1. See for instance Levinson 1983, p. 131; Neale 1992, p. 512, p. 523; Green 1998, p. 86. Research on this paper was supported in part by a USEM Research Grant from the University of Virginia, and is here gratefully acknowledged. Earlier versions of this paper were presented at the University of Virginia, July 2000, and at the Workshop on Saying, Meaning and Implicating, part of the Gesellschaft fur Analytische Philosophie, Bielefeld, Germany, October 2000. I am grateful to both audiences for their comments on those occasions. I have also benefited from comments from members of my Expression and Meaning seminar at the University of Virginia, particularly Joe Milton, Jacob Pease, and Kranti Saran. 2. Other challenges for the Gricean analysis of speaker meaning, such as whether the commonly invoked reflexive intentions are sufficient for their analysandum; whether an adequate analysans would involve attribution of a psychologically realistic set of attitudes; whether communicative intentions can be taken as logically prior to the notion of meaning that they are used to shed light on; these challenges will not detain us in the present study. For treatment of these challenges see Avramides 1989 and Clark 1996. from G. Meggle and C. Plunze (eds) Saying, Meaning, Implicating (Univ. of Leipzig Press, 2003): 200-19. MITCHELL S. GREEN Grice’s Frown: On Meaning and Expression 1. Introduction Paul Grice’s writings suggest that he views conversational implicature as a species of speaker meaning, and many writers concerned with implicature have endorsed that suggestion. 1 The myriad challenges facing the explication of speaker meaning raise some doubt, however, whether it is an adequate framework for the notion of conversational implicature. In particular, the “reflexive communicative intentions” (intentions to produce an attitude in an audience by means of the audience’s recognition of those very intentions) that Grice and his followers have taken to be crucial to speaker meaning do not, on closer scrutiny, seem to be necessary features of that concept at all. 2 At the same time, little consensus has emerged on how properly to understand speaker meaning