1 Hume, David, A Treatise of Hitman Nature, ed. Selby-Bigge, L. A. (Oxford: Clarendon Press, 1888; reprint edition, 1975). Hereafter cited as “THN”.Google Scholar

2 By “a serially uniform item”, I mean an item whose temporal stages are exactly alike in kind.

3 The naturalistic aspect of Hume's philosophy was first emphasized by Norman Kemp Smith in The Philosophy of David Hume (London: Macmillan, 1941).Google ScholarComparatively recently, Barry Stroud has accentuated Hume's naturalism in his Hume (London: Routledge & Kegan Paul, 1977)Google Scholar.

4 I use the term “item” because it is relatively free of ontological connotations. I avoid the terms “thing” and “object” since they are too easily interpreted as material thing and physical object. I want to keep the question of what sorts of items Hume's conditions are designed to individuate open. Though I have not the space to argue it here, my own view is that Hume restricts his conditions to Humean perceptions or qualia.

5 Hume writes, “We have a distinct idea of an object, that remains invariable and uninterrupted thro'a suppos'd variation of time; and this idea we call that of identity or sameness” (THN, I. 4. §6, 253) and “Thus the principle of individuation is nothing but the invariableness and itninterruptedness of any object, thro'a suppos'd variation of time, by which the mind can trace it in the different periods of its existence, without any break of the view, and without being oblig'd to form the idea of multiplicity or number” (ibid., I. 4. §2, 201).

6 For example, Hume comments, “…we must attribute a perfect identity to this mass [of matter], provided all the parts continue uninterruptedly and invariably the same…” (ibid., 1. 4. §6, 255).

7 Cf. Hume's comment on the origin of the attribution of identity to one's mind or self, “…our notions of personal identity, proceed entirely from the smooth and uninterrupted progressof the thought alongatrain ofconnected ideas…” (ibid., 1. 4. §6,260).

8 Identity is also on occasion misascribed to so-called “constant” series. The successive members of a constant series are exactly alike in kind and intermittent. A constant series, thus, satisfies the condition of serial uniformity but fails to satisfy the condition of temporal continuity. In this type ofcase, the mind fails to distinguish clearly between temporal continuity (as in bona fide exemplification of identity) and mere temporal contiguity. See ibid., 1. 4. §2, 199, 204–205.

9 Leyden, W. von, “Hume and 'Imperfect Identity'”, Philosophical Quarterly 7/29 (10 1957), 340–352. Hereafter cited as “von Leyden”.CrossRefGoogle Scholar

10 Penelhum, Terence, “Hume on Personal Identity”, The Philosophical Review 64 (1955), 571–589; reprinted in V. C. Chappell, ed., Hume (Notre Dame, IN: University of Notre Dame Press, 1966), 213–239.CrossRefGoogle Scholar

11 Patten, Steven C., ”Change, Identity and Hume“, Dialogue 15/4 (12 1976), 664–672. Hereafter cited as “Patten”.CrossRefGoogle Scholar

12 Ashley, Lawrence and Stack, Michael, “Hume's Theory of the Self and Its Identity“, Dialogue 13/2 (06 1974), 239–254. Hereafter cited as “Ashley and Stack”.CrossRefGoogle Scholar

13 Cf. Hume's comment about ascriptions of identity to the series of perceptions constituting a mental history or mind, “The identity, which we ascribe to the mind of man, is only a fictitious one, and of a like kind with that which we ascribe to vegetables and animal bodies” (THN, 1. 4. §6, 259).

14 Patten, moreover, is inclined to hold that imperfect identity is a kind of identity for Hume (Patten, “Change, Identity and Hume”, 670–671).

15 Hume nowhere straightforwardly asserts that an instance of imperfect identity is an item to which identity is misascribed. Hence, there is no passage to which one can point to settle the debate over how imperfect identity relates to identity. Isolated passages should be interpreted in accordance with the best overall account of Hume's procedure and shifting philosophical aims in §6. In part four of this paper, I provide an overall account which makes the various passages fall into place as parts of a coherent philosophical investigation.

16 l(a) above. 214–215.

17 Cf. Hume's genetic analysis of the concept of identity through time (THN, 1. 4. §2, 200–201).

18 See ibid., 1. 4. §6, 260.

19 In order to isolate the subject matter for his naturalistic study of ordinary thought and language, Hume on occasion adopts the standpoint of an ordinary person and speaks with the vulgar. Note his use of the pronoun “us” in the passages cited above, 218, from ibid., 1. 4. §6, 257–258.

20 See Hume's account of the genesis of the vulgar fiction of the “continu'd existence”, ibid., 1. 4. §2, 199, 208–209; 1. 4. §6, 253–256.

21 Penelhum also holds that for Hume ascriptions of identity to series or time-extended totalities are misascriptions. His view of what Hume is up to in his work on identity, however, differs strikingly from my own. Penelhum avers that Hume intends to give an account of the ordinary notion of identity-preservation and ordinary uses of “same”. Hence, he accuses Hume of misdescribing the conceptual and linguistic facts. Hume supposes that there is one set of conditions by which every kind of thing can be individuated and that one can arrive at this general set by analyzing uses of “same x”. But, Hume is mistaken. There is no single set of identity conditions. Identity criteria are concept-relative (Penelhum, in Chappell, ed., Hume, 221–228). As 1 understand Hume, he does not intend to put forth general conditions of identity. Rather, he restricts his conditions to qualia or Humean perceptions. Hume's criteria individuate the basic items of his revisionary ontology—perceptions—from which the mind constructs fictions, e.g., persons, trees, and commonwealths. Consequently, Penelhum is misguided in using ordinary concepts of enduring things and uses of “same x of kind k” as the standard by which to assess Hume's analysis of identity.

22 Undoubtedly, Hume has borderline cases (for which there are no conventions prescribing use) in mind when he comments that the use of “identity” in the extended sense is a verbal matter only. I, however, see no reason why Hume would not view uses of “identity” in the extended sense in standard or paradigm cases (for which there are conventions) also as of little or no philosophical importance as long as no fictions exhibiting identity are posited.

23 Supra, 214–215.

24 Supra, 215.

25 Patten suggests that Hume uses a second sense of “identity” which he applies to fictions (Patten. “Change, Identity and Hume”, 670–672). I concur with Patten that the series to which identity is ascribed are fictions that the imagination bundles together. Patten, however, like Ashley and Stack, does not appreciate the limited sanction from Hume. He has notning to say about Hume's semantic objections. Patten's comments are cursory, and he only hints that this second sense might be understood by mechanisms of Hume's naturalism. Unlike Patten, I provide an account that integrates Hume's statements on identity into both his critical philosophy, including his developmental semantics, and his naturalism.

26 “Imperfect identity,” means l(b) imperfect identity. See above. 215.

27 Discussed above, 216.

28 In §2, in dealing with the source of misascriptions of identity to successive, related items, Hume only explicitly considers the relation of resemblance among successive items. See THN, 1. 4. §2, 204, 209. In §6, however, Hume considers two general categories of relation in addition to resemblance, namely, (spatial or temporal) contiguity and causation (ibid., I. 4. §6, 260). The account of the origin of confusions which result in misascriptions of identity given in §2, thus, is expanded and enriched in §6.

29 Hume adopts the ordinary person's perspective in order to discover the kinds of cases to which people ordinaril y ascribe identity. These are the cases that he studies and classifies from the standpoint of the naturalist. The adoption of the ordinary person's perspective, hence, is instrumental in fulfilling his aims as a scientist of human nature.

30 The first is subsumed by the general relations of resemblance and contiguity; the second by causation; while the third seems to be subsumed by resemblance, causation, and perhaps contiguity as well.