Abstract: | In a previous investigation, we demonstrated the role of questions as an instructional resource in the social construction of conversation by depriving conversationalists of their use. The resulting conversational structure, measured in terms of the ability of third parties to reconstruct the dialogues after they had been randomized, was impaired. The importance of culturally shared knowledge for the reconstruction (and by implication the construction) task was apparent in the superiority of British over Canadian subjects in reconstructing unconstrained British dialogues. The present study was concerned with the effects of privately shared knowledge on the construction of conversation. Conversational structure and the frequency of questions were greater in strangers' than in mutual acquaintances' unconstrained dialogues. Strangers' dialogues, but not those of mutual acquaintances, were disrupted under a ban on questions. The results were taken to indicate that strangers need to provide one another with continual instruction in how to proceed, whereas mutual acquaintances, in virtue of their privately shared social knowledge, can construct orderly dialogue with less moment-to-moment guidance. |