I- The Ethnomethodological origin
II- Conversation analysis: views and tasks
III- The definitional problem.
Conversation analysis (C.A), in general, is characterized by a vast diversity of theoretical beliefs, research practices and findings. Sociologists, sociolinguists, social psychologists, anthropologists, discourse analysts, educators, and many others have come to recognize the relevance of conversation for their respective disciplines. Natural conversation has, thus, become a research site for different explorations, not only because social interaction is to a large extent verbal interaction but also because talk exhibits “orderly” features, which upon analysis, turn out not be features of language but features of interaction (see Streek 1980: 402).
More recently, it has become fashionable to explain the structural character of interpersonal verbal behaviour in terms of “common sense” knowledge and “intersubjectivity” (Garfinkel (1967, 1972), Schutz (1970)). Seen from this perspective, the fundamental presupposition of speaking is that when one speaks, one’s hearer can at least potentially understand what one is attempting to say (Tyler 1978:4). Conversational understanding is- in Brown and Levinson’s (1978) view- achieved by reconstruction of levels of intents beyond and above and integrative of those lying behind particular utterances or sentences (p.238).
In fact, without a commonsense thinking, defined as “a pragmatic attitude through which we manage our practical life and the fact that we obviously have things in common with other people”, Hizler and Keller (1989: 98) argues, “no intersubejective process can rise and no congruent subjectivities of the actors can come into play”. An important fact in everyday life is that one acquires/learns one’s social and linguistic competence as “an actor who is part of a larger group (negotiation, or innovative combination): ways of dressing, speaking, the rules of convenience often implied, acquired knowledge- in brief what is familiar” (Lalli (1989: 113)). These ways and rules provide substantive guidelines for the structural systematicity of behaviour in interactional situation. Given that language is the major means of social interaction, it is important to admit that it “has been acquired as an instrument for regulating joint activity and joint attention” (Brunner (1977:58)).Hence Conversation, as a language activity, is seen as part of this activity which involves joint attention and mutual collaboration
However, the primary requirement in the study of conversation is a precise and intersubjectively verifiable description of human action as behaviour which is meaningful to the actors. In this respect, Luckman (1989: 18) reports that the first step in satisfying this requirement must be the identification and reconstruction of the variety of ways in which actions are typically meaningful to the actors in human history.
Departing from the fact that interactional patterns are rule-governed and norm-following, the knowledge of these rules and norms is reciprocal but “taken-granted” (Garfinkel (1967) since attribution of “intentionality to one another on the part of social actors is an assumption commonly held in interpersonal communication. For Maffesoli (1989: 1) the social world of which these rules are constitutive and which is “taken-for-granted” can be understood as the product of permanent interaction, of a constant reversibility between the various elements of the social environment, inside the matrix of the natural environment. In other words, social interactions are seen as particular instances invoking the use of these rules and at the same time manifestation of this “taken-for-granted” knowledge.
I- The Ethnomethodological origin
Hopefully, the discussion to this point should make it clear that it is the discovery of the organization of ordinary activities such as conversation which is central to conversation analysis. Such a discovery has been also the goal of Ethnomethodological research.
Garfinkel (1967:74) , one of the first ethnomethodologists, argues that social knowledge cannot be adequately characterized in the form of statistically countable, abstract categories such as scalar ratings of roles, status, or personality characteristics (which are typical of conventional sociology and social psychology). Social knowledge is revealed in the process of interaction and the social world is created by interactants through the way they behave. For Benson and Hughes (1983:61) Garfunkel’s concern is “the study of “ethnic” (i,e participants’ own) methods of production and interpretation of social interaction”.
These “ethnic methods” seem to be part of the Hymesian communicative competence which includes knowledge of linguistic and related communicative conventions that speakers must have to create and sustain conversational cooperation (See also Gumperz (1982:209). This knowledge, accordingly, is to be inferred /induced from “naturally occurring conversations” whose importance is evidenced by “the grossly apparent facts” (Sacks, Schegloff and Jefferson (1974)). In this perspective Goffman (1975) emphasizes that “…(informal) talk is like a midden, a refuse heap in which bits and oddments of all the ways framing activities in the culture are to be found”.(498-499).
The focus on conversation as the simplest instance of a naturally organised activity is justified by its structural organisation which reflects the members’ methods in interaction. Ethnomethodological conversation analyses by Sacks and Schegloff and Jefferson (1974), Garfinkel and Sacks (1970), Turner (1974), Sudnow (1978) have produced evidence for the existence of a variety of structures in terms of which conversational interaction functions, as “a reaction within sociology to shortcomings resulting from the premature imposition of inadequate theoretical categories and frameworks on complex data” ( heritage (1989: 33)). Ethnomethodological conversation analysis emerged, thus, from Ethnomethodological observations about procedures of analysis within conventional sociology, and the crux of these observations centred on categorization practices (on this point, see Wooton (1982)).
II- Conversation analysis: views and tasks
Conversation analysis is the most productive and prolific form of analysis which has been developed with Ethnomethodological concerns in mind. It is primarily interested in tracing and describing “the locally managed structures of conversation” and not “in using interaction as a basis for making claims as to the development of attributes among participants” (Wooton (1981:166). Conversation analysis, for Bilmes (1988:161) aims to provide a model of analysis that is neither statistical nor inter-psychological. Instead, it is structural, done by reference to contextual features, especially sequencing, and to conventional understandings and procedures. It looks for mechanisms that produce and explain behaviour, but for social rather than psychic mechanisms.
Therefore, conversation is characterized by fundamental structural mechanisms of turn-taking (Sacks, Schegloff and Jefferson (1974, 1978); Duncan (1974, 1985), Sudnow (1972) regulating joint action, by preference organisation (Pomerantz (1984 b), Bilmes( 1988)); by adjacent positioning of turns at talk (Schegloff (1972)); by sequence of silence (Jefferson (1989, Edelsky (1981); by an enormous variety of prosodic and Kinetic features (Labov and Fanshel (1977), Argyle (1983), (1981));and interruptive behaviour (Zimmerman and west (1975), Schegloff (1977,1988)).
These structural mechanisms, however, are not analysts’ categories imposed on the data, but rather categories emerging from the structural details of conversational data. The conversation analyst, then, does not have to worry about imposing his analysis on the conversational data, for conversation, itself, to use Taylor and Cameron’s words, “wears its own inherent analysis on its own sleeve” (see Taylor and Cameron (1987)).
Additionally, however, the basic outlook of C.A is based on three fundamental assumptions:
First assumption: interaction is structurally organised
Second assumption: contributions to interaction are contextually oriented.
Third assumption: these two properties are inherent in the details of interaction so that no detail can be dismissed, a priori, as disorderly, accidental or irrelevant (Heritage (1984), Bull and Roger (1989)).
In relation to the first assumption, Heritage (1984) comments that “all aspects of social action and interaction can be found to exhibit organized patterns of stable, identifiable structural features. These organisations are to be treated as structural in their own right and social in character”(p.241). Also, the contributions to conversation in the form of adjacency pairs, which exhibit dialogic structure, are contextually oriented. That is, they are both “context-shaped” because they cannot be fully understood without reference to the context in which they occur, and context-renewing because they form the immediate context for subsequent ones.
Following these assumption, C.A should focus on conversation as a significant and engrossing activity in its own right. Interest in conversation as “raw material” for analysis makes of conversation analysis a sociology of everyday activities, which, in contrast to “conventional sociological interest in “wind”, personality and social structures as the product of interaction, looks upon talk as one place where these are located and revealed (see Psathas and Waksler (1989), Garfinkel and Sacks (1970), Sacks (1972), Garfinkel (1972)).The concept of “ structural organization”, an interactional fact, has induced analysts to use a structural approach to display this organisation, which is a methodological virtue dispensing, therefore, with motivational speculation and psychological characteristics of the particular participants (for further discussion, see the above references, also roger 1989:3).
The whole conversation analytic enterprise is designed to uncover not particular “action” (Searle (1969, 1975), Austin (1962)), within interaction, but sequences of behaviour which form conversation, considered as an “achievement” (Schegloff (1982)) and its content is not as interesting as the character of the work that participants do to maintain or alter it. It is this fact which licenses C.A to be strongly “data-driven”, studying phenomena which are inherent in the conversational data itself. Treating talk in conversation as an analytic resource has changed some of the current sociological theories of language by avoiding premature theory construction in favour of an empirical approach to the study of conversation.
Even more, C.A ,unlike other disciplines such as Speech act theory and discourse analysis, avoids recourse to intuitions of well-formedness or deviance – though the occurrence of repairs or substantial pauses is often treated as a kind of deviance or evidence of a breakdown in the smooth functioning of the rules and hence a testimony to them (see Dillon, Coleman, Fahnestock and Agar (1985: 447)).Conversation Analysts do not seek to explicate members’ intuition, and therefore, explore the structural details of conversation for the intuitions of member’s conversationalists are “foreign”. Jefferson and Schenkein (1978:160) state that in C.A we are “exploring certain organizational features of conversational interaction that may escape the scrutiny of ordinary intuitions… we are, in the end, examining the prospects of a “non-intuitive analytic mentality for investigating and describing organizational details of conversational interaction, not only thus far undescribed but thus far unnoticed as resources for conversationalists”.