Questions cartésiennes: Méthode et métaphysique (review)

452 JOURNAL OF THE HISTORY OF PHILOSOPHY 3o:3 JULY X992 from sex, often reduces to little more than one person exploiting someone else, or someone else's body. More attention in the last chapter to Augustine's powerful treatment of lust for domination--perhaps it is more or less omitted because of the limitations of the series, but Kirwan does talk about politics--would have been helpful. Kirwan's treatment of Augustine's arguments in themselves, however, is usually rewarding and deserves careful attention, though he sometimes draws later figures-Luther , Calvin, Jansen--into the debate to such an extent that it is harder for the reader (who may not be certain as to whether they are being treated fairly) to follow Augustine's views. These several authors might have been profitably relegated to footnotes . Nevertheless, among many illuminating passages, I would single out Kirwan's treatment of certainty (22-28, though it would have been good to have heard more about belief); of words, sentences, and names (43-52); of the sinlessness of God (7~76 ), though here and elsewhere it would have been interesting to hear of the relation between what might be dubbed "God's overall determination of nature" and God's "determination" of specifically moral behavior; of "degrees of free will"--despite my caveats above (82-9o); of God's foreknowledge (or rather "knowledge," 95-xo2), though the connection with predestination (144-46) is underdeveloped; of omnipotence (124-28), though this is too short a treatment of a topic which should have worried Augustine more than it did; of God's "favouritism" and "comparativejustice" 046-5o); of change and eternity 067-74); of lying and the general relationship between means and ends (196-2o4). Augustine's thoughts are scattered over many decades of writing; Kirwan pays little attention to the very important fact that his thought developed, thereby sometimes examining arguments which Augustine would never have been prepared to defend. But that is not always the case, and most of the analyses are penetrating and helpful for those looking for a more complete Augustine than the traditional books (and Kirwan's own book) have provided. Nonspecialists with logical interests will find it fascinating; specialists will find it necessary reading: an important corrective. JOHN M. RIST University of Toronto Jean-Luc Marion. Questions cart~siennes: M3thode et m~taphysique. Paris: Presses Universitaires de France, 1991. Pp. 264. FF loo. Jean-Luc Marion's Sur l'ontologie grise de Descartes, 2nd ed. (Paris: J. Vrin, 198x), Sur la theologicblanche de Descartes (Paris: PUF, 1981), and Sur la prisme rwttaphysiquedeDescartes (Paris: PUF, 1986) establish him as a major interpreter of Descartes. Questions cartksiennes is an excellent entr~ to his onto-theo-logical method and its results. Here his genius for architectonic elaboration is brilliantly exhibited: on the relations of metaphysics to method, on the ego and the cogito, and on the question of the existence of other minds and God. Besides the impact of their content, the clarity and reach of these essays force one to consider foundational questions concerning philosophy and its history. BOOK REVIEWS 453 The substance of the papers is too rich to summarize, but in them Marion develops at/east these theses: (1) that dreaming is itself thinking in which the cogito maintains its validity; (9) that Ego sum, ego ex/st0 has less metaphysical certainty than Cogito, ergo sum; (3) that simple natures are the primary Cartesian ideas; (4) that capab/e/capax shifts from a sense of receptivity to a sense of ability; (5) that Cogito, ergo sum is embodied in Descartes's notion of generosity; (6) that Descartes does and can prove the existence of only one mind, his own; and (7) that both Anselm and Descartes give nonontological proofs for the existence of God to the effect that because we cannot have a concept of God, God exists. I think these conclusions in general are correct. Nevertheless, Marion sets the arguments for each one of them on an exposed, undefended, and probably indefensible point of departure. These Achilles heels are: (1) taking the generalization "whatever thinks, exists" to be metaphysically secure; (9) taking Baillet's report of Descartes's dreams...