According to Gricean theory, conversational implicatures depend on what people presume or know about the speaker. For S to implicate p, others must presume that S is observing the Cooperative Principle, and they must know or at least believe that observance of the Cooperative Principle requires S to believe p. I argue here that these clauses in the Theoretical Definition are fundamentally misguided. They are the product of two mistakes: the assimilation of speaker meaning to communication, and of implicature to inference. What a speaker means or implies is determined by what the speaker intends. But one person's intentions do not depend on what others presume, believe, or infer. Conversational principles do play a role in the recognition of implicatures. But implicatures need not be recognized, and their recognition does not depend on any specialized reasoning process. The Cooperative Principle and associated maxims play no role in the generation of implicatures, and they play the same indirect and nonessential role in implicature recognition that known tendencies play in inductive inference generally.

THE COOPERATIVE PRESUMPTION CONDITION

The Theoretical Definition specifies that S conversationally implicates something only if S is presumed to be observing the Cooperative Principle. The Generative Assumption holds further that a conversational implicature exists in part because the speaker is presumed to be observing the Cooperative Principle. This cooperative presumption condition is vague in some respects: it is not clear who it is that is supposed to presume that S is observing the Cooperative Principle.