Sequencing rules and coherence in discourse

Abstract

This paper examines sequencing rules governing conversational organization. First, it argues against Levinson's (1983) position that it is impossible to formulate sequencing rules such as the one governing an adjacency pair which states the expectation of a certain speech act following the occurrence of a given speech act. Levinson (1983) argues that question can happily be followed by a range of speech acts other than answer. The present paper points out that while it is true that a question is not necessarily followed by an answer, it does not follow that the rule does not apply: it states what is expected to occur, not what actually occurs (see Berry 1982). I argue for the descriptive power of the sequencing rules governing an adjacency pair by demonstrating how they provide a basis for the interpretation of sequences which deviate from the adjacency pair sequence, and how they are deliberately violated to give rise to conversational implicature. Second, the paper points out that not only is there a rule governing what is expected to occur, but there is also a rule governing what is allowed to occur if the discourse is to be coherent. The rule governing coherent sequences is labelled the Coherence Rule, which states that in order for an utterance to form a coherent sequence with the preceding utterance, it must either fulfill the illocutionary intention of the latter, or address its pragmatic presuppositions. I argue for the existence of this rule by demonstrating that firstly a violation of this rule results in incoherent discourse which is noticed and attended to by interlocutors, and that secondly, a violation of this rule can usually be accounted for. © 1991.