Journal of the History of Philosophy

Descartes and Probable Knowledge JOHN MORRIS THB BEs'r-KNOW'N RULB of Descartes' method is surely the first: "never to receive anything as true which I did not know evidently to be so . . . and to include nothing more in my judgments than what presented itself so clearly and so distinctly to my mind, that I would not have any occasion to doubt it." t Such a rule leaves no room for knowledge which is merely probable. If our sciences are to be firmly grounded, then they must be based upon principles which are certain, not probable. And, indeed, until the lime of the Principles, there is no room for probable knowledge in Descartes' system. But the analysis of probable knowledge, or "moral certainty," which is developed there, is interesting in its own right, and it brings Descartes somewhat closer to a more modern, pragmatic view of science. I want, in this paper, to trace the development of Descartes' doctrine of probable knowledge, be~nning with a partial statement of the method as it appears in the early works, and contrasting this with the later analysis of the Principles. The difference between the two treatments seems to lie largely in the rather different status accorded to hypotheses or "suppositions," upon which scientilie explanations are based. 2 The method of the Discourse, and the accompanying physical treatises, requires Discourse, H, (Euvres de Descartes, edited by Charles Adam and Paul Tannery (13 vols.; Paris: Cerf, 1897-1913; hereinafter referred to as AT), VI, 18. A discussion of the role of hypotheses in Descartes' method may be found in Zydora Dambska, "Sur certains principes m~thodologiqttes dana lea 'Principia philosophiae' de Descartes," Revue de M~taphysique et de Morale, I.XII (1957), 57-66. Mine. Dambska contrasts Descartes' "official" methodology with an "unofficial" methodology which she finds in the Dioptrics, Meteorology, and especially in the Principles, where she finds the quam of Principles, IV, I, an anticipation of the als ob of Vaihinger. She also terms this latter method "positivistic" (p. 63). I do not believe, as does Mine. Dambska, that the earlier physical treatises anticipate the treatment of hypotheses in the Principles, since the early works always claim that the hypotheses are rendered certain by their effects. In another article, J.-P. Weber, "Sur une certaine 'M6thodologie offieieuse' chez Descartes," Revue de M~taphysique et de Morale, LXIII (1958), 246-250, the author points out that Mme. Dambska's characterization of the "unofficial" method could also be applied to the "official" method developed in the Rules, and that the distinction therefore cannot be maintained. The distinction which I develop here is somewhat different. In the early works, if a physical "hypothesis" or "supposition" has true consequences, then it is confirmed or "proved" by those consequences; in the Principles, the opposite claim is made: a hypothesis may be false, even though all its consequences are true. [3031 304 HISTORY OF PHILOSOPHY that we take, as hypotheses, certain theories concerning the subject under investigation . From these, we derive descriptions of individual events. If this derivation can be carried out, then the initial hypotheses are confirmed by the results. At the time of the Discourse, Descartes is ready to claim that this method gives certainty. He says that he has "demonstrated" the hypotheses from which he has started, and that the hypotheses "explain" the effects. "For experience renders the greater part of these effects very certain, the causes from which I deduce them serving not so much to prove them as to explain them; but, on the contrary, it is the causes which are proved by the effects." 3 The fallacy here is one that Descartes certainly should have recognized. It is that of atfirming the consequent. Given hypotheses A, B, C.... N, and propositions a, b, c.... m. Clearly to show that the hypotheses entail the propositions, and to know or assume that the propositions are true, certainly does not prove that the hypotheses are true. The claim, in the Discourse, that this method gives certainty has not been established. In 1637, the year in which the Discourse was first published, a letter to Plempins suggests that Descartes was considering another approach to the...