A task for any description dealing with the use of spoken language in natural situations is to account for the radical interdependency of speaking and listening. Speakers, in the very act of speaking, i.e. when holding a turn, are at the same time also listeners. They perceive the responses coming from listeners and may instantly, in their speaking, react to them and modify their course (see Jefferson 1973). Listeners, on the other hand, even though maintaining their stance as recipients of the developing turn of a current speaker, act at the same time to some degree as speakers: listening is an activity that has a global ecology, comprising facial, proxemic, gestural and bodily signals as well as purely verbal ones. Yet it also has a vocal, verbal and linguistic side: with small tokens like ‘hm’, ‘uh huh’, ‘yeah’, ‘yes’, ‘right’, etc., which a ‘listener may get in edgewise’ (Yngve 1970), which ‘may come between sentences’ (Schegloff 1982) or which act as ‘bridges between turn-constructional units’ (Goodwin 1986), recipients track the course of emerging talk in discourse, and display, in brief, their current understanding of that talk, co-constituting its continuation. It is in this limited sense, restricted to the small but still hearable, linguistic tokens inserted into the ongoing talk of a current speaker, that the term ‘recipiency’ will be used in the following.
To do the ‘intermediate’ work they do, the tokens of recipiency must remain brief and unobtrusive – they are designed to fit into slight gaps of ongoing talk and usually avoid obliterating parts of the ongoing talk to which they refer.