Abstract

Before the pioneering work of Norman Kemp Smith, most Hume scholars read him as a thoroughgoing skeptic. The dominant view today is that, for Hume, ‘natural beliefs’—paradigmatically, beliefs based on induction—are warranted in virtue of features of the psychological mechanisms that produce them; moreover, Hume would endorse a suitable naturalistic theory of warrant to sustain this position. I survey four naturalistic interpretations of Hume’s epistemology: Kemp Smith’s theory, proper-function theory, stability theory, and reliabilism. I do not argue for one of these interpretations over the others; instead, I focus on what they have in common: Hume provides a naturalistic response to any generalized skepticism. From within this broad interpretive framework, some commentators argue that Hume would extend the class of ‘natural beliefs’ to religious belief. The bulk of the evidence supporting this position is derived from the Dialogues; in particular, commentators argue that, instead of being supported by the argument from design, there is a natural propensity that causes one to form the belief in an intelligent designer upon noticing the order and regularity in the world. I argue that the evidence is insufficient to support the claim that, according to Hume, religious belief is a ‘natural belief’. I examine Hume’s Natural History, where he provides an account of the origin of religious belief, in conjunction with his epistemological observations about various belief-forming mechanisms in the Treatise. I show that, no matter which theory of naturalistic epistemology best fits Hume’s own, religious belief is not warranted naturalistically. Furthermore, I argue that on Hume’s view, polytheism, while still unwarranted, is epistemically superior to monotheism. I conclude that, for Hume, the psychological explanation of religious belief, in conjunction with the fact that religious belief cannot be warranted on the basis of any evidence or a priori or a posteriori argument, provides grounds to reject all forms of religious belief. The Natural History is best read as an important piece of a larger destructive project which has as its goal showing that religious belief is not warranted by any means—through reason or experience, by revelation, or by its naturalistic explanation.