James L. Kinneavy's A Theory of Discourse: The Aims of Discourse (PrenticeHall, 1971) has contributed much to the field of English. Evidence of its impact is that it is required reading for two NEH seminars-Edward P. J. Corbett's summer seminar at Ohio State and Dudley Bailey's year-long seminar at the University of Nebraska. This evident concern and the book's recent appearance in paperback (Norton, 1980) prompt a review of its strengths and limitations. Kinneavy clarifies the need for order in English studies, but-to use his own term for characterizing the field-his work is "preparadigmatic" in that his categories are static and his approach is too closely tied to literary criticism to be helpful in composition. Though he intends to rescue composition from "the present anarchy of the discipline,"' his theory is unsatisfactory for many who teach composition, largely because he fails to account adequately for rhetorical choices and composing processes. This review will focus on some of the underlying reasons for the limited success of Kinneavy's theory. Kinneavy seems aware of many of his presuppositions, including his assumption that he can side-step considering rhetorical processes. However, he does not always seem to be aware of the implications of his methodological decisions. His decision to analyze "the aim which is embodied in the text itself" (49) is based on a desire to concentrate on "discourse" rather than "composition." A theory of composition, he argues, would require attention to the process of composing, a concern he concludes "is not desirable" for an analysis of aims (4). He prefers to deal with "discourse analysis," with "the characteristics of the text," with the decoder, "who is the primary element in any communication situation" (49-50). Ironically, though he recognizes the rhetorical significance of the writer's audience, he fails to perceive that rhetoric, unlike discourse analysis, must deal with the process by which texts come into existence. He thus sets out to establish "the basic foundations of composition" and to provide a "framework of research" for all areas of dis-