The Unity of Understanding: A Study in Kantian Problems (review)

BOOK REVIEWS 309 strates, though Kant dismisses Reid's response to Hume as inadequate and/or missing the point, there is much that Kant seems to have learned from Reid's response. And finally, as if to acknowledge that contemporary academic specialization does not necessarily map well onto eighteenth-century thought, there are two essays that address concerns not necessarily regarded today as "philosophical." ROdiger Schreyer explores theories of the origins of language, particularly those of Smith and Reid, seeking to reveal what such theories can show us about Scottish Enlightenment conceptions of the "science of man." And Thomas Crawford offers a contribution on Boswell, addressing questions such as what the biography of a literary (as opposed to philosophica /) figure can reveal to us about this "science of man." This is an innovative, challenging, and satisfying collection. MARK H. WAYMACK Loyola Universityof Chicago Hubert Schwyzer. The Unity of Understanding: A Study in Kantian Problems. Oxford: Oxford University Press, x99o. Pp. 179. Cloth, $26.00. In the Introduction to this book, the author disavows any attempt at commentary or summary of any part of Kant's philosophy, and instead promises to clarify the "philosophical aims and convictions embedded in Kant's account" of human understanding, captured by the author in three "theses" that, he argues, Kant used as guidelines in working out his account of the understanding. These are 1) a "spontaneity thesis," namely, that the understanding is "not something that is a response to, or that conforms to, something else"; 2) an objective validity thesis, namely, that the understanding is capable of attaching a priori to objects; and 3) an "intelligibility" thesis that claims that our understanding is necessarily communicable, i.e., subject to "rules or conditions that govern the general intelligibility of what we say." Schwyzer explains that Kant's task as well as his own is to show how one and the same capacity--the understanding-can meet all of these conditions. The author explains that the spontaneity thesis requires a proof of the objectivity thesis, for how can something so far removed from sensibility as functions of unity derived from logic be connected to objects? Schwyzer argues in Chapters ~ and 3 that even though, for Kant, the ability to find some application is part of what it is to have a concept, still Kant cannot nonarbitrarily show that the unschematized categories have the particular schemata that he wants to show they have. That is, Kant cannot give an argument for why one schema fits a particular category, since he has only unschematized functions of unity to work with, and these cannot "point to, or even suggest, one criterion for its application rather than another" (27). Thus the schematism cannot provide a unified account of understanding (-- concept possession) that includes both the "ability to think coherently with the thought-element" and to apply it (40). The transcendental deduction, Schwyzer argues (in Chapters 4 and 5) does not solve the application problem. Schwyzer reads the argument of the deduction as begin- 31o JOURNAL OF THE HISTORY OF PHILOSOPHY 30:2 APRIL 1992 ning with an "original" spontaneous activity of the mind ungrounded in sensibility (the synthetic unity of apperception), and ending with the conclusion that this activity itself requires the functions of unity specified by the Clue (6e). Schwyzer repeats a standard criticism that even if it is the case that all the understanding can do is to judge (synthesize ), still Kant has not given an argument for why we must use these forms ofjudgment and not something else. He then argues that even though Kant may not need to deduce the individual categories, still "the question of what, in general, is to count as synthesis" cannot be left open here, and he suggests that what Kant needs to show is that representations must be synthesized "by those operations which are reflected in the structure of the (public) language, whatever those operations might be" (96-97). Hence what Kant needs is a general argument connecting the unity of apperception not simply with synthesis but with language (97). But Kant can make no such connection "since there is nothing in the argument [of the transcendental deduction] to even suggest the...