In speech act theory, there has been a tendency to concentrate investigations on the single utterance of a sentence. In my opinion, this tendency seems to originate based on a tacit and unproven premise: namely, an illocutionary force should dwell in each single utterance. In this paper, first, I take up the texts of Austin and Searle as examples, and try to show how this premise was smuggled into the theory. Then I argue that there are cases where we do an illocutionary act not by making a single utterance, but by making a group of utterances ("conversational sequence"). Through these examinations, I conclude that the premise in question is to be replaced by an alternative one: Sometimes-or maybe fundamentally -an illocutionary force dwells in a conversational sequence.
