Kant's Theory of Science (review)

206 HISTORY OF PHILOSOPHY phy of science, Gregorio Kiimovsky. The problem of evil in the Dialogues concerning Natural Religion is discussed by E. Chaves, in its logical and semantical aspects. It is impossible to give, in a short note, a clear idea of the significance of such a volume for philosophy and the social sciences in Latin America, where publications of this quality do not abound. The Argentinian papers on Adam Smith reveal the existence of a solid scholarship in economics in that country; and Flew's contribution is as illuminating as usual. As for the papers on Hume, they fully display the maturity of the Argentinian philosophic community. Criticism of some of the more polemical views exposed in these papers would be out of place here. But one can regret, in a publication dedicated to two philosophers with such strong mutual ties, the absence of any discussion of the relationship between them. This certainly is one of the most important links to be found by the historian of eighteenth-century science and philosophy. One of the doors to this kind of discussion might be the analysis of Hume's economical essays and of their relevance for Smith's work; another one might be the study of Smith's theory of moral sentiments, in its connection both with Hume's philosophy and Smith's economical theories. Such absence, however, does not diminish the importance of this volume in the Latin American philosophical and scientific context. J. P. MONTEIRO University of SSo Paulo Gordon G. Brittan, Jr. Kant's Theory of Science. Princeton: Princeton University Press, 1978. Pp. xii + ~15. $13.5o. This is unquestionably one of the better books that have recently been written about Kant's philosophy. Although by design limited in scope, it deals quite effectively with some of the recurring problems in interpretations of Kant. The arguments are always pertinent, clearly stated, and well balanced. The basic theme of the book--if such it may be called--is that "what Kant wrote constitutes not so much a reconcilation of 'rationalism' and 'empiricism' as the rejection of a feature they share in common" (p. 3). What the author rightly emphasizes is Kant's empirical realism. In the process of arguing his case, Brittan clarfies definitively (p. 13o) Kant's distinction between transcendental and empirical questions--a distinction that, incidentally, has a decisive bearing on Kant's conception of the thingin -itself (although Brittan does not mention this fact). Although I intend to comment on various chapters of the book, there is one general remark that must be made now. In conformity with the general misconception (which stems from Norman Kemp Smith's translation of the Critique of Pure Reason), Brittan in effect always uses Kant's term "a priori" in an adjectival sense: "Synthetic a priori judgments"; "synthetic a priori sentences"; "synthetic a priori principles ," and so on. But it would be a help in understanding Kant's position if all commentators were to realize that Kant uses the term "a priori" in an adverbial sense only. BOOK REVIEWS 207 Let me give a few examples: A4/B8: "wie weit wires in der Erkenntnis a priori bringen kOnnen." All/B25: "welche mithin v611ig a priori m6glich ist." A~o7: "liege also a priori allen Begriffen zugrunde." Alo9: "als a priori notwendig angesehen." Bxf: "was er nach Begriffen selbst a priori hineindachte." Bxvii: "wie man a priori yon ihr etwas wissen k6nne." B3: "ein Satz, der ein Urteil a priori ist." B159: "ihrer Verbindung nach, a priori zu erkennen." A156 "Die M6glichkeit der Erfahrung ist also das, was allen unseren Erkenntnissen a priori objektive Realit~it gibt." A 158/B197: "in einem synthetischen Urteile a priori [that is; a synthetic judgment made a priori]." There are dozens of passages of a similar type that I could have cited; but the samples I have given suffice. Brittan's discussion of Kant's theory of science is based exclusively upon the Critique of Pure Reason, the Prolegomena, and the Metaphysical Foundations of Natural Science; and one should not fault an author, as Kant would say, who intended to write a short book for not having written a long one. My point...