Hume’s Epistemological Evolution by Hsueh M. Qu (review)

This is a wonderful book that ambitiously and impressively brings to convergence two parallel, perennial lines of inquiry in Hume’s scholarship. One is the classic Kemp Smith question concerning the relation between Hume’s naturalism and skepticism. The other is about the relation of the first Enquiry to book 1 of the Treatise. Qu observes that the Treatise is most distinctively naturalist or descriptive, while the Enquiry is decidedly normative. His approach is to examine the two questions through a single lens by inquiring into the nature and causes of Hume’s epistemological evolution.

Any evolutionary account must identify what evolutionary biologists refer to as the “transitional form.” For Qu, this is part 4 of book 1 of the Treatise. In part 4, Hume discusses skeptical arguments about reason and the senses, the pitfalls of ancient and modern philosophy, and the immortality of the soul and personal identity. Qu perceives in these texts a Hume who becomes increasingly conscious and worried about the justification of our most fundamental beliefs. In part, this is triggered by Hume’s criticisms of other systems of philosophy, which force him to turn the table and articulate the principles guiding his own normative distinctions. According to Qu’s reading, in the “Conclusion” we witness the unraveling of Hume’s own commitments, leading up to Hume’s famous “dangerous dilemma” (T 1.4.7.6). Qu, like other scholars, identifies Hume’s solution as the “Title Principle” (TP). But Qu maintains that Hume’s TP fails to address the challenges raised by the dangerous dilemma. Qu identifies striking references to the “Conclusion” in the Enquiry and interprets Hume as taking a second, successful shot at addressing the problems he encounters in the “Conclusion.”

The book makes its case forcefully, carefully, and systematically. The range of literature it engages with is truly remarkable. Rival views are characterized in a conscientious and fair manner. In his criticisms, Qu anticipates and responds well to possible rebuttals. This book is an excellent exemplar of the art of philosophical argumentation.

There was a question that lingered in my mind after reading Qu’s book. What explains the transition, in book 1 itself, from descriptivism to normativism? Part of the answer given above is that Hume’s criticisms of other systems of philosophy force reflection on his own. But this just raises the same question in another form: If Hume is engaged in a descriptive project, then why turn to criticism of philosophical systems? Qu’s depiction of the Treatise as essentially descriptive also conflicts both with the stated purpose of the Treatise, namely, to establish a foundation for the sciences (T Intro. 7), and Hume’s later assessment of the Treatise in the Abstract (Pref. 2 and Abs. 3).

I find myself also disagreeing with Qu’s evaluation of Hume’s response to the dangerous dilemma. In the “Conclusion,” Hume is most exercised by the role of vivacity in our cognitive lives. One of the most important lessons of book 1 is that causal reasoning depends on the feature of vivacity to generate belief. Causal reasoning terminates in belief, but the element of vivacity requisite for belief does not seem to arise from reason. Indeed, Hume notes that it “seemingly is so trivial, and so little founded on reason” (T 1.4.7.3). Worse yet, vivacity seems indiscriminately promiscuous. It attends good beliefs as well as bad ones, such as superstition.

This is the train of thinking that lands Hume in the dangerous dilemma. It appears that he cannot assent to causal beliefs without assenting to “every trivial suggestion of the fancy” (T 1.4.7.6). The only alternative appears to be an endorsement of reason itself independent of the element of vivacity. But Hume has shown in the section “Of Scepticism with Regard to Reason” that the understanding, “when it acts alone, and according to its most general principles, entirely subverts itself, and leaves not the lowest degree of evidence in any proposition, either in...