A central document in the philosophy of language over the last two decades has been J.L. Austin's How to Do Things With Words, 1 which first proposes a distinction between "perforrnative" and " constative " utterances and then proceeds in systematic fashion to scrap the distinction in favor of the now familiar model of locutionary, illocutionary, and perlocutionary acts, wherein no formal division among kinds of speech acts is maintained. While philosophers have debated steadily the range and validity of these concepts ," students of discourse have rather more sporadically looked into their possible relevance to an understanding of rhetorical products." It is mainly to the second of these questions that the present essay addresses itself, growing primarily out of a theoretical interest in the classifications of discourse rather than in the theory of speech acts per se. Specifically, I wish to propose that Austin's original distinction between performative and constative has an application to whole discourses or "rhetorical acts;"" and that the notion of "performative" thus applied is a principal definer of the category of rhetoric traditionally known as "epideictic."