The Riddle of Hume's Treatise: Skepticism, Naturalism and Irreligion (review)

Professor Russell's book is a nice and rewarding book, well written, philosophically and historically instructive. It provides the reader with new and refreshing insights into Hume's Treatise, some of which are quite illuminating. The main argument is as follows:

How can we escape the skepticism/naturalism divide into which most commentaries get trapped? This riddle, this impossible conciliation, has become the test for any new book on the Treatise. But it might be the effect of not considering the true intention of Hume's book. It is a fact, if we carefully study the reactions to the book which followed after its publication, that the earliest critics believed that there was a close connection between Hume's skepticism and his more or less secret atheistic intentions. More precisely, there is evidence (and Russell goes deep into the details) that they regarded Hume's opposition to the dogmatic principles of Samuel Clarke as central for the understanding of this connection. Clarke, a much more important author in Hume's time than today, was committed to the defence of the rational credentials of the Christian religion and had a large influence at that time in Scotland and in the Universities. Beattie, especially, and one knows that wickedness is a good analyst, suggests that, along with other skeptics, Hume is much more obliged to Hobbes than he seems willing to acknowledge. And so we get to the crucial argument of the book (chapter six): "the scope and structure of Hume's Treatise is modelled or planned after that of Hobbes's The Elements of Law, and in this respect there exists an important and unique relationship between these two works" (61). Hence, "we are in a position to begin to excavate and systematically uncover Hume's fundamental irreligious intentions throughout the Treatise." This excavation is the constant motif of the following chapters which, part after part, investigate the three Books of the Treatise, to support Professor Russell's thesis.

Let us take for instance chapter XI: "Induction, Analogy and a Future state." Russell claims that "Hume's account of the problem of induction, as originally presented in the Treatise, is significantly motivated by irreligious objectives" (129). In 1736, Butler published his Analogy of Religion, partly as a response to Tindal's Christianity as old as creation (1730), a book that had an enormous impact. In a letter to Lord Kames (December 1737) Hume expressed his wish to meet Butler and to show him his own book. It is evident that Hume was acquainted not only with Butler's Sermons devoted to Ethics, but also with the Analogy. Now, a close comparison between the Analogy and the Treatise shows that (on the Bishop's side): (1) Butler's argument explicitly rests upon probable reasoning, which he clearly distinguishes from demonstrative reasoning; (2) analogy between the past and the future is the spring of probable reasoning; (3) Butler switches from the future in this world (A) to a future state in Heaven (B), arguing the uniformity principle; (4) he claims that (A) and (B) are equally credible and are naturally believed by the common sense. Now compare with Hume's doctrine on causation and the progress of part three in the first Book of the Treatise. It is easy to detect a relevant correspondence between the two texts and to see how much Hume's critical examination of causal inference antagonistically responds to Butler's philosophical commitment to the defence of religion. And so we discover the way out of the skepticism/naturalism riddle: the first stage of Hume's reasoning is to show that there is no foundation, a priori or a posteriori, for the uniformity principle; the second stage presents his positive "naturalism" for our causal inferences within this world (137ff.). It can be asked why Hume did not target his skepticism directly against religious inference and ran the risk of endangering every probable...