Conversation analysis, dialogism, and the case for a minimal communicative unit

Severinson Eklundh and Linell (1983) asked whether a minimal form of communicative interaction exists and, if so, how many moves it would require. In conversation analysis, the response to these questions has traditionally been that such a form exists and that it takes the form of a pair of adjacent utterances consisting of a first pair part (e.g., a greeting or a question) and a second pair part (e.g., a greeting in return or an answer to the question). Severinson Eklundh and Linell acknowledged that communicative exchanges could take the form of two-part sequences, but they argued that this format is relatively limited in scope. Instead, they proposed that the basic format for most communicative interactions is a three-part sequence and that this structure should not be reduced to a base pair with a sequence closing third as an expansion of the pair. This issue has been the subject of ongoing debate over the last four decades. In this article, we discuss how conversation analysis and extended dialogism have addressed the idea of a minimal form of communicative interaction. We review different approaches and how they overlap and diverge, and we make conceptual distinctions to account for their differences.