Turns-at-talk are fundamental units of participation in talk-in-interaction, and turn-constructional-units are the basic building blocks for turns. Possible completion of a TCU is, in principle, the possible completion of the turn, but multi-unit turns are not uncommon, and participants have practices for constructing multi-unit turns and for recognizing them in the course of their production. This article offers an account of one practice usable by speakers and recipients to convey and recognize the designed completion of a multi-TCU turn and/or a (...) multi-turn sequence in which ‘answering’ is being done: returning to, or ‘re-using’, a word or phrase from the start of the turn or sequence, whether articulated by same or different speaker, whether used to refer to same or different referents. This practice is one of the resources by which the overall structural organization of an interactional unit and its local realization are mutually realized. (shrink)

Conversation analysts have noted that, in psychotherapy, formulations of the client's talk can be a vehicle for offering a psychological interpretation of the client's circumstances. But we notice that not all formulations in psychotherapy offer interpretations. We offer an analysis of formulations that are diagnostic: that is, used by the professional to sharpen, clarify or refine the client's account and make it better able to provide what the professional needs to know about the client's history and symptoms. In doing so, (...) these formulations also have the effect of shepherding the client's account towards subsequent therapeutic interpretation. In a coda, we notice that sometimes the formulations are designed discreetly. We examine one such discreet formulation in detail, and show how its very ambiguity can lead to its failure as a diagnostic probe. (shrink)

Starting with Houtkoop and Mazeland’s study of discourse units, and touching upon recent studies aimed at detailing unit projection in interaction, this article argues that the drive toward abstract and discrete models for units and unit projection is potentially misleading. While it has been established that to engage in talk-in-interaction, as it unfolds in real time, participants rely on projectable units, research aimed at defining units unintentionally backgrounds the contingency inherent in interaction. A central function of language for collaborative action (...) is the management of simultaneously unfolding facets of action, sound production, gesture, and grammar – produced by multiple participants. This article draws upon classic and current data analyses foregrounding linguistic/interactional practices designed to manage local contingencies. It is argued that attention to participants’ regular methods for managing and exploiting contingencies be incorporated from the outset in our descriptions of language and the nature of unit building in interaction. (shrink)