REVIEW ESSAY

Conversation Analysis: A Quest for Order in

Social Interaction and Language Use

Ilkka Arminen
Finnish Foundation for Alcohol Studies, Helsinki, Finland

Paul ten Have: Doing Conversation Analysis - A it a worthy human science (ten Have
Practical Guide. (London: Sage, 1999). 1999:196-197). That is, a conversation cannot
be represented with a closed set of formal rules
Ian Hutchby and Robin Wooffitt: Conversation that would allow an infallible prediction of the
Analysis. Principles, Practices and Applications. next possible conversational move, or the set of
(Cambridge: Polity Press, 1998). next possible moves. Instead, every next con-
David Silverman: Harvey Sacks - Social Science versational move renews our understanding of
and Conversation Analysis. (Cambridge: Polity the priormove, so that each turn at talk orients
Press, 1998). to a preceding context, but also recreates the
context anew (Heritage 1984:242). Therefore,
Conversation Analysis (CA) has developed over
the course of the last 30 years. Its objective is to
a purely formal context-free description of a
conversation remains impossible. Instead, ana-
describe the competencies and procedures
involved in the production of any type of social lysis of conversation involves discerning parti-
cipants’ intersubjective understanding of the
interaction. In this essay I will discuss the state
of conversation as it evolves moment by
course
of the art of this enterprise and its relationship
moment during their orientation to the social
with the human sciences, drawing mainly from
a number of recent overviews of the field
action being accomplished. Consequently, as
tiny as the details of a conversation may be, they
(namely, ten Have 1999; Hutchby & Wooffitt are the building blocks of the architecture of
1998; Silverman 1998). My aim is to address
the working principles of the discipline and to intersubjectivity upon which the accomplish-
ment of social actions, simple and complex,
discuss their potential relevance for the human
rests.
sciences, including some disputes concerning
the applicability of CA.
The distinctiveness of CA as a social
scientific approach derives from its object of Basic ideas
analysis. CA treats talk and social interaction as
a sufficient object for analysis, rather than as a The basic idea of CA is so simple that it is
window to wider social processes or as a difficult to grasp. CA studies what an utterance
medium for data collection (Hutchby & Wooffitt does in relation to the preceding one(s) and the
1998:21). That CAdata collection methods implications it poses for the next one(s). As
rely on the tape recording of actual interactions Hutchby and Wooffitt ( 19 9 8 :15 ) put it, the next-
emphasizes the role of social interaction as an turn proof procedure is the most basic tool in CA.
autonomous reality sui generis. Traditionally, That is, the next turn is taken as evidence of the
however, sociologists have not seen the study of party’s orientation to the prior turn, there and
talk in its own right as a relevant enterprise. An then. This methodical procedure is CAs gateway
honest but paradoxical defence would be that to the participants’ own understandings as they
the very fact that it is impossible ever to achieve are revealed during actual interaction, thereby
a strictly formal analysis of conversation makes providing material for an analytic explication.
252 ,

Let us now look at an example, a brief perspectives has opened. E has provided an
instance of an exchange between E and M assessment, the upgraded quality of which M
(transcription simplified). has made plain through her mitigated second
assessment.
E: That Pat isn’t she a doll?
But, areader may protest, is all this pure
M: Yeah isn’t she pretty.
speculation? Can we say anything about the
In making use of the next-turn proof procedure, validity of this reading? Maybe the writer has
we should be able to say something about E’s got it all wrong. Is there any way to test the
turn with the help of M’s turn. Let us start from accuracy of the analysis? Actually, CA allows
the obvious. M’s turn is designed as an answer, the testing of its findings through the very same
but a particular kind of an answer. It answers next-turn proof procedure (Heritage 1984 :256-
through an assessment. The basic strategy for a 257). We should take up the turn following M’s
CA enterprise is to make comparisons that can turn to see whether our explication of the
be either imaginary or empirical. Here, to work interaction fits with the participants’ sense of
empirically, we would collect parallel instances the ongoing interaction as they reveal it turn by
to find regularities through which the kinds of turn. We might even imagine the set of
actions are accomplished. But as this text is just alternatives that E would use to counter M’s
an illustration, and not a demonstration, we downgraded assessment. In this way, even if we
shall only use our imagination to sketch out the are not able to make infallible predictions of the

meaning of an actual course of interaction next turns, we can give a reasonable description
through comparisons with imaginary cases. of the course of conversation and of potential
Note here that M continues her answer after next moves. Of course, proper empirical
the response token ’yeah’ and in so doing she research would be based on a collection of
treats her ’yeah’ as an insufficient answer to the cases, of which the analysis should reveal
kind of action E has produced. invariable regularity. Here we have the chance
At this stage the reader should be able to to practise our skills simply by imagining how
detect a hermeneutic circle. The next-turn proof the exchange will continue, and then looking at
procedure means that a reflexive relationship the extended sequence of the exchange (see ten
exists between adjacent turns. The next turn is Have 1999:4). If you feel that you have a better
used as an analytic resource for making sense of account of the exchange above, please, feel free
the prior turn, which, for its part, has provided to develop it further, and then check it against
the sequential implications that have made the empirical reality. Note also that this kind of an
next one relevant. Here M’s turn suggests that E analysis is not theory-bound. The analysis is not
has invited M to produce a second assessment. supposed to be measured against any theore-
In other words, despite its grammatical form, an tical account of interaction, but against the
assessment that is delivered through a yes-no reality of recorded interactions and their tran-
question format does not work like an ordinary scriptions.
question (or, at least, it is not treated so by M). M In their book, Hutchby and Wooffitt make a
does not treat E’s utterance as a straightforward useful analytic distinction between ’sequential
question, but as an invitation to assess the order’ and what they call ’inferential order’
person that E herself has described as a doll. We (1998:38-39), although sequential and infer-
could also note that M’s assessment is of a ential order presuppose one another. That is, the
specific kind, in comparison to E’s prior assess- participants’ inferential work allows them to
ment. M’s assessment is weaker and narrower build sequences of action upon which the
than E’s assessment. It is a downgraded assess- inferential work rests. The sequential order
ment, which suggests that M does not agree means the ’describable ways in which turns

very strongly with E. Now, when we are about to are linked together into definite sequences’
close our analysis (at least for a moment), we (1998:38); its analysis is CAbackbone. How-
are on the verge of making some sociological/ ever, it is tied to the inferential order that I
socio-psychological discoveries. The situation is would define as ’the kinds of implications and
rather juicy. E and M are talking about a third inferences participants draw about each other’s
party, Pat, and a particular type of relationship talk and conduct to make sense of it and to hold
is emerging between E and M. E has invited M to each other morally accountable’. In the final
join her in a joint display of their appreciation of instance, the inferential order is the basis for
Pat. However, M has declined the invitation everyday semiotics. To give an example, Viola’s
with a mild response, and a gulf between their encounter with Olivia in Shakespeare’s
 253

’Twelfth-Night: Or, What you Will’ reveals the 103), and left aside Viola’s and Olivia’s emerg-
way in which parties become accountable (i.e. ing relationship.
moral subjects) through the inferences their CA actually demands a disciplined
sequential conduct makes possible. Olivia’s approach that does not jump to conclusions,
distracted speech exposes to Viola (disguised in does not fall into immature theoretical specula-
men’s dress) that she has fallen in love with her: tions or rely on everyday assumptions. Paul ten
’She made good view of me; indeed, so much, Have (1999:107) goes so far as to make a
That, methought, her eyes had lost her tongue, distinction between ’pure’ and ’applied’ CA, and
For she did speak in starts distractedly. She loves wants the former to concentrate ’on talk
me, sure...’. &dquo;itself&dquo;, rather than its &dquo;context&dquo;’. To my
This distinction between sequential and mind, this strict division and the whole notion
inferential ’orders’ reveals the multidimen- of ’pure’ CA is misleading and inadvisable.
sionality of CA’s research object. Essentially, Moreover, the separation of talk from its context
CA is about the organization of interaction, is against all the basic tenets of CA, according to
that is, about the syntactic, semantic, and which the context-renewing properties of talk
prosodic qualities through which turns are amount to the endogenous construction of
designed, but also about the pragmatic context, as parties orient to the ’context’
connections through which turns are inter- through the management of talk-in-interaction
locked. Further, as Hutchby and Wooffitt as an observable part of performing social
stress (1998:39), these concerns interplay actions within that ’context’. A more sensible
with normative and inferential properties of way to address the issue of the applicability of
talk through which participants orient them- CA is to stress that CA allows, and even
selves to the sense and implications of their necessitates, selection of the focus of analysis,
interaction. The multilayered orderliness of which may be more closely connected to
talk makes it a ’deep’ object, so that a sequential or inferential properties of talk.
property of talk that looks innocent and
insignificant may be relevant for considera-
tion from another angle. CAs programmatic Origins and development
stance suggests that we should not a priori
assume the irrelevance of any detail of talk; All the recent overviews under consideration
instead, we should try to find order at all here give an account of the origins of CA, and
points (Hutchby & Wooffitt 1998:17-22). This even of the very moment of its discovery. In

methodological canon provides opportunity particular, David Silverman (1998) offers the
for unlimited new findings, but also makes first book-length introduction to Harvey Sacks’
the research process a never-ending quest. We iconoclastic thought experiments that, among
may think of CA as the reverse engineering of other things, paved the way for CA. Originally,
an immense complex of intersubjectivity. To Sacks studied law at Columbia and Yale in the
decipher this enigmatic structure also requires 1950s. Through his encounters with Erving
expert skill and craftsmanship from the Goffman and Harold Garfinkel, Sacks became
analyst for observing, detailing, describing involved in studies of sociology. Although both
and systematizing this fractal-like multitude. Goffman and Garfinkel were important teachers
On an analytical level it may be helpful to of face-to-face behaviour and everyday reason-
distinguish different styles of performing CA. ing processes, gradually Goffman’s and Sacks’
The analysis may focus on the sequential order, ways parted. Sacks obtained his PhD only after
and only passing remarks may be made about Goffman agreed to resign from his evaluation
the inferential properties of talk. For instance, committee in 1966 (Silverman 1998:28). Goff-
we could have concentrated on the properties of man found Sacks’ reasoning circular and
E’s and M’s turn design and on the relationship asociological. Although Goffman later became
between turns, and bypassed the potential social more interested in the emerging CA, he never
implications their exchange may have had; or did quite accept/understand it (depending on
we could have analysed the properties of Olivia’s the stance we take vis-a-vis Goffman and Sachs.
distracted speech, such as breaths and other In 1963 Sacks was working with Garfinkel
aspirations including laughter and laugh at the Center for the Scientific Study of Suicide
tokens, recognizable contexted silences, coughs, at UCLA. For their study, a set of calls to a
’y’knows’, ’uh’ in all its varieties, cut-offs, re- suicide prevention center was recorded. One
beginnings, re-directions, etc. (Schegloff 1996: task for center personnel was to try to obtain the
254

caller’s name. In most cases, the call takers were turn taking in conversation by Sacks, Schegloff,
successful in getting the caller’s name by giving and Jefferson in 19 74.
their name first. But then Sacks came across Sacks was tragically killed in a car accident
one call opening (see Silverman 1998:98-99; in 1975. However, developments in CA con-
Hutchby & Wooffitt 1998:18-20; ten Have tinued, one of which emerged in the late 19 70s
1999:13-15) which ran like this (transcription in the form of studies on interactions in
simplified): institutional settings.
A: This is Mr. Smith, may I help you?
B: I can’t hear you.
A: This is Mr. Smith. Applications
B: Smith.
As pointed out in all three recent overviews, the
In this case, the caller, B, has a hearing problem 19 74 paper on turn taking by Sacks, Schegloff,
that leads to a trajectory in which the place for a and Jefferson also prompted the idea of com-
reciprocal giving of names never materializes. parative studies on different systems of turn
Instead, after the caller’s hearing problem has taking. Roughly, turn taking allocates speaking
been solved, as above, the call taker is implied to time between parties so as to contribute towards
acknowledge the caller’s hearing with an item, the organization of the activity to which parties
such as ’yes’, and then to return to the opening are oriented. In ’ordinary conversations’, turn
of activity ’may I help you’. (Unfortunately, taking is coordinated turn-by-turn, locally, with
Sacks did not show how this call went on, but no pre-designed arrangements. By contrast, in
the course described above seems to be the ’formal speech-exchange systems’ turn alloca-
regular pattern; see Sacks 1992, Vol. 1, pp. 7- tion is based on a pre-designed order, an
76.) Moreover, as the call takers, out of learned institutional order in which parties are oriented
professional caution, avoided directly asking the to performing institutional tasks, as demon-
callers’ names, the hearing problem, such as strated by their commitment to that pre-
above, pre-empted the chance to get the caller’s designed turn order. Thus, the analysis of talk-
name. in-interaction in institutional settings such as
Thus, Sacks encountered a puzzle. Was this courtrooms, classrooms, and ceremonies aims
trajectory just a plain accident, or was there at specifying the very format through which the
something more in it? At this point, what he felt institutional practice is talked into being. Later
to be a ’wild’ possibility occurred to him. Rather studies on institutional interaction, however,
than as a string of propositions, could talk be have mostly focused on ’quasiconversational’
analysed as being composed of methodical ways institutional interaction, the specificity of which
of doing things at this level of detail? Could talk does not lie in formal turn taking, but in factors
be reduced to a set of methods and procedures such as turn design, lexical choices, sequence
through which given tasks were performed? In organization, etc. (ten Have 1999:168).
this example the hearing problem had been a A collection of key articles on institutional
methodical way to ’avoid giving a name without interaction, with a thorough, analytic introduc-
refusing to do so’. After this discovery, Sacks tion, edited by Drew and Heritage (1992),
started to apply the new methodology to his set provides a systematic exploration of a distinctive
of materials. His lectures from 1964 (published field. Regarding institutional interaction, the
in 1992, Vol. 1) allow us to follow his reverse engineering program of CA aims at
meditations. He studied problems such as ’how identifying the unique ’fingerprint’ of each
to get someone’s name without asking for it’ institutional practice. Notably, this fingerprint
(give yours), ’how to avoid giving help without is not the outcome of analysis, but its starting
refusing to give it’ (treat the circumstance as a point, with the help of which specific institu-
joke), ’how to get help for suicidal tendencies tional tasks, identities, and inferential proce-
without requesting it’ (ask ’how does this dures are opened for research. The analysis of
organization work’), etc. institutional interaction ultimately seeks com-
It took about 10 years for the key ideas of plex issues, such as strategic aspects of interac-
CA to be crystallized. Much of the early tion, the achievement of collaboration, or
development took place through collaboration procedures whereby participants’ differing per-
among Gail Jefferson, Emanuel Schegloff, and spectives are brought into alignment. In this
Harvey Sacks. A series of early CA papers respect, studies on institutional interaction are
culminated in the publication of a paper on very close to Sacks’ original idea to study
 255

members’ methodical ways of accomplishing disciplines. One such new field is so-called
social tasks in interaction. ’discursive social psychology’, a version of
Incidentally, or emerging from the inevi- discourse analysis (Hutchby & Wooffitt
table logic of scientific and social progress all 1998:202-228). This field addresses traditional
three recent overviews introduce the same new cognitive and epistemological concerns and
areas of institutional interaction. One new type discerns their interactional basis. Research foci
is the study of human-computer interactions include how the factuality of statements is
(HCIs). In a nutshell, the study of HCIs focuses interactionally designed, or what techniques
on the relationship between the design assump- speakers use to pre-empt a recipient’s scepti-
tions of the machine and user assumptions. cism. In addition, a new relationship has
These studies may be informed by the practical developed between linguistics and CA, as briefly
concerns of either to clarify and solve problems touched upon by ten Have (1999:199). A
in the user-machine interface, or to explicate sociologist may find the suggestion funny that
the properties of agent-client interaction for CA and linguistics intersect, as the whole CA
developing computer systems to replace the enterprise may look rather linguistic. But as I
agent (Hutchby & Wooffitt 1998:245-252). have tried to show, CAs perspective and set of
Another new area for research is the methods may be applied to various dimensions
survey interview interaction (ten Have of interaction, from social actions to syntax and
1999:170-180; Hutchby & Wooffitt prosody. Linguistically oriented CA analyses the
1998:173-178). The focus of this interaction intersection of grammar and interaction (Ochs
is the need to meet the requirements of both et al. 1996). It may address the ways in which
standardization and local interactional contin- grammar contributes towards the organization
gencies. These latter involve many types of of social interaction, but also the ways in which
problems, starting with understanding difficul- social interaction organizes grammar. We may
ties. The interviewee may display an inability to think of it both ways: forms of social interaction
understand a question as phrased in the survey, have gained their grammatical representations,
thereby inviting the interviewer to depart from grammar has been shaped by interaction, but
the standardized phrasing of the question. On grammar is also an inherent aspect of the
other occasions, the interviewee may depart organization of interaction, its mode. Finally,
from the scripted answering schemes and phonetics is also open to CA treatment. On the
produce answers that address new dimensions, one hand, prosody contextualizes and gives
or provide more or less elaboration than the interpretative cues for identifying interactional
standardized options allow. In both these cases, moves; on the other hand, sequential order
the interviewer may need to infer how to code contextualizes prosody and generates tasks for
the answers or negotiate with the interviewee. parties in designing the prosodic quality of their
The interviewee’s potential resistance poses a turns as being appropriate for the sequential
further set of dilemmas. In all, the analysis of position to which they are orienting themselves
the interactional accomplishment of survey (Couper-Kuhlen & Selting 1996).
interviews is a potentially fruitful methodologi-
cal field that may be applied to the development
of survey instruments. Hutchby and Wooffitt Divisions
(1998:178-197) also address the interactional
nature of semistructured or unstructured inter- In comparison to many social scientific schools
views. of thought, CA is a unified enterprise. Its
The third new area for CA is a focus on the uniformity starts with its methodology and
communication problems of ’deficient’ speakers, working principles. In CA, research focuses on
such as aphasics (ten Have 1999:189-192; the instances of actual behaviour that can be
Hutchby & Wooffitt 1998:252-257). As CAs observed and recognized intersubjectively, in
general aim is to describe speakers’ competen- contrast to ideal types, generalizations, and
cies, it may also help to define more exactly the averages that demand interpretive work, which
types of communication problems that deficien- makes the relationship between the finding and
cies pose in actual communication. The the investigated state of affairs indeterminate.
enhanced understanding of such communica- Consequently, most social scientific findings do
tion problems may open new therapeutic ways not cumulate; they are fuzzy, thereby preventing
to encounter clients. elaboration of their exact felicity conditions
CA applications also extend to various (Heritage 1984:231-236). The working princi-
256

ples of CA are bound to its research object. CA ability to recognize the activity types of partici-
analysis proceeds case-by-case and establishes pants’ turns of talk. As long as the analyst is a
observable patterns inductively Its analytical competent member of the culture studied, the
methods amount to an institutional form of co- ability to recognize social actions in interaction
operation, so-called ’data sessions’ (ten Have does not pose any practical problem, although
1999:102-104, 123-125). In a data session, a we may dispute the role of common-sense
group of CA workers attends to its material with knowledge in CA in analytic terms (Hutchby &
no pre-set objectives, so that interesting things Wooffitt 1998:112-113). A lack of ethno-
may simply be noticed. Then a focus line is graphic understanding may turn into a hin-
chosen from the set of noticed things, and the drance if the analyst studies foreign cultures,
major part of the work consists of collaborative isolated subcultures, or institutionally distinct
elaboration of patterns found. The data sessions settings. If the recognizability of participants’
provide an arena for educating prospective actions is not pre-given by shared cultural
researchers, but are also brainstorming occa- knowledge, ethnographic background knowl-
sions for any CA worker. Thus, the uniformity of edge may become indispensable. However, any
CA is maintained through its working style. ethnographic knowledge is always only a
CA’s distinctiveness and uniformity can be starting point for the discernment of partici-
emphasized on good grounds. Very few pants’ methodical design of their social actions.
accounts of CA have brought up its diversity. CA’s aim is not to go beyond participants’
Nevertheless, for analytic purposes it is useful to actions to recategorize them, but to reverse
stress the multidimensionality of CAresearch engineer participants’ machinery for perform-
object. As already discussed, CA studies may ing social actions (ten Have 1999:53-60).
focus on various layers of the architecture of The general line of argument as regards
interactional intersubjectivity. On a more subtle quantification proceeds in parallel. CA aims at
level, CA workers have individual styles and establishing a unique basis for its findings so
distinct methodological solutions. Of the recent that they reveal participants’ orientation and
overviews, ten Have’s merits distinction for the methodical course of action in every singular
openness of its methodological debates. The instance. Consequently, coding and quantifica-
methodological bifurcation points are worth tion are problematic, as they may lead the
consideration, as they single out fields that need analyst away from a sufficiently close analysis of
further development, thereby indicating new the data (Hutchby & Wooffitt 1998:115-119).
directions for research. Of the potentially However, there are some notable exceptions (ten
disputable issues, ten Have takes up CA’s Have 1999:144-148). For instance, research
relation to ethnography and quantification. may focus on a compound unit composed of
Generally, CAs findings are indifferent to several identifiable elements, the relationships of
ethnographic data. Through its own way of which may be too complex to be described
working with the data, CA reconstitutes and comprehensively, but the aggregate level out-
specifies participants’ ongoing orientation to come of which may be an interesting phenom-
interaction, thereby demonstrating their endo- enon in its own right.
genous construction of context. That is, Interestingly, CAs
basic unit of analysis, a
through the detailed inspection of interaction turn of talk, behaves much this way. A
the analyst will reconstruct the sense of an conversational interaction depends on partici-
ongoing social action, and no reference to pants’ ability to recognize the completion of a
external data sources is needed. For instance, turn so that the turn taking can take place and
in our illustration we were able to point out the conversation proceed in an orchestrated
what was going on between E and M without manner. Ford and Thompson (1996) made an

any reference to ethnographic background analytic distinction among syntactic, intona-

knowledge. From an ethnomethodological tional, and pragmatic elements of talk, and
point of view, the basis of social order is not studied statistically the interrelationship of
hidden under the observable surface of partici- these three elements. They concluded that
pants’ conduct, but that very conduct, in its syntax alone does not project the end of a
intersubjective availability, forms the methodi- turn; it does so only when linked with intona-
cal basis for them to maintain and manage the tion and pragmatics. A statistical analysis can
understandability and orderliness of their even improve our understanding of the basic
actions. However, the analysis of a sequential units of talk. However, Ford and Thompson
course of action presupposes the analyst’s worked case-by-case to achieve intersubjectively
 257

valid coding, and used individual examples to may revolutionize a reader’s understanding of
demonstrate their findings. mundane matters. That was something Harvey
Finally, macroscopic entities can also be Sacks was very good at. As a pioneer, Sacks also
complex phenomena, the individual contingen- enjoyed a considerable degree of intellectual
cies of which may fall outside the interests of freedom. Since Sacks, CA has developed into a
analysis. For example, the way doctors design tight discipline with its own methodological
the delivery of a diagnosis is related to several canons. Some amount of openness to CAs
factors, such as the location of the diagnosis in internal heterogeneity might help it to fulfil its
the encounter, the opacity of the diagnostic promise in the human sciences, in both
information, the certainty of the diagnosis, and linguistics and the social sciences.
how controversial it is (Perakyla 1998).
Together these factors amount to too many
contingencies to be comprehensively described,
but regularity in doctors’ behaviour can still be
References
established through individual case demonstra- Couper-Kuhlen, E. & Selting, M. (eds.). 1996. Prosody in
tions and their statistical aggregation. Conversation - Interactional Studies. Cambridge: Cambridge
University Press.
Drew, P. & Heritage, J. (eds.). 1992. Talk at Work - Interaction in
Institutional Settings. Cambridge: Cambridge University Press.
Conclusion
Ford, C. & Thompson, S. 1996. Interactional Units in
Conversation: Syntactic, Intonational, and Pragmatic
Conversation Analysis is still a young enterprise. Resources for the Management of Turns. In E. Ochs, E.
It provides detailed and firm analysis of the Schegloff & S. Thompson (eds.), Interaction and Grammar, pp.
intersubjective meanings of social actions. 134-184. Cambridge: Cambridge University Press.
However, CA’s internal complexity is one of the Heritage, J. 1984. Garfinkel and Ethnomethodology. Oxford: Polity
Press.
obstacles that may hinder the fulfilment of its
Ochs, E., Schegloff, E. & Thompson, S. (eds.). 1996. Interaction
promise. Although these recent overviews of CA and Grammar. Cambridge: Cambridge University Press.
are useful introductions that make the enter- Peräkyla, A. 1998. Authority and Intersubjectivity: The
prise accessible to students and social scientists, Delivery of Diagnosis in Primary Health Care. Social
they also demonstrate how tremendous an Psychology Quarterly, 61, 301-320.
effort it is to learn the craft of CA. In particular, Sacks, H. 1992. Lectures on Conversation. 2 vols. Cambridge:
Blackwell.
ten Have, but also Hutchby and Wooffitt, are Sacks, H., Schegloff, E. & Jefferson, G. [1974]1978. A Simplest
much more at home with institutional interac- Systematics for the Organization of Turn Taking for
tion and less convincing in their analyses of Conversation. Language, 50, 696-735. Reprinted in J.

everyday talk. It is a demanding task to tackle Schenkein(ed.), Studies in the Organization of Conversational
Interaction, pp. 7-56. New York: Academic Press.
seemingly trivial pieces of interaction and turn Schegloff, E. 1996. Turn Organization: One Intersection of
them into meaningful statements about the Grammar and Interaction. In E. Ochs, E. Schegloff & S.
hidden rationalities and complexities of the Thompson (eds.), Interaction and Grammar, pp. 52-133.
procedural forms of human behaviour that Cambridge: Cambridge University Press.