Abstract This conversation-analytic study takes as its focus one kind of adjacency pair, comprising a first pair-part that is a tentatively (i.e., provisionally and/or approximately) formulated questioning turn, and a second pair-part that supplies an answer (or sometimes a non-answer response). Such tentative questioning turns, through a range of turn design features, are routinely built to (1) relax the expectations for any particular mentioned candidate answer to be confirmed by the answerer, (2) display difficulties with finding an optimal wording of the question, (3) treat the prospective answerer as better placed for setting the terms in which the answer will be framed, and (4) defer to the prospective answerer some rights to shape the topic and/or action agenda. Qualitative and quantitative evidence is provided showing that answerers orient to this tentativeness by usually beginning to answer in overlap with the questioning turn. The answers themselves also attest that TQTs set only loose constraints for the response turn.
Rasmus Persson (Tue,) studied this question.