Conversation Analysis

Full text

1
SAGE Research Methods Foundations
Published: 2019
DOI: http://dx.doi.org/10.4135/9781526421036
Methods: Conversation Analysis
Online ISBN: 9781526421036
Conversation Analysis
Ian Hutchby
Introduction
Conversation analysis (CA) is an approach to social research that investigates the sequential
organisation of talk as a way of accessing participants' understandings of, and collaborative
means of organising, natural forms of social interaction. A distinctive methodological feature is
that CA gathers its data of naturally-occurring interactions as they unfold in real time using
video- or audio-recording technology. Recordings are transcribed in close detail to allow for
fine-grained analysis of the design, exchange and coordination of actions within social
interaction. This chapter introduces the intellectual foundations and basic principles of CA;
discusses its approach to understanding turn-taking systems; outlines its approach to data
collection and transcription; illustrates its analytical procedures; and discusses the application of
CA to institutional interaction and other sociological topics.
While, as its name suggests, some practitioners have focused mainly on the investigation
of everyday conversation, examining talk as a social institution in its own right with its own
structures, CA has in fact been applied to a wide variety of different forms of talk. These range
from ordinary telephone conversations to consultations in doctors' surgeries, from family
dinnertime talk to communication between airline pilots and ground crew, from job interviews to
television interviews with celebrities or politicians, to speeches given at political rallies.
Conversation analysts use the term speech exchange systems, of which conversation itself is but
one; and maintain that any such system, in which two or more participants exchange turns at talk
for whatever purpose, can be analysed to reveal significant detail about the organisation of social
life and institutions.
For that reason, conversation analysts prefer to call their object of study talk-in-
interaction:
The term talk-in-interaction [is] prefer[red] to 'conversation' so as to circumvent the
connotation of triviality that has come often to be attached to the latter term, and to
broaden the scope of what we mean to be dealing with to interactional settings that
clearly fall outside the common-sense meaning of 'conversation'. (Schegloff, 2006: xiii)
The human ability to engage in talk-in-interaction is, of course, underpinned by the
ability to use language in the first place. But what is at issue for CA is much greater than the

2
structures of grammar, syntax and semantics, as these tend to be analysed in linguistics. To study
talk-in-interaction is to explore some of the most basic questions about the nature of human
sociality.
For example, talk-in-interaction is central to the organisation of social life in all known
human cultures. It acts as the primary medium for the establishment and maintenance of
interpersonal relationships, for the exchange of information, and for the conduct of social affairs
both at home and at work. Even as societies have developed technologies for mediating
interpersonal communication, ranging from the letter to the internet, the domains of domestic
and institutional life, from family meal times to discussions about international military strategy,
continue, at root, to involve the exchange of talk-in-interaction.
Conversational activity is also one of the earliest and most important means by which
children develop the linguistic skills that underpin their general social competencies. Even before
they are able to understand, let alone produce, words and sentences, children understand turn-
taking, routinely engaging with their caregivers in quasi-conversational exchanges of turns in
both verbal and gestural formats. In developing language, children do more than simply acquire
the lexis and grammar of their mother tongue. Through talk-in-interaction, they acquire the tools
that afford participation as ratified members in everyday society.
Origins and key influences
CA originated in the work of American sociologist Harvey Sacks (1935-75), much of which
initially appeared in the form of lectures he gave for a course on conversation that he developed
in the University of California during the years 1964-72. These lectures were tape-recorded and
transcribed. Sacks collaborated closely with two other important figures in the development of
the field: Emanuel Schegloff, and Gail Jefferson (1938-2008), who herself was the principal
architect of the specialised transcription system that is used in CA (Jefferson, 2004). Between
them these three figures published many of the foundational papers in the field (Sacks, 1963,
1972; Schegloff, 1968; Schegloff and Sacks, 1973; Sacks, Schegloff, and Jefferson, 1974;
Schegloff, Jefferson, and Sacks, 1975). Jefferson's edition of Sacks's original lecture transcripts
was posthumously published as a two-volume collection with an extensive introduction by
Schegloff (Sacks, 1992).
Sacks's ideas were influenced by the form of sociology of everyday life and interpersonal
behaviour that Erving Goffman had developed in the 1950s (Goffman, 1956). He was also
influenced by the ethnomethodology of Harold Garfinkel (Garfinkel, 1967). Sacks had a close
association with Garfinkel in the early stages of his career, and the two co-authored an important
paper (Garfinkel and Sacks, 1970) which drew key parallels between the methods of practical
reasoning and sense-making within cultural settings (the central concern of ethnomethodology)
and members' mastery of ordinary language resources as the medium for this sense-making (the
central concern of CA).
Sacks's originality lay in the fact that he was the first sociologist to take seriously the
sequential organisation of talk as a members' resource for contextualising and therefore
understanding situated social actions. That position was formalised in an influential paper, 'A
Simplest Systematics for the Organisation of Turn-Taking for Conversation' (Sacks, Schegloff
and Jefferson, 1974), in which the main aim is to reveal how the technical aspects of turn-taking
represent structured, socially organised resources by which participants perform and coordinate
activities through talk-in-interaction. This paper contains both a generalisable model of the
management of turn-taking in everyday conversation, and a theory of 'speech exchange systems'
by which that model can be extrapolated to other, non-conversational forms of talk-in-

3
interaction, or so-called 'institutional' discourse types.
This approach has, over the ensuing decades, provided inspiration and grounding for a
vast body of work on the management of activities in everyday social interaction (Atkinson and
Heritage, 1984; Lerner, 2004; Schegloff, 2006; Drew and Heritage, 2007) as well as on the role
of talk in more specialised or formal social settings (Boden and Zimmerman, 1991; Drew and
Heritage, 1992; Arminen, 2005; Richards and Seedhouse, 2005).
In the latter kind of work (sometimes thought of as 'applied' conversation analysis) the
turn-taking system for conversation is used as a benchmark against which other forms of talk-in-
interaction may be recognised for their distinctiveness. Ordinary face-to-face conversation is
treated, for analytic purposes, as an interactional baseline. In comparative terms, all forms of
institutional interaction can be characterised by a systematic reduction and/or specialisation of
the array of practices observable in ordinary conversation. In such a view, the observably
specialised nature of institutional discourse is therefore actively produced by participants. The
upshot is that emphasis is placed not on how the setting determines the activities, strategies and
procedures adopted within it, but on how those activities, strategies and procedures make
available (for participants and analysts alike) participants' orientation to, and reproduction of, the
setting's specialised institutional features.
Analysing naturally-occurring talk
This way of thinking derives from CA's basic methodological standpoint, which is that in all
cases analytical claims should be based on the members' own displayed understanding of their
actions (Schegloff and Sacks, 1973). CA insists that is it is more important to explicate the ways
that the participants in any interaction display their own understanding of what they are doing
and the context in which they are doing it, than to begin from theoretically-driven assumptions
about what participants may be doing or how it might be affected by pre-determined features of
the context. That position evolved in the context of a critique of prevailing methodological
approaches in the social and linguistic sciences; in particular, approaches in structural linguistics
for which language analysis must be carried out in abstraction from actual contexts of use, and
perspectives in sociology and social anthropology for which the language used by societal
members is treated as a window onto the thoughts, beliefs and attitudes of sample populations.
In sociology, anthropology and related social sciences, while recent years have seen a
'turn to language' in important senses, major approaches tended to view the mundane activity of
talking as an unremarkable element of social life. The most widely adopted research methods in
sociology – the interview and the questionnaire survey – and in anthropology – field observation
and informant reports – rely on language as a tool for finding out about social structures and
large-scale social phenomena. But for the most part, research papers render the language used in
the production of their findings invisible.
Sacks, like Garfinkel, argued that members' practical reasoning about social structures
should become a topic of analysis rather than a resource. Sacks's key insight was that members'
practical reasoning becomes observable in the mundane activity of talking. Subsequently, Sacks's
position has been extrapolated much more explicitly in terms of how sociological claims about
the relevance of social variables must be based on the demonstrable orientations of members
themselves to those variables, as displayed in the organisation of talk (Schegloff, 1997).
In linguistics, a prevailing notion was that to analyse language in its most organised form,
it needs to be removed from any actual instantiations in use. Many agreed with the views of
Chomsky (1965), who argued that while it was possible to analyse linguistic 'competence' (innate
knowledge of grammatical structure that enables humans to produce meaningful sentences), the

4
actual use of language (linguistic 'performance') could not be scientifically described because of
the disfluencies and 'noise' sources that occur in natural speech. Linguistics should thus be
concerned with an 'ideal speaker-hearer' whose sentences were purely grammatical, and analysis
was based on invented sentences, rather than utterances made by actual people in everyday
contexts of language use.
Invented sentences were also used, though for practical rather than theoretical reasons, in
the influential 'ordinary language philosophy' of Wittgenstein (1953) and Austin (1962).
However Sacks was critical of the reliance on invented data, observing that while this approach
to research can clearly yield important insights, as a method for investigating interactional uses
of language it is inherently limited. This is because utterances occur not as isolated sentences but
in the fluid context of conversational sequences. As he pointed out (Sacks, 1992, Vol. 2, 5),
there is a major difference between inventing a sentence that would make sense as an utterance,
and inventing a sequence of turns that one could confidently predict the occurrence of. In short,
while we might feel confident in proposing that a particular sentence is grammatical or makes
sense, we feel much less confident if we have to predict a sequence of turns beyond, say, a
question and an answer.
Sacks' aim was to show that everyday language, though it sometimes appears chaotic and
grammatically imperfect, is nevertheless a highly ordered, socially organised phenomenon,
which is therefore scientifically describable. However, he realised that one could not use
intuition to understand the orderliness of ordinary conversation (or any other form of speech
exchange). Rather, it was necessary to gain access somehow to the actual production of
utterances in sequences, in real time.
This leads to a methodological commitment that is one of the hallmarks of CA: an
emphasis on recording interaction in the naturally-occurring settings of social life. The use of
photographic, audio and video recording technology in both the collection and analysis of data
on human communication was pioneered in the 1940s and 50s by human ethologists such as
Birdwhistell (1952), Bateson (1956) and Mead (Bateson and Mead, 1942), who in turn
influenced researchers in facial and bodily kinesics (Scheflen and Scheflen, 1972; Kendon,
1990). Its use in the study of speech interaction was strongly advocated by conversation analysts
from the beginning. Even though recordings themselves are limited in the sense that they may
not capture every potentially relevant thing that occurs in and around the vicinity of the
recording device, the advantages far outweigh those limitations. As Sacks put it:
Such materials had a single virtue, that I could replay them. I could transcribe them
somewhat and study them extendedly – however long it might take. The tape-recorded
materials constituted a 'good enough' record of what had happened. Other things, to be
sure, happened, but at least what was on the tape had happened. (Sacks, 1984: 26)
All CA research is therefore based on the analysis of tape recordings of naturally-
occurring behaviour. The term 'naturally-occurring behaviour' refers to behaviour that would
have taken place whether or not the researcher intended to record it. Hence, behaviour recorded
in an experimental laboratory set-up, or in an interview initiated for specific research purposes, is
usually thought to fall outside this definition.
Data and transcription
Initially, CA researchers restricted their attention mainly to recordings of telephone
conversations. These were relatively easy to record in naturalistic settings (people's homes), and

5
due to the absence of face-to-face cues and gestures, enabled a focus entirely on the mechanics
of speech exchange (Hopper, 1992). In recent times, video-recordings have become more widely
used. However, CA's explicit focus on the organisation of talk-in-interaction means that gesture,
body movement and facial expression tend not to be studied in their own right, as may be the
case in the field of interactional kinesics (Kendon, 1990), but rather in exploring the relationships
between speech and body movement (Heath, 1986).
It is important to stress that, for CA, it is the tape-recording itself rather than the
transcript which is thought of as the primary data. The aim is to analyse the data (the recorded
interaction) using the transcript as a convenient referential tool. The transcript is thus seen as a
'representation' of the data; while the tape recording is viewed as a 'reproduction' of a
determinate social event. The tape is only one form of reproduction; and whether it is an
audiotape or a videotape, it does not reproduce everything that went on in the vicinity of the
recording device during the time it was switched on.
At the same time, clearly there are innumerable phenomena in any given stretch of talk
which could be transcribed to varying levels of detail. No transcription system exists which is
able, or lays claim to being able, to capture all the possible features of talk that may be
observable. As Kendon (1982: 478) remarked:
It is a mistake to think that there can be a truly neutral transcription system, which, if
only we had it, we could then use to produce transcriptions suitable for any kind of
investigation...Transcriptions, thus, embody hypotheses.
Similarly, Ochs (1979: 44) describes transcription as 'a selective process reflecting theoretical
goals and definitions'.
A CA transcript therefore embodies in its format, and in the phenomena it marks out, the
analytic concerns which conversation analysts bring to the data. These concerns are of two types.
First, the dynamics of turn-taking. On this level, transcripts seek to capture the details of the
beginnings and endings of turns taken in talk-in-interaction, including precise details of overlap,
gaps and pauses, and audible breathing. Second, the characteristics of speech delivery. Here,
transcripts mark noticeable features of stress, enunciation, intonation and pitch.
A list of the principal transcription conventions used in CA is provided in the Appendix
to this chapter. Below I illustrate some of these key conventions with reference to actual
transcripts.
Overlap onset is marked in transcripts by the use of left hand square brackets:
(1) [SBL:1:1:11:5]
1 B: Uh huh and I'm so:rry I didn' get Mar:gret I
2 really['ve been wan'ing to.
3" D: [W'll I think she mu:st've stayed out'v
4 to:wn
5 (0.2)
6 B: I thi[nk so too.
7" D: [in Fre:sno sh- see she'n Pe::g (0.7)
8 dro:ve over to 'er sister's 'oo lives in Fresno::.
As we see with this extract, the aim is to be as precise as possible about marking the point
at which overlap begins: even when, as in lines 6 and 7, that is in the middle of a word. By
marking the precise points of onset in overlapping talk, very close calibrations in the
understandings that speakers display of each other's talk become available for analysis.

6
Other forms of overlap may also be marked. For instance, speakers may start a turn
simultaneously. This is marked by the use of a double left-hand bracket:
(2) [Heritage:I:II:3]
1 N: Ye:s.
2" I: [[It's uh:
3" N: [[And and who did you go to.
Utterances also may sound 'latched' together: they may occur right next to each other
with absolutely no gap, but also no overlap, between them. This is marked by equals signs, thus:
(3) [NB:IV:10]
1 E: Is the swimming pool enclosed with the
2 gla:ss bit?=
3 N: =No::, it's uh: ou:ts- (.) eh no outside
Equals signs are also used to deal with a problem inherent in the attempt to represent
naturally-occurring talk on the page: the fact that page lines are strictly limited in length, whereas
conversational turns are not. This means that a single utterance may need to be broken up, if, for
instance, another speaker says something during its course. In such a case, the two parts of the
longer turn are connected by equals signs, with the embedded utterance transcribed on a line
between:
(4) [NB:IV:14]
1 N: But eh- it's- it's terrible to keep people
2" ali:ve and [you know and just let them=
3 E: [Right.
4" N: =suffer day in and day out,
In a similar sense, a protracted spate of simultaneous talk may require a combination of
left and right brackets (marking the end of overlap) and equals signs, if the spate extends across a
number of lines on the page:
(5) [NB:IV:14]
1 E: Well, we don't know what it's all about
2 I g-I- ((sniff)) Don't get yourself=
3 N: =[[Oh I'm not. I just- you know I wish ]
4 E: [[Honey you've got to get a hold of your- I know]=
5 N: =I'd- I'd kind of liked to gone out there but
6 I was afraid of the fog
Because CA transcripts seek to represent the production of talk in real time, the length of
a word's enunciation is marked if it is noticeably extended. This is done by inserting colons into
the word at the point the speaker extends it:
(6) [SBL:1:1:11:5]
1 B: I thi[nk so too.
2" D: [in Fre:sno sh- see she'n Pe::g (0.7)
3" dro:ve over to 'er sister's 'oo lives in Fresno::.
The longer the audible extension, the more colons are inserted:

7
(7) [HG:28]
1" N: A::nywa::y,
Conversation analysts seek to time intervals in the stream of talk relatively precisely, in
tenths of a second. This is not just a matter of accuracy. Research has shown (Davidson, 1984;
Pomerantz, 1984) that pauses even as short as two or three tenths of a second can have some
interactional, and therefore analytic, significance. Jefferson (1989) produced an extensive
exploration of the interactional significance that is attached to silences of one second in length
during conversation. It is clear that using a catch-all device such as writing 'pause' in transcripts
would not have enabled many of the finer analytic points contained in these papers to be made.
Timings of pauses, then, are important features of transcripts. The timings, which are
usually done with a stopwatch, are inserted in the transcript at the precise point of their
occurrence in the recording. They may occur within a turn:
(8) [SBL:2:7:20]
1 A: I- if you want to uh(b) (1.1) maybe get up a game...
Or they may occur between turns:
(9) [SBL:2:7:20]
1 A: I- if you w ant to uh(b) (1.1) maybe get up a game
2 some morning while you're out the:re,=why that's
3 always fu:n,
4 B: Mm hm.
5" (0.5)
6 A: So let me kno:w.
Pauses that are detectable, but run for less that 0.2 of a second, are indicated by a period
within parentheses (micropauses):
(10) [SBL:2:7:20]
6 A: So let me kno:w.
7" (.)
8 B: Yah will do:.
Breathiness which is audible to the transcriptionist is marked by 'h' for exhalation and '.h'
for inhalation. This feature of recordings is transcribed because (among other reasons) audible
in-breaths may be involved with the management of turn-taking. For example, an open-mouthed
in-breath may mark a participant's attempt to start a turn. Notice how in the following extract,
speaker E draws in a long breath at line 3, which overlaps substantially with P's invitation. This
inbreath signals her attempt to take the floor and respond to the invitation.
(11) [NB:52:2:66]
1 P: Oh I mean uh: you wanna go to the store er anything
2 over at the Market [Basket er anything?-
3" E: [.hhhhhhhhhhhhhhhhhh-h Well honey...
In transcripts, the longer the breath, roughly, the longer the line of 'h' or '.h' provided in
the transcript:

8
(12) [HG:28]
1" N: .hhh
2 (0.5)
3 N: A::nywa::y,
4 (.)
5" H: eh-eh .hhhhhhh Uh::m,
Transcripts also mark plosive aspiration within a word: this happens, for instance, when
someone speaks 'through' laughter (that is, laughs while enunciating a word or phrase). This is
indicated by placing the 'h' in parentheses. However, due to the nature of the phenomenon, this
can be very hard to do, and the transcriptionist may need to listen to the same word or group of
words many times before he or she can achieve a satisfactory textual rendering of the sounds that
occur on the tape (see especially the enunciation of the word 'eyes' in line 6 of extract 14):
(13) [EB:1]
1 S: I hope by next semester it'll be a bi(h)t
2 b(h)edd(h)er heh heh heh heh .hh
(14) [H:2.2.89:4:3]
1 H: Yes but you ca:n't actually:, take anybody to
2 la::w, .h jus:t on:, an accusation.
3 (1.1)
4 C: .p .hhh (.) No I kno:w I'm not just making
5 accusations I've got proof of my own
6" ey(h)(huh huh hih)es! .hhhh
As shown here, laughter, which may previously have been represented descriptively by
the transcriber simply writing '(laughs)', should be transcribed as literally as possible in the form
of onomatopaeic renditions of laugh particles: 'ha ha', 'heh heh', 'hih hih', and so on. The
particles are designed to represent as closely as possible the sounds emitted by the participants:
(15) [Goodwin:GR:40]
1 J: So I said look Gurney, yer just a big ass
2 kisser, (0.4) en [yer getting yer wa:y, ]
3" B: [AAHh hah-uh hah-uh huh]=
4 J: =[I(h) ju(h)st ] lai[d it a:]ll on,
5" B: [.hhhhhhhhhhh ] [ hhah ]
6 (.)
7" B: ehh huh uh-huh uh-huh
One of the difficulties with transcribing laughter is that participants frequently laugh
together. This can make for a highly complex transcribing job, especially when the recorded talk
involves more than two people. In the following fragment, there are only three participants; but
their extended stretch of laughing together leads to a rather daunting-looking section of
transcript:
(16) [Goodwin:AD:56:r]
1 B: he:uh[he-uh-ha ]
2 C: [he[ha: : ]ah ha-ha-ha-ha-ha-ha-a-ha]
3 L: [ah!ah!]ah!ah!ah!ah!ah!ah!ah!ah! ]=

9
4 C: =[.hhh [he]he]:h=
5 L: [ah!a![a!]
6 B: [ah hh]
7 C: =he[:h he-ehh-e-he- [he- [.e-.hhee[hh!]
8 B: [Oo:::::ps, n[he [u -huh [
9 L: [eh!]=
10 C: =[e::::a::yee:]ee::
11 L: [uh!ah!ah!ah!]
To summarise, the transcription system used in CA goes to produce transcripts that are
accurate at the relevant levels of detail (levels of turn-taking and actual speech delivery), while
avoiding being technically inaccessible to the majority of readers. The purpose of the
transcription conventions, however, is not merely to produce accurate representations of talk, but
to focus attention on those features of talk-in-interaction that are analytically significant from the
standpoint of CA.
Turn-taking: Some basic observations
Following is an initially unremarkable extract of conversation recorded during a telephone call.
However, as we look in detail at the transcript, it reveals a wide range of the most basic aspects
of the organisation of talk-in-interaction that have been described within conversation analysis.
We start, in line 1, at the beginning of the conversation, just as Nancy picks up the phone to
answer Edna's call.
(17) NB:II:2:1-2
1 Nancy: Hello:,
2 Edna: .hh HI::.
3 (.)
4 Nancy: Oh: 'i:::='ow a:re you Edna:,
5 Edna: FI:NE yer LINE'S BEEN BUSY.
6 Nancy: Yea:h (.) my u-fuhh! h- .hhhh my fa:ther's wife
7 ca:lled me,h .hhh So when she ca:lls me::, h I
8 always talk fer a lo:ng ti:me cuz she c'n afford it
9 en I ca:n't.hhh[hhhh[huh]
10 Edna: [↑OH:[:::]: my [go:sh=Ah ↑th]aht=
11 Nancy: [↑↑AOO:::::hh!]
12 Edna: =my phone wuz outta order:
13 (0.2)
14 Nancy: n[:No::?
15 Edna: [I called my sister en I get this busy en then I'd
16 hang up en I'd lift it up again id be: busy.
17 (0.9)
18 Edna: .hh How you doin'.
19 Nancy: .t hhh Pretty good I gutta rai:se.h .hh[hh
20 Edna: [Goo:u[d.
21 Nancy: [Yeh
22 two dollars a week.h
23 (.)
24 Edna: Oh [wo:w.
25 Nancy: [↑Ih:::huh hu[:h huh,
26 Edna: [Wudee gun: do with it a:ll.
27 Nancy: Gol' I rilly I jis' don't know how Ah'm gunnuh
28 spend all that money.
29 (0.2)

10
30 Edna: Y'oughta go sho:pping,
31 Nancy: .hhhh Well I should but (.) yihknow et eight
32 dollars a mo:[n:th:, anything I'd] buy'd, be using=
33 Edna: [hm hmm hm-mm-hm. ]
34 Nancy: =up my raise fer 'alf [a YEA:R:] ((smile voice))
35 Edna: [Ye:a:h. ]
This transcript shows a number of relevant features of the socially organised nature of
talk-in-interaction. At the most basic level, it is designed to display how the talk is organised into
a series of turns. For conversation analysis, however, turns do not just occur in a serial order (one
turn followed by another, and so on): they are sequentially organised (Sacks, 1987). That is to
say, there are orderly ways in which one turn is related to a next; and in which turns are therefore
coordinated into patterned sequences through which particular activities are accomplished.
That orderliness is described by treating the transitions between turns as revealing two
kinds of things. First, 'next turn' is the place in which speakers display their understanding of a
prior turn's possible completion. Second, next turns are places where speakers display their
understanding of a prior turn's 'content,' or more specifically, the action it has been designed to
do.
In terms of the first issue, we find that overwhelmingly, turn-transitions are coordinated
by participants with minimal gap and overlap between utterances. Note in the transcript above,
for instance, that only five between-turn gaps occur (in lines 3, 13, 17, 23 and 29), and the
longest of these is no more than two-tenths of a second. (I return to the issue of overlaps
presently.) Participants are able to achieve this level of coordination between turn endings and
next turn beginnings because of a basic set of features in the actual make-up of turns themselves.
Sacks, Schegloff and Jefferson (1974) proposed that turns are made up of 'turn-
constructional units' (TCUs) - examples are: a sentence, a clause, a phrase, or a single word such
as 'Hello' - which are recognisable by members of a language culture. The end of any TCU
represents a point at which a next speaker may legitimately make a bid for the floor. Sacks et al.
captured this feature by referring to TCUs as presenting 'transition-relevance places' (TRPs) at
their completion. At a TRP, a candidate next speaker may, but need not, attempt to take a turn;
while a current speaker may, but need not, attempt to produce a next TCU. Current speakers may
also select a next speaker, in which case the one selected is obliged to take a turn at that point.
These rules for turn-taking are context-free: that is, they allow for such local contextual
variations as the identities and number of speakers, length and content of turns, and so on. But
they are also context-sensitive in that they apply to the local circumstances of particular turns in
particular conversations.
The crucial point about these rules is that they are observably oriented to by members. An
orientation to the possible completion of a turn at TCU completion, and the legitimate relevance
of turn-transition at that point, can be illustrated using the extract above if we focus on the
occurrence of overlap. On the face of it, overlapping talk may be considered evidence of an
incoming speaker's failure adequately to attend to the status of a current speaker's turn: that is,
they might be seen as starting 'too early' on their turn; or starting up at a point where the turn has
not yet reached a recognisable transition-relevance place. However, it is possible to show that
most instances of overlap in extract (17) clearly occur in the environment of TRPs.
For instance, in line 9, what Edna's 'OH:::::' overlaps is a quiet laugh particle, 'hhh hhh
huh,' which Nancy fits onto the end of a TCU: 'So when she calls me::, h I always talk fer a lo:ng
ti:me cuz she c'n afford it en I ca:n't.' In line 21, what Nancy overlaps is the last phoneme of a
recognisable TCU: Edna's assessment, 'Goo:ud,' of Nancy's announcement 'I gutta rai:se.'
Other instances appear more complex, but can still be accounted for as orderly. For

11
example at line 11, Nancy's 'AOO:::::hh!' is in fact a high-pitched laugh, which seems to be
produced in overlap with - but prior to the recognisable completion of - Edna's remark that she
thought her phone was out of order (since she had tried numerous times to get through).
However, notice that in her immediately prior turn Nancy had offered a joke of sorts about
talking on the phone for a long time when her father's wife calls, 'cuz she c'n afford it en I ca:n't.'
She then begins quietly to laugh. Edna's turn is begun with a loud, and high-pitched 'OH::::: my
gosh,' to which Nancy responds with her similarly-pitched 'AOO:::::hh!' Nancy hears the action
that Edna is doing as that of responding to her joke, and starts to laugh by reference to 'OH:::::
my gosh' as a recognisable, and possibly complete, joke-response. The overlap is complicated
because Edna carries on her turn, following up Nancy's quip by herself quipping about the length
of time Nancy's phone has been engaged: 'OH::::: my gosh=Ah thaht my phone wuz outta order.'
Focusing on this instance of overlap, and wondering why it occurs in the places that it
does, illustrates a principal policy of CA, and also suggests the analytical pay-off from that
policy. The policy is to treat anything that occurs in talk-in-interaction as possibly orderly - to
dismiss no detail a priori as disorderly, trivial, or irrelevant. The pay-off is that we thereby gain
an insight into the nature of participants' own understandings of what is going on at any moment
in interaction, as displayed in the ways their turns address the turns they are sequentially next to.
CA refers to this as the 'next-turn proof procedure'.
Next-turn proof procedure
CA's interest in how the sequential organisation of talk can be used to reveal the ways
participants exhibit understanding of one another's utterances can be illustrated further by
looking at the unfolding of a real-time exchange piece of data which occurred during a breakfast
conversation between a mother and her school-age son. Let us look first at one particular turn:
(18) KR:2
1 Mother: Do you know who's going to that meeting?
Taken in the abstract, this can be seen as an example of what linguists would call an 'indirect
speech-act' (Levinson, 1983). These are utterances which are not necessarily meant to be
understood literally. For example, the sentence 'Can you pass me the salt?' is not usually taken to
be a literal question about the recipient's ability to pass the salt, requiring a literal answer such as
'Yes, I can' or 'No, I can't'. Rather, its basic interactional function is to request the recipient
actually to pass the salt. The speaker's intention is not communicated directly in the words that
are spoken (as they might be, for instance, in the sentence 'Pass me the salt!') but by means of
implication.
In terms of extract (18) above, Mother's utterance could in fact be interpreted as doing
one of at least two different types of indirect action. It could represent a genuine request for
information about who is attending the meeting; or it could be a 'pre-announcement': a
preliminary to some information Mother wishes to announce about who is, in fact, going
(Terasaki, 2004). In the first case, the required response would be an answer which informs
Mother about the meeting's attenders ('Yes, Mr So-and-so and Mrs So-and-so'); whereas in the
second case, the response would be something like, 'No, who?', which would provide the
opportunity for her to announce what she knows about the attenders.
But seeing this simply as an example of the abstract category 'indirect speech-act' does
not tell us about the kind of utterance it actually was in the unfolding sequence in which it
occurred (Schegloff, 1987). As noted, for CA the crucial issue is how the participants in that

12
interaction understand, or make sense of, any given utterance. In using the next-turn proof
procedure, therefore, the analyst must observe how the recipient(s) of such utterances interpreted
them.
As the sequence unfolds, this is what we find:
(19) KR:2
1 Mother: Do you know who's going to that meeting?
2 Russ: Who
3 Mother: I don't know!
4 Russ: Ouh:: pro'bly: Mr Murphy an' Dad said
5 prob'ly Mrs Timpte an' some o' the teachers.
Russ's first response, 'Who', clearly shows that he initially interprets Mother's utterance as a pre-
announcement: he is inviting her to tell him what he assumes she already knows about who is
going to the meeting. However, Mother's next turn, 'I don't know!', displays that Russ's inference
was incorrect: she was actually asking him to tell her what he knows about who is going to the
meeting. Notice how Russ then backtracks, re-interpreting the first turn as a genuine request for
information, and produces (in lines 4-5) an amount of information about who might be at the
meeting.
This sequence demonstrates a number of things. First, that people's understandings of one
another's actions can actually unfold as sequences themselves unfold. This shows clearly the
relevance of the next-turn proof procedure: if an understanding happens to be incorrect, that in
itself can be displayed and repaired in the following turns in the sequence. The sequence above
also demonstrates that people's utterances in conversation are not necessarily determined by their
individual beliefs, preferences or mental states but can be determined instead by their orientation
to the structural organisation of conversation. The fact that Russ displays in the final turn that he
does possess knowledge about who is going to the meeting despite initially responding with
'Who?', shows that it was not that he did not know, but that he was orienting (incorrectly as it
turned out) to Mother's first turn as doing a particular kind of action (a pre-announcement)
requiring a particular kind of response.
Adjacency pairs: The conditional relevance of next position
Mother's utterance in the previous extract is also an example of a first part in an adjacency pair:
one of the central concepts in CA research. Adjacency pairs are sequences in which one category
of utterance makes relevant a circumscribed class of responses in next position. Easily
recognisable examples are: a question, which makes an answer relevant as the next move; a
greeting, which makes a return greeting relevant in next turn; an invitation, which makes an
acceptance or declination relevant in next position; or an accusation, which makes a rebuttal or
justification relevant next.
The adjacency pair concept illustrates the way in which particular types of utterance can
be made conditionally relevant by prior turns. The production of a first part of a pair-type, such
as a question, sets up the constraint that a next speaker should respond by producing the relevant
second part from that type - in this case, an answer. Moreover, whatever does follow a first pair-
part will be monitored for exactly how it works as a response to that move. In other words, not
producing something hearable as an answer in the next turn following a question is an
accountable event: something for which a participant may be sanctioned, as in the following
extract:

13
(20) TW:M:38
1 Child: Have to cut these Mummy.
2 (1.3)
3 Child: Won't we Mummy.
4 (1.5)
5 Child: Won't we.
6 Mother: Yes.
The child first announces that they will 'Have to cut these', then, following a pause (line 2)
displays that she was expecting a response of some sort from the mother: in line 3 the child asks
her to confirm the initial observation. Getting no response again in the 1.5-second pause in line
4, child pursues that response in line 5, after which the mother finally answers.
By saying that a second pair-part is conditionally relevant given a first, conversation
analysts are pointing to something specific about the adjacency pair relationship. It is not that
they are merely describing the fact that certain types of turns follow other types. Rather, they are
pointing to the fact that the relationship between adjacency pair-parts is a normative one: in other
words, one that has moral dimensions. This can be seen on two levels. First, motivational
inferences can be drawn from the non-occurrence of a second part following the production of a
first. For instance, not returning a greeting may be taken as a sign of rudeness; not providing an
answer to a question may be taken as indicative of evasiveness; while not proffering a defence to
an accusation may be taken as a tacit admission of guilt. Second, the oriented-to relevance of
second parts following the production of a first can remain in play across time: it is not limited to
cases of literal adjacency.
Thus, instances in which, say, a question is followed by another question, rather than an
answer, may seem to militate against the force of the adjacency pair concept. But such cases in
fact can quite strongly display the temporally extendible relevance of the adjacency pair
framework, once we see that the second question routinely represents a first move in an insertion
sequence (Levinson, 1983: 304-6). Insertion sequences defer a second pair-part's production, but
they do not negate its relevance. A speaker may respond to a question such as, 'Can I borrow the
car?' with another question: 'How long do you need it?' The response to that inserted question -
say, 'Only a couple of hours' - provides a next slot in which a response to the first question is
once more relevant and to be monitored for.
Another aspect of the normative properties of adjacency pairs lies in the systematically
different ways that recipients of first parts design the alternative actions to be done in second
position. Invitations, for instance, can be accepted or declined; requests can be granted or
rejected. The significant point is that these alternatives are non-equivalent. That non-equivalence
is traced in the features of turn design through which alternative second parts are proffered.
Broadly, responses which agree or are congruent with the expectation projected by a first pair-
part are produced contiguously and without mitigation. Responses which diverge from that
expectation - which in some way disagree - tend to be prefaced by hesitations, discourse markers
such as 'Well...,' and, unlike congruent responses, are accompanied by accounts for why the
speaker is responding in this way (Pomerantz, 1984).
Extracts (21) and (22) show the first type of response, where the second speaker agrees
with the assessment produced in the first turn, and their turns are produced immediately and
straightforwardly:
(21) JS:II:28
1 Jo: It's a beautiful day out isn't it?
2 Lee: Yeh it's just gorgeous.

14
(22) VIYMC:1:2
1 Pat: It's a really clear lake isn't it?
2 Les: It's wonderful.
Extract (23), on the other hand, shows someone seeking to decline an invitation, in which rather
than straightforwardly and immediately declining, the second speaker produces talk that softens
or mitigates the declination. This consists of (a) delaying the start of the turn with a slight laugh
('hehh') and the word 'Well'; (b) issuing an appreciation of the invitation ('that's awfully sweet of
you'); and (c) providing an account for why the invitation is being declined:
(23) SBL
1 Mary: Uh if you'd care to come over and visit a little
2 while this morning I'll give you a cup of coffee.
3 Ida: hehh Well that's awfully sweet of you, I don't
4 think I can make it this morning .hh uhm I'm
5 running an ad in the paper and- and uh I have to
6 stay near the phone.
These different response types are termed preferred and dispreferred respectively. The concept
of preference in CA is not used to refer to the psychological dispositions or motives of
individuals; but to point to just this structural feature of the sequential organisation of some types
of adjacency pair. Research has additionally shown that the design features of dispreferred
responses can be used as a resource for the maintenance of social solidarity in talk-in-interaction.
This is so not only in the way that dispreferred responses may be accompanied by accounts or
explanations; but also in the way that hesitations and other means of marking a dispreferred
response can provide a source for a first speaker to revise the original first pair-part in such a
way as to try and avoid disagreement or rejection (Davidson, 1984).
This happens in the following extract, where Edna, hearing that Nancy may be about to
turn down her invitation to come over for lunch, issues a candidate reason for Nancy to decline
the invitation (line 4):
(24) NB:II:2:4
1 Edna: Wanna come down an' have a bite a' lunch with
2 me:?=I got some bee:r en stu:ff,
3 (0.2)
3 Nancy: Well yer real sweet hon:, uh::m (.) [let- I hav-
4→ Edna: [Or do yuh have
5 sum'pn el[se t-
6 Nancy: [No: I have to uh call Bob's mother. .h I
7 told 'er I:'d ca:ll 'er this morning.
These points bring out again the centrality, for CA, of the inferential properties associated
with speakers' moves in interaction sequences. They also address the ways that those inferences
have a distinctly moral, or evaluative, dimension. Speakers can be seen not only to be
establishing and maintaining mutual understanding of one another's actions in sequences of talk,
but also to be holding each other accountable for those actions. In this sense the adjacency pair
framework, and the preference organisation that operates for some types of adjacency pair,
constitute an important site in which to observe the relationships between patterns of language
use and structures of social action.

15
The application of conversation analysis
CA's approach to sequence organisation has enabled it to develop a distinctive perspective on
how participants themselves play a central role in establishing and reproducing the context-
specific nature of their interaction. This is based on the idea that different forms of talk should be
viewed as a continuum ranging from the relatively unconstrained turn-taking of mundane
conversation, through various levels of formality, to ceremonial occasions in which not only who
speaks and in what sequential order, but also what they will say, are pre-arranged (for instance,
in wedding ceremonies). By selectively reducing or otherwise transforming the full scope of
conversational practices, concentrating on some and withholding others, participants can be seen
to display an orientation to, and at the same time to be reproducing, particular institutional norms
as relevant for their current state of interaction.
Using this approach, what is sometimes thought of as an 'applied' wing of CA has
developed a basic typology of institutional discourse, characterised at root by two main forms
known as formal and non-formal. The formal types are represented by courts of law, most kinds
of interview, certain kinds of classroom environment, and various ceremonial occasions. In such
settings there is a close relationship between the social identities adopted by participants and the
types of turn that they produce in interaction (Drew and Heritage, 1992). For example, one role
incumbent (the attorney, the interviewer) asks questions while the other (the witness, the
interviewee) attempts to answer them.
The category of formal institutional interaction incorporates only a small number of
institutional settings. Far more widespread are what are referred to as the non-formal types,
which occur in medical, psychiatric, social service, and business-related environments. In such
settings, much less uniformity in the patterning of conduct is evident. The interaction is more or
less explicitly directed towards carrying out official tasks such as diagnosing illness or making
decisions about clients' health or welfare needs. Typically, these official tasks and activities are
managed within turn-taking frameworks that allow for considerable variation, improvisation and
negotiation in terms of the participation status adopted at given moments by lay and professional
participants alike.
For this reason, CA refers to non-formal types of institutional interaction as having a
quasi-conversational character. The differences with ordinary conversation tend to be seen in the
form of asymmetries that emerge out of patterns of interaction based on a mutual orientation to
specific activities associated with the situation's task-oriented work. For example, in doctor-
patient consultation the interaction can have a conversational feel at certain points; but in the
overall structure the activities of questioning and answering, and the forms of information sought
and given, are asymmetrically distributed as between doctors and patients (Maynard and
Heritage, 2004).
This more applied wing of CA acts as one type of challenge to some of the critiques that
have come from within sociology and linguistics. One critical argument is that CA lacks an
adequate sense of the contextualisation of utterances within a wider set of social relations. A
second, related claim has been that CA in general is unwilling to make links between the micro
details of talk-in-interaction and the macro levels of sociological variables – class, gender, power
relations, ideologies, cultural values and so forth.
Such critiques are fuelled by the fact that much CA writing does often come across as
reluctant to engage explicitly with sociological concepts such as power, gender, class, and so on.
Partly, this has to do with its theoretical lineage. As noted, CA emerged in the context of a
critique of conventional sociological thinking which sees talk mainly as a resource for
demonstrating the existence of macro-level phenomena (power, gender, class, etc.) that, for

16
sociological theorising, inevitably affect people's behaviour. For instance, a characteristic claim
might be that 'our social practice in general and our use of language in particular are bound up
with causes and effects which we may not be at all aware of under normal conditions'
(Fairclough, 1995: 54). Another might be that social institutions 'are characterised
by...heirarchical relations of power between the occupants of institutional positions' and,
consequently, in their actions institutional agents 'exercise the power which is institutionally
endowed upon them' (Thompson, 1984: 165).
For CA, such views simplify the question considerably. Rather than taking it for granted
that unconscious causes or power relations determine the course of encounters, CA aims to
describe the ways that participants display their active awareness of – or orientation to – context-
specific factors, in and through the design of their talk. By focusing on the local management of
talk-in-interaction, the CA approach can provide compelling, but differently structured, accounts
of how power comes to operate as a feature of, and is used as a resource in, institutional
interaction. The question, in short, becomes one of how participants manage role-related
distributions of interactional resources to achieve effects that are differentially available to others
in the setting (Hutchby, 1996).
An important phenomenon here (though by no means the only one) is the question-
answer sequence that is central to much of the institutional interaction that CA researchers have
analysed. In many forms of institutional interaction, questions do indeed get asked primarily by
institutional figures, such as attorneys, doctors, and news interviewers. Questions are a powerful
interactional resource for the simple reason that the asking of a question places constraints on the
discourse options available to its recipient. And while individual questions constrain, sequences
of questions can constrain more strongly.
For example, in Atkinson and Drew's (1979) courtroom studies, the fact that the attorney
is able to ask sequences of questions which the witness is restricted to answering gives particular
powers to the attorney. One of these is what Drew (1992) later described as the 'power of
summary'. The questioner 'has "first rights" to pull together evidence and "draw conclusions'''; in
other words, to define the meaning and the terms of a particular set of answers, which is
something that the witness cannot do. 'The witness is left in a position of addressing and trying to
deal with the attorney's selection of which items to pull together: she has no control over the
connections which are made...nor over the inferences which may be drawn from such
juxtapositioning' (Drew, 1992: 507).
In such a formal type of institutional interaction, the normative constraints that inform
participants' actions are severe, and there may be particular consequences of breaking them (for
example, 'contempt of court'). In more informal settings, institutional figures may not have the
support of such sanctions in their role of asking the questions. This in itself can have distinctive
consequences for relations of power within institutional encounters. For instance, studies of
doctor-patient interaction have shown how patients, by withholding talk at certain points in a
diagnostic sequence for example, may be complicit in the construction and maintenance of a
power situation in which the doctor not only determines the topics that will be talked about, but
also defines the upshots and outcomes of their discussions (Frankel, 1984; Heath, 1992; Maynard
and Heritage, 2004).
Thus, by showing how participants display an orientation to institutional settings by
engaging in certain activities and refraining from others, CA can be used to demonstrate how
power is actively produced as a feature of those activities. To put it another way, oriented-to
activity patterns, such as differences in questioning and answering moves, are themselves
intrinsic to the play of power in institutional interactions.
In summary, CA is not entirely resistant to linking the properties of talk with higher-level

17
sociological variables. But CA is resistant to assuming linkages between the properties of talk
and higher-level sociological variables. The debate is not so much about the existence of factors
such as power in interaction. It is, rather, a methodological argument, about the nature of claims
that can legitimately be made about the data we gather to analyse language use in social
interaction.
Conclusion
In very simple terms, it is possible to distinguish between two traditions in the study of language
use within society. Clark (1992) has called these the 'product' tradition and the 'action' tradition.
He distinguishes between approaches that focus largely on mental processes, and those that focus
largely on social actions. In the product tradition, which is associated with psychology and some
areas of linguistics (or psycholinguistics), the aim is to understand the underlying cognitive
processes which are involved in the production and understanding of language. The basic idea is
that speaking and listening are 'autonomous processes...that investigators can study by looking at
individuals in isolation' (Clark, 1992: xxvi).
The action tradition, on the other hand, is more concerned with observing what people
actually do with language, and understanding how communication is immersed in the course of
interaction itself. Here, in sociolinguistics and related approaches, the view is that 'speaking and
listening are parts of collective, or joint, activities' (Clark, 1992: xxvi). In other words,
investigators are primarily interested in studying the practical accomplishment of communicating
and understanding within social interaction, rather than the mental processes that lie behind it.
Conversation analysis is an approach that belongs very firmly within the action tradition.
In this chapter we have seen how, beginning from its commitment to collecting recorded data in
natural social settings, CA proceeds through the detailed transcription of those data and the
analysis of turn design, turn-taking and the collaborative construction of actions, to investigate
the organisation and orderliness of human social life through the prism of talk in naturally-
occurring behaviour.
Finally, then, if we were to try and summarise the basic ideas behind CA as an approach
to social research, it could be in terms of the following five statements:
1. Talk is not just forming words or conveying information; it is a vehicle for
accomplishing social actions.
2. Talk is produced in specific interactional contexts, and how people talk is highly
sensitive to its interactional context.
3. Talk-in-interaction is orderly; that is, wherever we look we can find systematic patterns
and structures in the ways that people use talk to interact.
4. Talk is organised sequentially; that is, by focusing on how people take turns at talking
we can understand how they interpret the immediate interactional context, since turns are
systematically related together.
5. The best way to analyse this is by looking at recordings of naturally-occurring
interaction, rather than using fieldnotes, as in ethnography, or intuition, as in many kinds
of linguistics.
Appendix: Transcription Conventions
Underline Underlining of a word or part of a word indicates speaker emphasis.

18
(0.5) Numbers in brackets indicate a gap timed in tenths of a second.
(.) A dot enclosed in brackets indicates a ‘micropause’ of less than one
tenth of a second.
= Equals signs are used to indicate ‘latching’ or absolute contiguity
between utterances; or to show the continuation of a speaker’s
utterance across intervening lines of transcript.
[ ] Square brackets between adjacent lines of concurrent speech indicate
the points of onset and cessation of overlapping talk.
(( )) Double brackets are used to describe a non-verbal activity: for
example ((banging sound)). They are also used to enclose the
transcriber’s comments on contextual or other relevant features.
( ) Empty brackets indicate the presence of an unclear utterance or other
sound on the tape.
.hhh h’s preceded by a dot are used to represent audible inward breathing.
The more h’s, the longer the breath.
hhhh h’s with no preceding dot are used in the same way to represent
outward breathing.
huh Laughter is transcribed using ‘laugh tokens’ which, as far as the
heh transcriber is able, represent the individual sounds that speakers make
hih while laughing.
stre:::tch Colons indicate the stretching of a sound at the preceding lexical item.
The more colons the greater the extent of the stretching.
wor- A dash indicates a sudden cut-off of the word being uttered.
Punctuation marks are not used grammatically, but to indicate
. prosodic aspects of the talk. A full stop indicates a falling
, tone; commas indicate fall-rise or rise-fall (i.e. a ‘continuing’
? tone); question marks indicate a marked rising tone.
!¯ Upward and downward arrows are used to mark an overall rise or fall
in pitch across a phrase.
a: Underlining of a letter before a colon indicates a small drop in pitch
during a word.
a: Underlining of a colon after a letter indicates a small rise in pitch at
that point in the word.
CAPITALS Capitals mark a section of speech markedly louder than that
surrounding it.
® Arrows in the left margin point to specific parts of the transcript
discussed in the text.
° ° Degree signs are used to indicate that the talk between them is
noticeably quieter than surrounding talk.
< > Outward chevrons are used to indicate that the talk between them is
noticeably slower than surrounding talk.
> < Inner chevrons are used to indicate that the talk between them is
noticeably quicker than surrounding talk.

19
References
Arminen, Ilkka. 2005. Institutional Interaction. London: Ashgate.
Atkinson, J. Maxwell and Paul Drew. 1979. Order in Court: The Organisation of Verbal
Interaction in Judicial Settings. London: Macmillan.
Atkinson, J. Maxwell and John Heritage, 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: Clarendon Press.
Bateson, Gregory. 1956. The message 'This is play'. In Bertram Schaffner (ed), Group
Processes: Transactions of the Second Conference, pp. 145-242. New York: Josiah Macy
Jr. Foundation.
Bateson, Gregory and Margaret Mead. 1942. Balinese Character: A Photographic Analysis. New
York: New York Academy of Sciences.
Birdwhistell, Ray. 1952. Introduction to Kinesics: An Annotation System for the Analysis of Body
Motion and Gesture. Washington DC: Foreign Services Institute, US Department of State.
Boden, Deirdre and Don Zimmerman, eds. 1991. Talk and Social Structure. Cambridge: Polity
Press.
Clark, Herbert H. 1992. Arenas of Language Use. Chicago: University of Chicago Press.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. The Hague: Mouton.
Davidson, Judy. 1984. Subsequent versions of invitations, offers, requests, and proposals dealing
with potential or actual rejection. In J.M. Atkinson and John Heritage (eds), Structures of
Social Action: Studies in Conversation Analysis, pp. 102-128. Cambridge: Cambridge
University Press.
Drew, Paul. 1992. Contested evidence in courtroom cross-examination: The case of a trial for
rape. In Paul Drew and John Heritage (eds), Talk At Work: Interaction in Institutional
Settings, pp. 470-520. Cambridge: Cambridge University Press.
Drew, Paul and John Heritage, eds. 1992. Talk at Work: Interaction in Institutional Settings.
Cambridge: Cambridge University Press.
Drew, Paul and John Heritage, eds. 2007. Conversation Analysis (4 Volumes). London: Sage.
Fairclough, Norman. 1995. Media Discourse. London: Edward Arnold.
Frankel, Richard. 1984. From sentence to sequence: Understanding the medical encounter
through microinteractional analysis. Discourse Processes, 7 (1): 135-170.
Garfinkel, Harold. 1967. Studies in Ethnomethodology. Englewood Cliffs: Prentice Hall.
Garfinkel, Harold and Harvey Sacks. 1970. On formal structures of practical actions. In John C.
McKinney and Edward A. Tiryakian (eds), Theoretical Sociology, pp. 338-366. New York:
Appleton-Century-Crofts.
Goffman, Erving. 1956. The Presentation of Self in Everyday Life. New York: Random House.
Heath, Christian. 1986. Body Movement and Speech in Medical Interaction Cambridge:
Cambridge University Press.
Heath, Christian. 1992. The delivery and reception of diagnosis in the general practice
consultation. In Paul Drew and John Heritage (eds), Talk At Work: Interaction in
Institutional Settings, pp. 235-267. Cambridge: Cambridge University Press.
Hopper, Robert. 1992. Telephone Conversation. Bloomington: Indiana University Press.
Hutchby, Ian. 1996. Confrontation Talk: Arguments, Asymmetries and Power on Talk Radio.
Mahwah NJ: Lawrence Erlbaum Associates.
Jefferson, Gail. 1989. Notes on a possible metric which provides for a 'standard maximum
silence' of one second in conversation. In Derek Roger and Peter Bull (eds), Conversation,
pp. 166-196. Clevedon: Multilingual Matters.

20
Jefferson, Gail. 2004. Glossary of transcript symbols with an introduction. In Gene Lerner (ed),
Conversation Analysis: Studies from the First Generation, pp. 13-34. Amsterdam: John
Benjamins.
Kendon, Adam. 1982. The organisation of behaviour in face-to-face interaction: Observations on
the development of a methodology. In Klaus R. Scherer and Paul Ekman (eds), Handbook
of Methods in Nonverbal Behaviour Research, pp. 440-505. Cambridge: Cambridge
University Press.
Kendon, Adam. 1990. Conducting Interaction. Cambridge: Cambridge University Press.
Lerner, Gene, ed.. 2004. Conversation Analysis: Studies from the First Generation. Amsterdam:
John Benjamins.
Levinson, Steven. 1983. Pragmatics. Cambridge: Cambridge University Press.
Maynard, Douglas and John Heritage, eds. 2004. Interaction and Medical Practice. Cambridge:
Cambridge University Press.
Ochs, Elinor. 1979. Transcription as theory. In Elinor Ochs and Bambi Schieffelin (eds),
Developmental Pragmatics, pp. 43-72. New York: Academic Press.
Pomerantz, Anita. 1984. Agreeing and disagreeing with assessments: Some features of
preferred/dispreferred turn-shapes. In J.M. Atkinson and John Heritage (eds), Structures of
Social Action: Studies in Conversation Analysis, pp. 79-112. Cambridge: Cambridge
University Press.
Richards, Keith and Paul Seedhouse, eds.. 2005. Applying Conversation Analysis. Basingstoke:
Palgrave Macmillan.
Sacks, Harvey. 1963. Sociological description. Berkeley Journal of Sociology, 8 (1): 1-16.
Sacks, Harvey. 1972. An initial investigation of the usability of conversational data for doing
sociology. In David Sudnow (ed), Studies in Social Interaction, pp. 31-74. New York: Free
Press.
Sacks, Harvey. 1984. Notes on methodology. In J.M. Atkinson and John Heritage (ed),
Structures of Social Action: Studies in Conversation Analysis, pp. 21-27. Cambridge:
Cambridge University Press.
Sacks, Harvey. 1987. On the preferences for agreement and contiguity in sequences in
conversation. In Graham Button and John R.E. Lee (eds), Talk and Social Organisation,
pp. 54-69. Clevedon: Multilingual Matters.
Sacks, Harvey. 1992. Lectures on Conversation. Oxford: Blackwell.
Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson. 1974. A simplest systematics for the
organisation of turn-taking for conversation. Language, 50 (4): 696-735.
Scheflen, Albert and Alice Scheflen. 1972. Body Language and the Social Order. Englewood
Cliffs: Prenctic Hall.
Schegloff, Emanuel A. 1968. Sequencing in conversational openings. American Anthropologist,
70 (6): 1075-1095.
Schegloff, Emanuel A. 1988. Presequences and indirection: Applying speech act theory to
ordinary conversation. Journal of Pragmatics, 12 (1): 55-62.
Schegloff, Emanuel A. 1997. Whose text, whose context? Discourse and Society, 8 (2): 165-187.
Schegloff, Emanuel A. 2006. Sequence Organisation in Interaction. Cambridge: Cambridge
University Press.
Schegloff, Emanuel A. and Harvey Sacks. 1973. Opening up closings. Semiotica, 7 (3): 289-327.
Schegloff, Emanuel A., Gail Jefferson and Harvey Sacks. 1975. The preference for self-
correction in the organisation of repair in conversation. Language, 53 (3): 361-382.
Terasaki, Alene. 2004. Pre-announcement sequences in conversation. In Gene Lerner (ed),
Conversation Analysis: Studies from the First Generation, pp. 171-224. Amsterdam: John

21
Benjamins.
Thompson, John. 1984. Studies in the Theory of Ideology. Cambridge: Polity Press.
Wittgenstein, Ludwig. 1953. Philosophical Investigations. Oxford: Oxford University Press.