Beyond Our Senses: Recasting Book I, Part 3 of Hume's Treatise

Hume Studies Volume XX, Number 2, November 1994, pp. 241-259 Beyond Our Senses: Recasting Book I, Part 3 of Hume's Treatise SAUL TRAIGER The early sections of Book I, Part 3 of A Treatise of Human Nature1 are widely studied, and with good reason. They contain Hume's skeptical arguments about what we now call inductive inference or what Hume called reasoning from experience. Very little attention, however, has been paid to Hume's extensive treatment of the social context of belief formation and correction which dominates Sections iv-xiii of Part 3. When these sections are noticed at all, they are seen as, at best, embellishments and digressions, at worst, as unfortunate muddles. Purified of Sections iv-xiii, Part 3 is simpler (and shorter) than the actual Part 3; its arguments are well understood and their influence on Kant and the subsequent history of philosophy is well documented.21 will purposefully dirty the waters by attempting to account for the rich examples and the explicit treatment of the social component of belief in these neglected sections. While there are excellent reasons for stressing the importance of Hume's skeptical arguments, the text shows that Hume had a second major concern, namely to account for the formation and regulation of beliefs as they typically arise in social contexts. What exactly is at stake? It's easy to demonstrate that Hume treats belief in social contexts in a significant chunk of Part III. There is extensive discussion of such phenomena as testimony, credulity, and education throughout . It does not follow from this, however, that the analysis of testimony, for example, is central to Hume's epistemology. As Part 3 is usually interpreted, Hume's core account of belief formation and revision is decidedly non-social. Saul Traiger is at the Department of Philosophy, Occidental College, Los Angeles, CA 90041. Internet e-mail: traiger@oxy.edu 242 Saul Traiger Beliefs are formed and corrected by one's private stock of perceptions. On this interpretation, the social contexts are just fluff in Hume's account; they provide interesting but eliminable examples. Cases of belief formation through testimony, on this view, can be accounted for by a theory of causal inference which makes no reference to social factors, to other believers and their perceptions. It is my contention that Hume's purpose in Book I, Part 3 is to offer a theory of belief which does justice to the social nature of belief formation and correction. Most beliefs are transmitted through testimony and education. We tend to believe what we are told; yet we can also reject testimony on reflection. While causal inferences, like those from flame to heat, sometimes make reference only to our own perceptions, most of the time those inferences take us "beyond our senses" by inferentially connecting us to the impressions and ideas of others. In what follows I assess the importance of testimony and other social elements in Hume's epistemology. Social elements make reference to persons and perceptions other than those of the epistemic agent. I will highlight and attempt to account for Hume's attention to social-epistemic contexts in Part 3 of Book I, beginning with Hume's claim that causal inference takes us "beyond our senses." The claim can be read both individualistically and socially. I argue that it must be interpreted socially in at least some key passages . A further claim is that Hume not only illustrates causal inference with social examples, but makes use of the complexity of such examples to account for the inference. When our causal inferences involve testimony, as they often do, the impressions from which the inferences are made are of a special sort. They are impressions of words or utterances. Hume pays special attention to this type of causal inference when discussing the difference between reading fiction and history at T 97-98. This and neighboring passages on the interpretation of the words and utterances of others will figure prominently in my interpretation of the core of Book I, Part 3. Some Preliminaries: Knowledge and Probability Part 3 is entitled "Of knowledge and probability." Its sixteen sections span 107 pages of the Selby-Bigge/Nidditch edition...