Hume's Second Thoughts on the Self

36. HUME'S SECOND THOUGHTS ON THE SELF* 1_. Although the appendix in which Hume confesses disillusionment with the Treatise theory of personal identity is very puzzling and confusing, there have been few serious attempts to explicate it. Wade L. Robison's recent paper, "Hume on Personal Identity," goes a long way toward making up for this lack, and I concur with much of what Robison says. Nonetheless, I think further light can be shed on Hume's thinking, and I will attempt to do so in what follows . 1_. Hume reports that a review of his discussion of personal identity reveals problems so severe and difficult that he does not "know how to correct my former opinions , nor how to render them consistent." After summarizing the arguments that led him to the Treatise view of personal identity, he offers this explanation of the discovery inconsistency : In short, there are two principles which I cannot render consistent, nor is it in my power to renounce either of them, vis. that all our distinct perceptions are distinct existences, and that the mind never perceives any real connection among distinct existences. (636)¿ The trouble with this explanation is that there appears to be no inconsistency at all between these two principles. Many commentators have simply passed over this fact, perhaps because Hume's theory of self-identity has seemed so preposterous that they have felt it obvious that he would be unhappy with it. * This paper grew out of my contribution to a colloquium at the 1974 Eastern Division Meeting of the American Philosophical Association. I would like to express my thanks to Norman Melchert, whose paper stimulated my werk on this topic. 37. Others -- such as Norman Kemp Smith and John Passmore, in addition to Robison -- have claimed that Hume was incorrect in placing the inconsistency and have suggested more suitable candidates for the source of Hume's doubts. This failure to take the author at his word about what disturbed him may appear too presumptuous a move for a commentator to make, but the appendix provides strong evidence that Hume was quite confused and could easily have misstated or misidentified his problem. For example, Hume begins by saying that he will offer the arguments "on both sides" -- that is, the arguments that led him to adopt this theory of personal identity and those now causing him to doubt it. In fact, however, while he gives a neat summary of the arguments for the Treatise theory, he says nothing about the grounds for his new doubts. In addition, the statement of the two principles is introduced by the words "in short," as if Hume were providing a mere summary of what he had just said. The preceding passage, however, makes no mention of and is in no way concerned with the two principles. What is offered as a summary is no summary at all. Added to the apparent consistency of the two principles, these facts provide strong grounds for thinking Hume was quite confused. As such, they support those commentators who have felt forced to look elsewhere for the source of Hume's dissatisfaction. Robison' s most concise statements of his own interpretation of the problem are as follows: Hume's explanation of why we think we have an idea of the self depends upon there being a self... But he has argued that we can have no idea of such a self so that we cannot conceive it as existing, let alone guarantee its existence. (190) ...we all think we have an idea of the self... Hume has to explain this fact, and ...the sort of explanation he used appeals to the existence of an active self distinct from any bundle of perceptions and propensities ... (192) Robison is correct, I think, on the basic point: the inconsistency is between Hume's bundle analysis of the self and his explanation of why we mistakenly believe there is a per- 38. sistent self. I differ with him, however, on two points. First, I think that his contrast between an active mind and a set of propensities is incorrect. That the Humean self has propensities is itself sufficient to generate Hume's problem...