Abstract

This dissertation provides a systematic account of Hume's bundle theory of the nature and identity of persons. What distinguishes this project from the existing scholarship on this subject is, first, its attention to topics that have been given short shrift despite their importance for understanding and appreciating the force of Hume's theory and, second, its sustained engagement with a range of standard objections to the view. ;In Chapter I, I set out the competing substance theory of the self, then examine Hume's arguments against this theory. In Chapter II, I explore three crucial notions of identity developed by Hume. By discussing Hume's account of perfect identity in light of his theory of time, I show that this is for Hume a genuine notion, one based on the way we apprehend objects. I then delineate the distinctions Hume develops between different types of identity so as to make sense of his positive account of the identity of changing things, including persons. ;In Chapter III, I offer a detailed interpretation of the bundle theory, then turn to the vexed question of what prompts Hume, in the Appendix to the Treatise, to confess himself caught in a labyrinth of difficulties. I chart a course through this labyrinth, critically engaging a number of readings of the problem of the Appendix. This clears the way to a more successful interpretation while at the same time allowing me to defend the bundle theory against a number of charges. ;A final objection to the bundle theory is the Butlerian criticism that if the theory is true, past- and future-directed self-concern are unjustified. In Chapter IV, I reply to this objection by sketching a Humean account of past- and future-directed self-concern. This requires the resolution of a number of issues, among them the basis for Hume's distinction between personal identity as it regards the thought and imagination and as it regards the passions, and the sense in which the latter "corroborates" the former. Addressing these issues reveals a theory possessed of far more breadth and complexity than has frequently been acknowledged