The article begins with a sketch of the relation of interaction to language and to culture, and of the students of interaction to the students of language and of culture. A 10-second segment of recorded interaction at a family dinner is then examined in a fashion meant to preserve the integrity1 of what is being done interactionally while incorporating attention to the deployment of various facets of the language that is used, and its relationship to simultaneously ongoing bodily doings. An (...) interactional practice – whining – from that episode is then juxtaposed with the same practice in several other segments of interaction in the interests of developing a more formal, transsituational account. The viability of research focused on phenomena in an analytically distinct domain of events while preserving the integrity of the occasions in which instances of the phenomenon occurred is then reviewed, using a case study of the conjoint use of phonetic analysis and conversation analysis. The article concludes with a reply to Levinson's article in this special issue of the journal, and uses the occasion to sketch the relationship between interaction and so-called ‘macro’ social and cultural formations such as kinship. (shrink)

The article begins with an effort to clarify and differentiate a variety of terms used by analysts in dealing with mentions of persons in conversation and other forms of talk-in-interaction — such terms as person-reference, identifying, describing, categorizing, and the like. This effort leads to the observation that `reference to persons' and `membership categorization' are quite distinct sets of practices, with most reference to persons not being done by membership categories, and most uses of membership categorization devices being in the (...) service of actions other than referring. Two interactional sequences whose analysis turns on a connection to talk earlier in the occasion are then examined; first, to establish what is going on interactionally without respect to the mentioning of persons, and then as exercises in examining the various ways person-reference and membership categorization can figure in a stretch of interaction. (shrink)