1 Hume, David A Treatise of Human Nature, Selby-Bigge, L.A. ed. (Oxford: Clarendon Press 1888)Google Scholar; 2nd ed. revised by Nidditch, P.H. (Oxford: Clarendon Press 1978)Google Scholar. All subsequent references to Hume are to the Treatise and, unless I indicate otherwise, are to section and page numbers of Book I, Part iv.

2 Hume proceeds as well in Section 2 to explain how the philosophical concept of the ‘double existence’ arises out of a rejection of the ordinary notion (211-16). Furthermore, as the section’s title suggests, he argues that the ordinary view that one directly perceives objects is false (210-11) and that the philosophical theory that some impressions are caused by objects they resemble is unverifiable (212). However, my focus is on Hume’s account of the ordinary notion of objects that arises in the imagination as a virtually automatic response to having certain impressions.

3 See, for example, Smith, Norman Kemp The Philosophy of David Hume (London: Macmillan 1941), 471CrossRefGoogle Scholar, n. 2, and Stroud, Barry Hume (London: Routledge & Kegan Paul 1977), 100Google Scholar.

4 Stone, Jim ‘Hume on Identity: A Defense,’ Philosophical Studies 40 (1981), 277CrossRefGoogle Scholar

5 Rene Descartes, Meditations on First Philosophy, Med. VI, in The Philosophical Writings of Descartes, John Cottingham, Robert Stoothoff, and Dugald Murdoch, eds. And trans. (Cambridge: University Press 1984), Vol. ll., 52; John Locke, An Essay Concerning Human Understanding, IV. xi. 5 in Nidditch, Peter H. ed. (Oxford: Clarendon Press 1975), 632Google Scholar; and George Berkeley, Three Dialogues Between Hylas and Philonous, Third Dialogue, in The Works of George Berkeley, Luce, A.A. and Jessop, T.E. eds., Vol. ll (London: Thomas Nelson & Sons 1949), 235Google Scholar

6 Descartes, Med. VI, 61-2; Locke, Essay, IV. xi. 7, 633-4; and Berkeley, Third Dialogue, 235

7 For a detailed defense of this interpretation, see Kornegay, R. Jo ‘Hume on Identity and Imperfect Identity,’ Dialogue 24 (1985) 213-26CrossRefGoogle Scholar.

8 For explications of the imagination’s role in the production of the fictitious continuant, see Smith, Kemp The Philosophy of David Hume, 473-8Google Scholar and McRae, Robert ‘The Import of Hume’s Theory of Time,’ Hume Studies 6 (1980), 124-7Google Scholar.

9 Consider, for example, ‘We find by experience, that there is such a constancy in almost all the impressions of the senses, that their interruption produces no alteration on them, and hinders them not from returning the same in appearance and in situation as at their first existence’ (§2, 204; Hume’s emphasis).

10 For a provocative, critical analysis of Hume’s causal inferential account, see Bennett, Jonathan Locke, Berkeley, Hume: Central Themes (Oxford: Clarendon Press 1971), 322-33Google Scholar.

11 For a detailed analysis of the porter case, see Price, H.H. Hume’s Theory of the External World (Oxford: Clarendon Press 1940), 50-9Google Scholar.

12 See H.H. Price for a detailed and schematized account of this type of reasoning whereby one postulates the missing items of what he calls ‘gap-indifferent’ series in order to complete the pattern found among similar series of varying degrees of incompleteness (Hume’s Theory of the External World, 60-4).

13 Hume might object to this label. Though he does refer to this inference as ‘a kind of reasoning from causation’ (§2, 195), he writes that it is not based on habit formed by observed constant conjunctions. In contrast to strict causal inference, this inference ‘arises from custom in an indirect and oblique manner’ (§2, 197). In it the imagination completes patterns found spottily in experience. One infers a greater degree of regularity than one ever senses. ‘[A]s the mind is once in the train of observing an uniformity among objects, it naturally continues, till it renders the uniformity as compleat as possible’ (§2, 198). Standard causal inference is based on observed constant conjunction of states or events and a habit to expect the other usual member of a pair on the basis of observation of the one. This extended, causal inference is based on observed imperfect patterns of states or events from which the imagination completes the template. It is not based on habit, but on what H.H. Price called ‘the inertia principle.’ For a classic account of how this inference differs from standard causal inference see Price, Hume’s Theory of the External World, 53-9. For a recent analysis see Wilson, Fred ‘Hume’s Fictional Continuants,’ History of Philosophy Quarterly 6 (1989), 174-6Google Scholar.

14 Note that Hurne offers what I term the ‘central account’ discontinuously. Thus, I am in effect restructuring Hurne’s analysis by emphasizing that the account of constancy in Section 2, pp. 199-218 and that of coherence in Section 6, pp. 253-5 are sub-parts of one genetic account of how misascriptions of identity to constant series, on the one hand, and to coherent series, on the other, lead to the notions of objective things and events.

15 Some scholars miss this point as well. For example, both Popkin, Richard (‘David Hume:His Pyrrhonismand His Critique of Pyrrhonism’in Chappell, V.C.,ed.,Hume (Garden City, NY: Doubleday 1966), 64-7, 77-9)Google Scholar and Wilson Fred (‘Hurne’s Fictional Continuants,’ 171-80) fail to appreciate it when they address the issue of whether Hurne holds that the vulgar belief in objects’ continuance has a rational basis. Popkin contends that Hurne denies a rational basis. Popkin ignores the causal inferential account, which puts a strain on his reading. Wilson, who argues against Popkin’s reading, holds that Hurne takes this vulgar belief to be rationally justifiable. Wilson emphasizes the causal inferential account. However, he restricts his discussion to Section 2 and thereby ignores the account of coherence in Section 6, which comports better with Popkin’s skeptical interpretation.

16 Any serious scholar should contend with Hume’s dismissal of the causal inferential account as ‘too weak to support alone … the continu’d existence of all external bodies’ (§2, 198-9). Barry Stroud, for example, somewhat plausibly suggests that this claim means that one can originally develop the concept of objects only as a reaction to constant series. However, once one develops the notion, one uses the hypothesis of continuing, mind-independent things to explain the coherence of series also (Barry Stroud, Hume, 100 and 259-60, n. 4). Thus, one may originally develop the notion of, and belief in, the continuance and mind-independence of, say, one’s coal and grate and thereafter suppose that the fire also exists objectively. In part IV, section 2 below I expound a deep reason for the incompleteness of the causal inferential account.

17 There are special circumstances (e.g., if there were a loud, interfering noise) in which one would posit unheard notes.

18 I am indebted to an anonymous referee of CJP for pointing this out.

19 Whether or not Humean ‘temporal contiguity’ is the same notion as temporal continuity or what Hume terms; ‘uninterruptedness’ is not beyond debate. My interpretation is that series that exhibit temporal contiguity include series of temporally continuous items as well as series of sequential items punctuated by noticeable temporal gaps. My textual evidence comes from what I call the ‘central account’ and includes Hume’s claims that ‘[h]owever at one instant we may consider the related succession as variable or interrupted, we are sure the next to ascribe to it a perfect identity, and regard it as invariable and uninterrupted’ and ‘the objects, which are variable or interrupted, and yet are suppos’d to continue the same, are such only as consist of a succession of parts, connected together by resemblance, contiguity, or causation’ (Treatise, I. iv. 6, 254 and 255, respectively). These passages I take to mean in part that a series of ‘variable’ items (those related by inexact resemblance or causation, such as expected gradual differences) and of ‘interrupted’ items (related by temporal contiguity) can be confused with a single ‘invariable’ item (the stages of which are exactly similar) that is also an ‘uninterrupted’ item (the stages of which are temporally continuous).

20 Hume alludes to the same sort of reasoning when he writes, ‘One who concludes somebody to be near him, when he hears an articulate voice in the dark, reasons justly and naturally; tho’ that conclusion be deriv’ d from nothing but custom, which infixes and inlivens the idea of a human creature, on account of his usual conjunction with the present impression’ (§4, 225).

21 This conclusion supports Hume’s claim (§2, 198-9) that causal inference cannot account for the objectivity of all events so considered.

22 Not just any causal relation will suffice, for one needs to rule out illusory impressions. If action by an hallucinogen causes the alleged impressions of the tune, the experience has an objective cause. Yet it is not thereby rendered an objective experience. The right sort of cause must involve action on the sensory apparatus in the usual way.

23 My adoption of this strategy is inspired in part by P.F. Strawson’s work on the objectivity of sounds in Individuals (London: Methuen 1959), Ch. ll, ‘Sounds,’ 58-86, esp. 59-74. Strawson argues that one should explicate the objectivity of sounds by reference to a spatial system of objects in which events occur. One can explicate the objectivity of sounds in terms of relations to material objects, but not vice versa. The objectivity of material objects is basic, according to his descriptive metaphysics. On Hume’s behalf, I give a genetic account of the objectivity of sounds, analogous to Strawson’s logical account, in which the careers of external objects are basic. The strategy seems promising, not only because I believe that it is a sound one, but also because Hume does provide an account of the objectivity of material objects’ careers which can underpin this analysis.

24 In part V, I further contend that neither constancy nor coherence is a sufficient condition for being considered an objective experience.

25 The ascription of spatial properties to sounds, odors and tastes is yet another fiction of the imagination, viz., a notion whereby an idea, e.g., spatial location or dimension, is applied to an idea, i.e., sounds, odors and tastes, from which it could not originally be derived. The notion is a construction of the imagination, not a copy of some impressions. Hume proceeds in Section 5 to argue that this commonsensical notion is objectionable on philosophical grounds as it has alleged incoherent implications, e.g., that tastes and odors are shaped (§5, 238-9).

26 This challenge was put to me by John Wright in correspondence.

27 Bricke comes to a similar conclusion about one of Hume’s difficulties in accounting for the notion of mind as a discontinuous, time-consuming entity. He writes, ‘Minds must be taken to contrast with bodies in suitable ways. At the very least they must differ on the point of independence. And that means that the observational situations that give rise to the notion of mind cannot possess the gap-indifference [that feature which makes one ascribe identity while failing to notice the interruptions] that is the mainstay of Hume’s account of the idea of bodies’ Bricke, John Hume’s Philosophy of Mind (Princeton: Princeton University Press 1980), 80-1Google Scholar; his emphasis).

28 Of pleasures and pains (bodily feelings which are not appetites), Hume explicitly claims, ‘The third [kind of impression of sensation] are the pains and pleasures, that arise from the application of objects to our bodies, as by the cutting of our flesh with steel, and such like …. Both philosophers and the vulgar … esteem the third to be merely perceptions; and consequently interrupted and dependent beings’ (§2, 192).

29 Interestingly, Hume does not clearly sort out the unpleasant, organic feelings (e.g., hunger-pangs and feeling parched) from the desires to eat and to drink to relieve the discomfort, as Hutcheson had done before him. The latter differentiated between the ‘uneasy Sensations’ which precede and accompany the appetitive desires of hunger and thirst proper (Frances Hutcheson, An Essay and Conduct of the Passions and Affections, 1728; 3rd ed., 1742; facsimile with introduction by Paul McReynolds (Gainesville, FL: Scholars’ Facsimiles & Reprints 1969), 52, 91-3). Rather, Hume conceives of the bodily appetites of hunger, thirst, and lust as organic impressions that have physiological causes, internal to one’s body. He writes, ‘The palate must be excited by an external object, in order to produce any relish: But hunger arises internally, without the concurrence of any external object’ (Treatise, II. i. 5, 287). Moreover, these bodily appetites have ideas of food, drink, and sex as effects. ‘The sensations of lust and hunger always produce in us the idea of those peculiar objects, which are suitable to each appetite’ (Treatise, II. i. 5, 287). Hence, for Hume, hunger, thirst, and lust are unpleasant impressions that are causally, not logically, related to the ideas of food, drink, and sex. Given Hume’s analysis of bodily appetites and their origins, one would expect him to classify them, along with bodily pains and pleasures, as impressions of sensation. Indeed, this is what he does in the first book of the Treatise when he introduces the distinction between impressions of sensation and impressions of reflection. Hume writes, ‘An impression first strikes upon the senses, and makes us perceive heat or cold, thirst or hunger, pleasure or pain of some kind or other’ (Treatise, I. i. 2, 7-8). However, Hume confusedly classifies the bodily appetites among the passions or violent emotions, along with the calm emotions, as impressions of reflection in the second book (Treatise, II. i. 1, 275; II. iii. 9, 439). This classification is made despite the fact that in Book II he draws the distinction between the two types of impressions as follows: ‘Original impressions or impressions of sensation are such as without any antecedent perception arise in the soul, from the constitution of the body, from the animal spirits, or from the application of objects to the external organs’; and ‘[s]econdary, or reflective impressions are such as proceed from some of these original ones, either immediately or by the interposition of its idea’ (Treatise, II. i. 1, 275). Since hunger, thirst, and lust, as Hume analyzes them, lack impressions of sensation or ideas as causes, they cannot coherently be categorized as passions, a type of impression of reflection. This inconsistency has been noted by Alfred B. Glathe in Hume’s Theory of the Passions and of Morals (Berkeley: University of California Press 1950), 30; and by PallS. Ardal in Passion and Value in Hume’s Treatise (Edinburgh: Edinburgh University Press 1966), 9.

30 N.B.: as note 29 explains, hunger-pangs and feeling parched, which Hume takes hunger and thirst to be, are included in his category of passions.

31 See Treatise, IT. i. 2, 277.

32 Alternatively, one might suppose that Hume does believe the ordinary person locates one’s pains and pleasures in one’s body, say, a burning sensation in one’s hand. Accordingly, Hume’s task would be understood somewhat differently. He would need to explain why bodily pains and pleasures are externally located, yet not taken to be mind-independent.

This alternative interpretation raises a number of issues. Do ordinary people attribute physical locations to their bodily pains and pleasures? Does Hume suppose they do? Does one’s answer to the latter question raise a serious challenge to my analysis of Humean subjectivity?

No doubt people say, for example, ‘the pain is in my hand.’ Such expressions can be interpreted in at least two ways: (1) the pain seems to be in my hand, viz., is in phenomenological space and (2) the pain is located literally in my hand in the same way as bones are in my hand, viz., in physical space.

How Hume would decode such expressions is not easily decided by scrutiny of the text. However, a case can be made that Hume would accept (1), not (2). First, Hume does pry apart phenomenological and physical space when he writes that sounds, tastes, and odors do not seem to be in space, yet the ordinary person locates them in physical space (§2, 191; §5, 235-8). Moreover, he would not want to hold that one automatically infers physical location from apparent location since ordinary people conceive of dreamt or hallucinated, visual, or tactile things as only in phenomenological space, not physical space. Thus, Hume could consistently hold that in ordinary thought bodily pains and pleasures are viewed as in phenomenological space, but not as externally locatable.

Second, Hume does contrast ‘internal impressions,’ which subsume bodily pains and pleasures as well as emotions, with ‘external objects’ (§2, 194-5). Granted, one could read this as the contrast between internal to the mind (= mind-dependent) and external to the mind(= mind-independent), as this alternative interpretation would. However, a more natural reading would be to take this as the contrast between the mind-dependent and unlocated in physical space, on the one hand, and the mind-independent and located in physical space, on the other. Furthermore, when Hume introduces ‘distinctness,’ he states that it covers both external location and mind-independence (§2, 188). There is no passage in which he suggests that ordinary people take some impressions to be distinct in only one of these ways, e.g., externally located but dependent on someone’s mind for existence.

Fortunately, there is little at stake here to undermine my main argument about Humean subjectivity. In either case, Hume would need to debar the move from location in phenomenological or physical space of bodily pains and pleasures to mind-independence. In either case, Hume could argue that these pains and pleasures are not gap-indifferent, viz., there is no need to hypothesize unfelt stages to make these impressions conform to usual causal patterns.

33 See note 28 above.

34 Bodily pains and pleasures are classified as ‘impressions of sensation [which] are such as without any antecedent perception arise in the soul, from the constitution of the body, from the animal spirits, or from the application of objects to the external organs’ (Treatise, II. i. 1, 275). These ‘depend upon natural and physical causes’; hence, their explanations lie not in Hume’s science of human nature but in ‘anatomy and natural philosophy’ (Treatise, II. i. 1, 276).

35 I am indebted to anonymous referees and an editor of CJP for suggestions on how to shorten and improve this paper.