Calibrating understanding in interaction

This study examines how Mandarin Chinese speakers deploy candidate understandings in other-initiated repair in everyday interaction. Drawing on 15 hours of naturally occurring conversation, the analysis shows that candidate understandings recurrently emerge when some aspect of prior talk is treated as not yet sufficiently interpretable for the ongoing interaction. In such cases, speakers produce proposals that specify either the referent or the interpretive scope within which a prior turn is to be understood. In Mandarin, candidate understandings are frequently formatted in nonclausal forms with final particles to invite confirmation from the prior speaker. Such practices do more than resolve problems of comprehension. Rather, they constitute an interactional space through which participants negotiate what can remain shared and what requires further specification within the local interactional context, thereby recalibrating understanding as the interaction unfolds.