reportedspeech for himself or herself (or at least to appear to do so)may help to explain its recurrent association with certain types

of activity sequences Golato’s studies of reported speech inGerman focus on the use of a particular quotative (‘und ich so/und er so’ – ‘and I’m like/and he’s like’) to report embodiedactions (2000) and on self-quotation to report past decisions(2002) Both Drew (1998a) and Holt (2000) examine complaintswith respect to reported speech, and propose that in complaints therecipient can be given ‘access’ to a reprehensible comment, enab-ling him or her to offer a negative assessment of it and thus supportthe teller’s own evaluation Both find that reported speech is recur-rently associated with recounting the climax of a story involving acomplaint.

Conversation analytic research has revealed other environmentswith which reported speech is recurrently associated Reportedspeech is often associated with laughter and can occur in making

a joke or telling an amusing story (Holt, 2000, this volume).Goodwin (1990), in a study of African-American children, findsthat reported speech is used to report contentious comments by athird party to the child targeted by those comments The reduction

of responsibility for a reported utterance partially accounts for theassociation between reported speech and gossip Being able todelegate responsibility for ‘forbidden expressions’ (Bergman1993:113) gives gossips more freedom to transgress normal rules and

‘enjoy playing with taboo modes of expression and turns of phrasethat offend good taste’ (1993: 117) In a study of racial discourse on

a college campus, Buttny (1997) found that speakers would usereported speech in order to evaluate themselves and others, butoverwhelmingly to criticise others (see also Buttny and Williams,

2000)

Existing CA work has thus already done much to illuminate one

of the central concerns in the linguistics literature – that of theexistence or otherwise of distinct forms of reported speech It hasalso, as we have seen, ranged substantially beyond past work, tohitherto uninvestigated contexts In this respect the current volumeconsolidates and enhances existing conversation analyticallyinformed work by grounding the accounts in sequential analysis

As we shall see, it engages with two of the central concerns ofprevious literature – the authenticity of reported speech and whatreported speech does in interaction

1.5 This volum e

The report ed speec h in this volum e is shown in its diver se manifes ation s in a range of context s The convers ation an alytic focus of thechapte rs that follow reveal the extent to which it is the collabora-tive achie vement of all the particip ants The chap ters by Char lesGoodw in and Elizab eth Holt underl ine the inte llectual debt owed

t-by today’s work in report ed speech to Bakhtin /Volosin ov an dGoffm an whi le also explor ing the limitations of the framew orksdevelo ped by them Good win examines both the tellin g of a story inconvers ation a nd the activities of an aph asic man able to speak onlythree words (‘yes’ , ‘no’ an d ‘and’) Video data of these two types ofinteract ion reveal the im portanc e of analys ing foo ting and report edspeec h withi n embodi ed, multi-part y proces ses of interact ion Hol t,using telepho ne data, similarl y shows how report ed speech is theprodu ct of collabo ration by both pa rticipan ts By focusing on whatshe calls ‘enac tments’, wher e participa nts shift footing to enact acharac ter, she throw s light not only on the joki ng scenar ios that theyconstr uct but also on the more specifi c charac teristic s of this type

of repo rted speech , such as the lack of any dist inct intr oductorycompone nt

The ch apters by Elizabet h Couper-Kuhlen and Rebecca Cliftboth focus on hitherto little-studi ed forms of reported speec h –those produced in non-na rrative con texts Coup er-Ku hlen, examin-ing both reported speech and reported thought, discusses twoframing environments for these: assessments and accounts Shegoes on to claim that such non-narrative uses of reported speechconsist maximally of one turn-constructional unit (TCU) Clift, inChapt er 5, produces additional evidence to endorse this claim, andfurthermore examines a recurrent context for such non-narrativeuses of reported speech: competition over rights to assess She alsoinvestigates the means by which reported speech is necessarily not areproduction of what was – or might have been – originally said.The chapters by Markku Haakana and John Rae and JoanneKerby return reported speech and thought to what is perhaps itsmost common environment: story-telling Haakana, examiningFinnish data, focuses on reported thought in complaint stories

as examples of what is not said, and proceeds to examine theframing of the reported utterance through the various introductory

components in Finnish Rae and Kerby, examining a corpus ofstories told by young offenders of their encounters – often with thepolice – reveal how speakers design contexts for reported speech Inaddition, they explore how responsive actions are represented

in reported speech

Rae and Kerby’s chapter is one of four which examine reportedspeech in institutional contexts Renata Galatolo focuses on thereported speech deployed during a notorious Italian murder trial,and discusses the moral and evidential work it performs in suchlegal contexts Steven E Clayman, in work which once morereveals a debt to Goffman, investigates how broadcast journalistspresent themselves as speaking on behalf of the public, showing thestrategic importance of such a practice in the increasingly adversar-ial domain of journalistic interviews In the final chapter, RobinWooffitt examines how reported speech is used by mediums tosupport their claims to have established contact with the dead

By grounding their investigations in the analysis of sequences ofaction, the articles in this volume can thus be seen to return to many

of the issues that have preoccupied students of reported speech overthe years In doing so, they have aimed to anchor what havehitherto been analytic preoccupations in the orientations of theparticipants themselves And, in taking participants’ orientations

as the starting point, they have ranged far beyond the traditionaldomains of inquiry, illuminating further the diversity and richness

of reported speech

is well demonstrated by the large body of significant research onboth reported speech and the dialogic organisation of language andculture that has flourished since the 1970s in a number of differentfields.

In reported speech the voices of separate actors are found in aparticular place, a complex strip of talk produced by a singlespeaker, albeit one quoting the talk of another While recognisingboth the originality and the importance of Volosinov’s insights,

I will argue here that the precise way in which he conceptualisedreported speech actually served to hide, and render invisible toanalysis, crucial aspects of the very dialogic organisation of lan-guage that he sought to probe These include: 1) dialogue as multi-party sequences of talk in which the voices of different participantsare not only heard but actually shape each other; 2) the visible

actions of hearers and thus the multi-party interactive organisation

of utterances (which would seem central to Volosinov’s (1973)interest in a word as shared territory); and 3) utterances which lackthe syntactic and other complexity required to incorporate reportedspeech

To investigate such issues I will look first at one of the mostpowerful and influential models for analysis of the different kinds

of ‘speakers’ that can co-exist within a strip of reported speech:Goffman’s deconstruction of the speaker in Footing (1981) (seealso Goffman,1974/1986) In presenting this model Goffman alsooffered an important framework for the study of participation,and indeed participation seems absolutely central to the dialogicorganisation of human language (C Goodwin, 1981, 1986a;

M H Goodwin, 1990, 1997, 2000; Goodwin and Goodwin, inpress,1987; Heath,1986; Rae,2001) There are, however, seriousproblems with Goffman’s approach to participation What he pro-vides is a typology of participants rather than analysis of howutterances are built through the participation of structurally differ-ent kinds of actors within ongoing courses of action To probe howsuch issues are consequential for the investigation of actual talk

I will first use Goffman’s model of the speaker to describe thedifferent entities visible within a strip of reported speech in a story.This model provides important analytic tools However, its limita-tions become visible when analysis is expanded to include the ac-tions of silent (though consequential) participants, such as the partywhose talk is being quoted To further examine the dialogic organ-isation of both utterances and the speaker I will then look

at the impoverished talk of a man with aphasia so severe that helacks the syntax to construct the rich, laminated utterances required

by the frameworks of both Goffman and Volosinov It will beargued that a quite different notion of both participation and thedialogic organisation of language is necessary to explicate the way

in which this man functions as a powerful speaker by incorporatingthe complex talk of others within his own limited utterances Suchphenomena shed light on the constitution of the speaker and thehearer – the two participant categories that are most central tohuman language – and to the dialogic processes that provide organ-isation for the construction of talk through their interaction witheach other within this framework

2.2 Complex speakers

The deconstruction of the speaker offered by Goffman in Footingdemonstrates the genuine power of an analytic framework thatfocuses on the dialogic interplay of separate voices within reportedspeech Figure2.1is a story in which a teller quotes something thather husband said The story is about one of the prototypical scenes

of middle-class society Friends have got a new house As guestsvisiting the house for the first time, the speaker and her husband,Don, were in the position of admiring and appreciating their hosts’new possessions However, while looking at the wallpaper in thehouse Don asked the hosts if they were able to pick it out, or wereforced to accept it (lines 13–16).1

Who is speaking in lines 14 and 16? The voice that is heard isAnn’s, the current story-teller However, she is reporting somethingthat her husband, Don, said, and moreover presenting what he did

1 This same story was analysed from a different perspective, withoutreference to Footing, in Goodwin (1984) I am indebted to Gail Jeffersonfor transcribing this talk

Figure 2.1 Extract (1)

as a terrible faux pas, an insult to their hosts in the narrated scene.She is both reporting the talk of another and also taking up a parti-cular stance toward what was done through that talk In a very realsense Ann (the current story-teller) and Don (the principal charac-ter in her story) are both ‘speakers’ of what is said in lines 14 and

16, though in quite different ways The analytic framework offered

by Goffman in Footing for what he called the Production Format

of an utterance provides powerful tools for deconstructing the

‘speaker’ into a complex lamination of structurally different kinds

of entities (see Figure2.2)

In terms of the categories offered by Goffman, Ann is the mator, the party whose voice is actually being used to produce thisstrip of speech However, the Author of this talk, the party whoconstructed the phrase said, is someone else, the speaker’s husband,Don In a very real sense he is being held accountable as not onlythe Author of that talk, but also its Principal, a party who is sociallyresponsible for having performed the action done by the originalutterance of that talk Goffman frequently noted that the talk

Ani-of speakers in everyday conversation could encompass an entireFigure 2.2 Production Format

theatre And indeed here Ann is putting Don on stage as a character

in the story she is telling, or in Goffman’s terms animating him as aFigure

Moreoever there is a complex laminated and temporal itation among these different kinds of entities within the space ofAnn’s utterance Thus it would be impossible to mark this as aquotation by putting quotation marks before and after what Donsaid In addition to the report of this talk, the utterance also con-tains a series of laugh tokens, which are not to be heard as part ofwhat Don said, but instead as the current speaker’s, Ann’s, com-mentaries on what Don did through that talk Through her laughtokens Ann both displays her own stance towards Don’s utterance,formulating his talk as something to be laughed at, and, throughthe power of laugh tokens to act as invitations for others to join inthe laughter (Jefferson,1979), invites others to join in such treat-ment Ann thus animates Don as a figure in her talk while simul-taneously providing her own commentary on what he said byplacing her own laugh tokens throughout the strip of speech beingquoted

interdig-In brief, in Footing Goffman provides a powerful model forsystematically analysing the complex theatre of different kinds ofentities that can co-exist within a single strip of reported speech.The analytic framework he develops sheds important light on thecognitive complexity of speakers in conversation, who are creating

a richly inhabited and textured world through their talk In ition to producing a meaningful linguistic sentence, Ann, within thescope of a single utterance, creates a socially consequential image ofanother speaker His talk is thoroughly interpenetrated with an-other kind of talk that displays her stance toward, and formulation

add-of both what he said (e.g as a laughable add-of some type), and the kind

of person that would say such a thing Goffman’s deconstruction ofthe speaker provides us with genuine analytic insights, and tools forapplying those insights to an important range of talk

2.3 Recovering the social and cognitive life of hearers

Goffman’s speaker, a laminated structure encompassing quite ent kinds of entities who co-exist within the scope of a single utter-ance, is endowed with considerable cognitive complexity However,

differ-no comparable semiotic life animates Goffman’s hearers In a ate section of the article they are described as cognitively simplepoints on an analytic grid listing possible types of participation inthe speech situation (e.g Addressee vs Overhearer, etc.).

separ-However, Ann’s talk is actually lodged within a participationframework that has a range of structural features that carry it wellbeyond either a typology of participants, or dialogic text instanti-ated within the talk of a single speaker Don, the principal character

in Ann’s story, the party whose faux pas is being reported, is notjust a figure animated through the talk of the story, but an actualperson who is present at the telling Indeed he is seated right next tothe story-teller Elsewhere Goffman defined a social situation, such

as the gathering where this story was told, as ‘an environment ofmutual monitoring possibilities’ (Goffman, 1972: 63) Central tothe organisation of the participants’ monitoring of each other is theway in which those present ‘jointly ratify one another as authorizedco-sustainers of a single, albeit moving focus of visual and cognitiveattention’ (Goffman,1972: 64) Within the field created by Ann’sstory it is appropriate and relevant for the others present to look atDon, the author of the terrible faux pas, when it is at last revealed.That place for scrutiny of the co-present offender being animatedwithin the talk is defined by the sequential organization of thestory, that is at its climax As principal character in the story Don

is faced with the task of arranging his body for the scrutiny it willreceive when that moment arrives When a videotape of the telling

is examined it can be seen that, as Ann quotes what he said duringlines 14 and 16, Don’s face and upper body perform visual versions

of her laughter Indeed, on looking at the video, it appears thattwo separate bodies are performing the same laugh For example,there is quite precise synchrony between escalation in Ann’s vocallaughter and Don’s visual displays Thus, just as laugh tokens firstappear in ‘wa(h)llpa(h)p(h)er’ in line 14, Don’s face starts to form asmile/visual laugh As Ann’s laughter becomes more intense in line

16 Don’s face matches her escalation with more elaborate headmovements, wider opening of his mouth, etc

The participation framework relevant to the organisation ofAnn’s story, and most crucially the quoted speech within it, thusextends far beyond structure in her talk to encompass the embodiedactions of others who are present Don is faced with the task of