There is a fairly traditional view in philosophy of language (stemming from the work of Paul Grice) which holds that there are two kinds of meaning which together make up the complete content of an utterance: on the one hand, there is what the sentence means and, on the other, what the speaker conveys. However, in recent years this orthodox position has been superseded by the view that there are three kinds of meaning which make up the complete content of an utterance: what the sentence means, what the speaker explicitly asserts by their utterance and what she (merely) implicates. Following Sperber and Wilson I’ll refer to the intermediate kind of content, between literal sentence meaning and implicatures, as the explicature of the utterance. My aim in this paper is to take a closer look at the notion of an explicature, asking what pragmatic processes might give rise to explicatures and how they are supposed to be individuated. My argument will be that advocates of explicatures face a tough job in answering both these questions and I’ll suggest that these difficulties may be instructive. Perhaps, I’ll contend, the problems show us that explicatures don’t really exist in any interesting sense, in which case the door is open for a return to the simple binary model of Grice (and Kripke). Thus, in conclusion, I’ll sketch a model of communication which appeals only to the traditional Gricean two-way distinction.