- Millican’s “Abstract,” “Imaginative,” “Reasonable,” and “Sensible” Questions about Hume’s Theory of Cognition
In a 1998 Hume Studies book symposium, Peter Millican provided excellent critical comments on my Cognition and Commitment in Hume’s Philosophy (Millican, “Hume on Reason,” 141–60), and I am grateful that he has done the same for Hume (Millican, “Skepticism,” Hume Studies 40 [this issue]). Many of the new or revised interpretations in the latter book result, directly or indirectly, from his extraordinary stimulus, both in his writings and in person, as a philosophical scholar and interlocutor. His comments range over much of the book, but the majority of them concern chapter 2 (“Principles of Perceptions”), chapter 3 (“The Mind and its Faculties”), chapter 4 (“Sense-Based Concepts”), and chapter 8 (“Morality and Virtue”). In a brief concluding section, he also touches on chapter 9 (“God and Religion”), rightly observing that the source of our main differences about Hume on miracles lie in the crucial role I see Hume giving to his distinction between “proofs” and mere (that is, non-proof) “probabilities.” Millican focuses chiefly, however, on abstract ideas, imagination and reason, and the moral sense, and I will limit my replies to those topics. In doing so, I will follow his section divisions.
It may be helpful to note from the outset, however, two respects in which Millican and I differ in our overall approaches to Hume’s philosophy. Both of these are, of course, matters of degree. First, he is readier than I am to dismiss aspects of Hume’s philosophy as confused or contradictory; or, to put it the other way [End Page 227] around, I tend to find Hume to be, upon close examination, a more consistent philosopher than he does. Second, Millican tends to see the Treatise in particular as a work containing many errors that Hume corrects, most typically by simple omission, in the Enquiries. In my view, each of the Enquiries has more narrowly defined aims than does the corresponding book of the Treatise, and each strives for greater impact through greater accessibility, with the result that they generally seek to minimize the amount of psychological detail provided that is not essential to achieving those narrower aims. Whereas Millican tends to interpret the omission of a doctrine from the Enquiries as itself positive evidence that Hume rejected it, I do not.
1. Teaching and Tone
Whereas Cognition and Commitment aimed to shed light on Hume’s philosophy through the solution of a discrete set of interpretive puzzles, Hume—in keeping with the series of which it is a part—seeks to explain within a manageable compass Hume’s most important ideas across the full range of philosophical topics he discussed. As Millican notes, it tries to do so in a way that will be accessible to readers who have no specialized training in philosophy and little prior knowledge of Hume, while at the same time presenting those ideas and revealing their relations in a distinctive way that will make the book of substantial interest to historians of philosophy and to other philosophers grappling with questions like those that animated Hume. I am gratified that Millican finds the book to be mostly successful in both respects, which can easily be in tension with one another.
To maintain the book’s manageable compass despite its scope, I generally refer readers to the annotated “Further Reading” following each chapter, both for competing interpretations and for work of my own that provides further textual support for interpretive claims I make in the text. I do try to signal within the text itself distinctions among what Hume says (or at least seems to say); what we might reasonably infer that he believes, in order to explain what he says; and what he is more or less committed to granting by what he says, whether he recognizes it or not. In doing so, I also try to observe the distinction between what Millican calls “agreed facts” and “interpretive claims” (207) about each of these classifications.
Given the truly vast variety of interpretations of Hume, however, one reader’s “agreed facts” are often another’s dubious “interpretive claims.” For example, Millican...