Hume: Radical Sceptic or Naturalized Epistemologist?

Hume Studies Volume XXIV, Number 1, April 1998, pp. 31-52 Hume: Radical Sceptic or Naturalized Epistemologist? KEVIN MEEKER According to Barry Stroud, we should view David Hume neither as a prototypical analytic empiricist1 nor as "the arch sceptic whose primary aim and achievement was to reduce the theories of his empiricist predecessors to the absurdity that was implicitly contained in them all along."2 Rather, he urges us to follow the watershed work of Norman Kemp Smith3 and view Hume primarily as a naturalist—i.e., as a philosopher who approached the study of human nature in roughly the same way that Newton studied the physical world.4 Insofar as Stroud and Smith are correct, it seems reasonable to consider Hume a forerunner of what today is called "naturalized epistemology."5 In this paper, I will argue that even if we can label Hume as a naturalized epistemologist, the "traditional" construal of him as a sceptic is in some sense much more apt because the philosophical system that he constructs has far-reaching sceptical implications.6 Before arguing for my interpretation, however, let me briefly outline how I will proceed. In section I, I will define scepticism more precisely and provide a prima facie case for a sceptical interpretation of Hume's system. The burden of section II will be to argue that the standard "naturalist" attempts to counter this sceptical reading are misguided. Before concluding, I will examine (in section III) other naturalistic readings of Hume in the context of naturalized epistemology to show why my sceptical interpretation is preferable. Kevin Meeker is at the Department of Philosophy, Humanities 124, University of South Alabama, Mobile AL 36688 USA. 32 Kevin Meeker I. The Prima Facie Case for a Sceptical Interpretation of Hume A. Hume's Attack on Reason. In this paper, I will understand scepticism as a thesis denying that humans have knowledge. Philosophers often associate Hume's alleged scepticism with some particular topic.7 That is, they view Hume as sceptical about inductive inferences, miracles, and so on.8 But I will argue that Hume's system leads to a more thoroughgoing scepticism because his system is sceptical with regards to knowledge claims in all areas (what some call global scepticism). The main passage upon which I want to focus to support my interpretation is found in I iv 1 of A Treatise of Human Nature, which is entitled "Of scepticism with regard to reason."9 Although I will not critically analyze Hume's argument for scepticism, we should familiarize ourselves with, its reasoning for two reasons. First, it will reveal why Hume believed that this argument had momentous consequences for philosophy. Second, it will help us consider how others try to reconcile a naturalist reading of Hume with this section. So let us turn to an exposition of the text. Hume's first major point in this section is that it is always possible for us to be wrong about a specific judgment. While this is obvious in perceptual cases, Hume insists that even in areas such as simple arithmetic we can still make mistakes. Now his point is not that the fundamental theorems of arithmetic are not necessarily true; rather, his point is that our epistemic grasp of such principles is always fallible. That is, all beliefs have some degree of uncertainty attached to them. Or, in Hume's own illustrious terms: "all knowledge resolves itself into probability..." (T 181). Hume's second point is that our beliefs do not arise ex nihilo; they have their source in one of our cognitive faculties. Consequently, according to Hume, when we try to estimate the probability that a particular belief is true we should not only evaluate that belief per se, but also compute the reliability of the faculty that produced that belief. Because all of our beliefs are fallible, the probability calculus dictates that we assign both the probability that a particular belief is true and the probability that that particular belief is true given the reliability of the faculty that produced it a value of less than one. But elementary mathematics also tells us that when you multiply two real numbers that are less than...