My critical comments on Part I of P. J. E. Kail's Projection and Realism in Hume's Philosophy are divided into two parts. First, I challenge the exegetical details of Kail's take on Hume's important distinction between natural and philosophical relations. I show that Kail misreads Hume in a subtle fashion. If I am right, then much of the machinery that Kail puts into place for his main argument does different work in Hume than Kail thinks. Second, I offer a brief criticism of Kail's argument for reading Hume "as a realist about the external world" (Kail, 67). The two parts are (loosely) tied together because it turns out that Kail and I disagree about how Hume thinks of philosophers' activity generally.
One caveat: in what follows I do not offer a review of the highlights of Kail's terrific book. In particular, my arguments do not touch his analysis of the very helpful distinction between feature projection and explanatory projection that plays a crucial role in the two other parts of the book. There is much to admire in Kail's approach, which combines analytic rigor and exegetical care with wide historical learning. There are many illuminating excursions to Malebranche, Spinoza, Leibniz, Berkeley, and even Freud. Moreover, Kail is to be praised for [End Page 67] treating Hume's Treatise as a unified book and not as two distinct contributions to epistemology and moral philosophy. If I were to do justice to the book's many merits, it would far exceed the brief criticism that follows.
I: Natural and Philosophical Relations and Probable Reasoning
Much of the interest in Kail's book is in the nitty-gritty details. Here I focus on Kail's treatment of Humean reason and Hume's treatment of relations. Kail writes, "Hume will argue that our capacity to reason upon philosophical relation of causation presupposes that we have habits of inference—natural relations among ideas—drawn through experience of cause and effect." This is Hume's conclusion to the key section T 1.3.6, "Of the inference from the impression to the idea." Kail then quotes Hume's crucial sentence: "Tho' causation be a philosophical relation, as implying contiguity, succession, and constant conjunction, yet 'tis only so far as it is a natural relation, and produces an union among our idea, that we are able to reason upon it, or draw any inference from it" (T 188.8.131.52; SBN 94) (Kail, 42-43; emphases in Hume and Kail).
At first glance I very much liked Kail's emphasis on Treatise 1.3.6 and 184.108.40.206, in particular.1 Kail then adds: "Part of the significance of this conclusion lies in the consequence that there is no faculty of probable reason that is independent of the mechanism of association. . . . Prior to Hume, associative inference was identified with animal inference and, while humans did share in that kind of associative inference, such inferences were sharply distinguished from reason. For Hume, the basis of probable reason is animal reason" (Kail, 43). Kail thinks the result so important that he repeats the claim a few pages later (Kail, 46).
So far, so good. But upon closer inspection it turns out that Kail and I disagree on some crucial matters. In treating of Hume's account of reason, Kail makes the following four-fold distinction: (1) reasonF is a psychological faculty, which, itself, comes in two "forms:" demonstrative and probable reason; (2) reasonN is a reason qua a normative consideration in favour of a judgment; it turns out that on Kail's account "reasonF is our means of detecting or being sensitive to reasonsN" (Kail, 36). (3) ReasonI is reasoning qua a form of inference or mental transition from one judgment to another; it turns out that this third type of reason is a species of, (4) reasoning A, a mental activity of a certain kind in which ideas are compared (Kail, 36-37).
Let's accept temporarily for the sake of argument the utility of this fourfold distinction in thinking about the matters treated by Hume. But is it really Hume? According to Kail "there...