Hume's Of Scepticism with regard to reason: A Study in Contrasting Themes.

121. HUME'S Of Scepticism with regard to reason: A STUDY IN CONTRASTING THEMES.* This paper attempts to describe the complex dialectical interplay among the contrasting rational, sceptical and naturalist elements which appear in Section I, Part IV of Book I of Hume's Treatise of Human Nature. At the same time we shall try to show that, contrary to Hume's own evaluation of that section, it is the sceptical element, in which the unreliability of reason is supposedly demonstrated , as opposed to the naturalist element, in which its unreliability is supposed to remain hidden from us, that deserves to triumph. With these two goals in mind and in order to make it easier for the reader to follow the inevitable twists and turns of the dialectic, we have divided the paper into two three-sectioned parts. The first part bears the title "Reason and Scepticism", while the second part bears the title "Naturalism". The first section of Part I is concerned with proving that Hume's attempted reduction of all knowledge to probability is frustrated by his failure to explain how we can make mistakes in simple operations with small numbers. If so, the first phase of his sceptical assault against reason fails to dislodge her from her throne. If the second section is right, the second stage of the sceptical assault fares no better. For Hume's attempted reduction of all probabilities to zero is shown to rely on an unsatisfactory formula for calculating the probabilities in question. In the third section an effort is made to show that the attempted reduction of the second section, i_f it had been a success instead of a failure, would have destroyed reason but left scepticism intact. *I should like to thank Howard Sobel for his penetrating and unsparing criticism of an earlier version of this paper. 122. Hume's conviction, on the other hand, that the flawed reduction in question is unflawed and destroys both leads to a prolonged discussion, initiated in the first section of Part II, of his naturalism. Indeed, this theme dominates the rest of the paper. Thus it deserves a part to itself. If we are correct in the first section of Part II, Hume's naturalism is best interpreted as a doctrine of instincts relying for its doctrinal status upon a failure to accept the conclusion of the sceptical argument directed against reason. But, as we try to bring out in the second section, the specific psychological inability with which Hume could in the final analysis give substance to this failure of acceptance is left without any causal foundation. Finally, in the third section that inability is seen to run afoul of reason as Hume himself understands it. And note is taken of the fact that reason so understood and the sceptical argument directed against reason that Hume adopts for his own both commit him willy-nilly to rejecting for good or ill one standard definition of knowledge. Reason and Scepticism Nowhere are the rational elements in Hume's philosophy more in evidence, ironically enough, than in Section I, Part IV of Book I of the Treatise entitled Of Scepticism with regard to reason. For it is there we are told in the very first paragraph that in all demonstrative sciences the rules are certain and infallible . Our reason, moreover, must be consider ' d as a kind of cause, of which truth is the natural effect. (T180) In other words, reason, left to its own devices, would produce truth and, presumably, only truth. Anyone who steadfastly adheres to such principles will reject out of hand one kind of scepticism with regard 123. to reason. That is the kind which results from the alleged discovery that reason, left to its own devices, compels us to embrace one or more contradictions, an obvious form of 2 falsity. It is a kind of scepticism which Bayle cultivated so assiduously. And yet, despite the influence which Bayle 's writings exerted upon Hume, they never convinced him of the viability of this kind of scepticism. On the contrary, apart from some wavering in the Enquiry, he was as adamant as any Leibnizian rationalist that right reason3 ing leads only...