Hume's sense impressions

HUME’S SENSE IMPRESSIONS This paper is devoted to the nature and roles of sense impressions in Hume’s account of perception. At first sight Hume’s Treatise of Human Nature, a book written within the frame of reference of the Lockean ‘way of ideas’, is mostly devoted to the examination of ideas: ideas of memory and ideas of the imagination, general ideas, the ideas of space and time, the idea of cause and effect, etc. And it is true that in p. 8 of the Treatise Hume declares sense impressions to be beyond the bounds of his discussion, and only goes back again to issues directly relative to sensory perception as a source of knowledge towards the end of Book 1, in the 30-page section devoted to what he calls the ‘scepticism with regard to the senses’. Yet the pages of this work are full of vivid references to sense impressions: patches of colour such as a missing shade of blue, the purple surface of Hume’s own table, red and blue points, a hand spread against the blue colour of the firmament, the colours reflected in the clouds, a spot of ink, black characters on the white pages of history books—visual images prevail, but there are also the taste of a pineapple, the sweetness of a fig and the bitterness of an olive, the creaking of a door, a succession of notes on a flute, the warmth of the fire and the coolness of water, etc. Sense impressions are major protagonists of Hume’s theory of knowledge. I shall start by considering how Hume introduces sense impressions at the beginning of the Treatise. My purpose is to show that even though he explains the distinction between impressions and ideas on the basis of their different strength and liveliness, the crucial difference between them is in fact that ideas are copies of impressions, while impressions are their ‘originals’. Hume’s sense impressions, in turn, do not copy anything. They are what ideas represent, the objects of our thought. In his well-known study of Hume, after rehersing a number of criticisms against Hume’s treatment of the difference in strength and liveliness, Barry Stroud suggests, as a possible way to salvage the Humean distinction between impressions and ideas, that we could define impressions as the perceptions taking place in the presence of their objects, while ideas are the perceptions we have in their absence. But, he goes on, in this way the difference between feeling and thinking, also invoked by Hume in this connection, is still left unaccounted for.1 My reading of Humean sense impressions has to deal with two problems, very well summarised by Stroud’s criticisms: the first is, if impressions do not represent anything, how can Hume talk about ‘objects’ at all—in fact, what are Humean ‘objects’? Dealing with this problem is the main subject of the present paper. The second problem I shall only consider 1 See B. Stroud, Hume (London: Routledge and Kegan Paul, 1977), pp. 27-29. On impressions as the perceptions occurring in the presence of external objects see also Cass Weller, ‘Impressions of sense as perceptual takings’, forthcoming in Archiv für Geschichte der Philosophie. Also J. Bennett, Locke Berkeley Hume. Central Themes (Oxford: Clarendon Press, 1971), p. 225, for a crisp analysis spiced up with the usual rude remarks. 26/1/15 in a tentative manner at the end, and it has to do with the difference between feeling and thinking: why does Hume make such heavy weather of the different strength and liveliness of impressions and ideas, if their distinction is, in fact, based on their original-to-copy and object-to-representation relation? 1. Hume’s sense impressions and ideas For a start, Hume introduces what he calls impressions in real time, as the sensations and emotions that I, his reader, am having now, while perusing page 1 of his Treatise [apologies for this small fiction!]. These include the sight of these black marks on this white page, the feeling of the crispness of the corner of the page between my fingers and of the bulk of the fat volume I am holding in my hand, etc., and ‘the pleasure or uneasiness’ occasioned by all these sensations. ‘All our sensations, passions and emotions, as they make their first appearance to the soul’ (T/1) are impressions. These perceptions are immediate and compelling, commanding my instant unreserved assent—in other words, it simply does not cross my mind that I may not really be seeing the colours and shapes, and feeling the mass and textures that constitute my old copy of Hume’s Treatise. Impressions, as he puts it, ‘enter with most force and violence’ . Impressions are constantly accompanied by a second kind of perceptions: ideas, which are the mental copies of former impressions. ‘By ideas, Hume says, I mean the faint images of [impressions] in thinking and reasoning’. For example, at this moment—please remember that I am still reading p. 1 of the Treatise—they include everything which is going on in my head, except all that I listed a moment ago under the heading ‘impressions’: my reactions to Hume’s ‘discourse’, and presumably, as far as ‘images’ stricto sensu go, that of the dark green cover of my book—I cannot see it in this moment because I’m holding it, but I remember it distinctly from former sense impressions I had in the course of my earlier visual acquaintance with this copy of the Treatise.2 This is Hume’s so-called copy principle: all our ideas are faithful, if fainter, copies of former impressions. This principle presupposes that the umbrella term for all mental contents is not ‘ideas’, as standard within the Lockean frame of reference, but rather ‘perceptions’; Lockean 2 This discussion presupposes that in calling ideas ‘images’ of impressions Hume is using the term ‘image’ in a metaphorically extended sense. For the standard view that Hume literally equates ideas with mental images see for example J. Laird, Hume's Philosophy of Human Nature (London: Methuen & Co., 1932), pp. 26-7, and the lengthy and sensitive treatment in C. Maund, Hume's Theory of Knowledge (London: Macmillan & Co., 1937), pp. 165-85, esp. 169-70; also taken for granted in A. Flew, Hume’s Philosophy of Belief (London: Routledge and Kegan Paul, 1961), p. 22, and J. Bennett, Locke Berkeley Hume. Central Themes (Oxford: Clarendon Press, 1971), p. 222. For a full examination of this issue see my Space and the Self in Hume’s ‘Treatise’ (Cambridge: Cambridge University Press, 1998), ch. 1, pp.15-20. See also the excellent article by S. Everson, ‘The difference between feeling and thinking’, Mind 97 (July 1988): 401-13. 2 26/1/15 ideas are, Hume is saying, only one of the two kinds of perceptions. Hume is explicit in stating that this distinction between impressions and ideas introduces something new in contemporary philosophical idiom: ‘I here make use of these terms, impressions and ideas, in a sense different from what is usual, and I hope this liberty will be allowed me’. But then he also suggests—in a way which is somewhat typical of contemporary criticisms of Locke3— that perhaps this appearance of originality is illusory, by adding: ‘Perhaps I rather restore the word, idea, to its original sense, from which Mr. Locke had perverted it, in making it stand for all perceptions’ (T2n). A notorious problem with this distinction has to do with the exact characterisation of the two kinds of perceptions relative to one another. Hume’s first approach to this issue is apparently straightforward: ‘the difference betwixt these consists in the degrees of force and liveliness, with which they strike upon the mind, and make their way into our thought or consciousness’ (T1). These lines are in the very opening paragraph of the Treatise; and the same point is repeated, in more elegant terms, in the opening paragraphs of the Enquiry, sect. 2: ‘the most lively thought is still inferior to the dullest sensation... our thought is a faithful mirror, and copies its objects truly; but the colours which it employs are faint and dull, in comparison with those in which our original perceptions were clothed’ (E11). The possible objections to this approach are obvious. There are counterexamples with a solid early modern tradition at least since Descartes: as Hume is quick to admit, ideas can, sometimes, be pretty lively and strong. For instance, Hume mentions the classic case of very vivid dreams or of hallucinations, which it would be very hard to tell from impressions on the basis of their liveliness and strength—so much so, that in fact many of us would probably be inclined to regard them as impressions. On the other hand, many impressions may be particularly calm and faint (T2). Strength and liveliness provide a criterion which is both too vague to justify the distinction, and too ambiguous to make it actually work.4 And it is easy to construct cases of simple ideas which are not, in fact, copied from corresponding impressions by imagining a continuous gradation of a quality (Hume’s choice is the colour blue) with just one gap—wouldn’t our mind be able to fill the gap...?5 3 See for example An Essay on the Origin of Evil. By Dr. William King, late Lord Archbishop of Dublin, (trans. by Edmund Law), Cambridge, 1731, ch. 1, King’s note A, esp. p. 6, and Law’s defence of Locke’s notion of ‘idea’, note 3, p. 7. This text may have been known to Hume, who takes notes about it in his early Memoranda (see E.C. Mossner, “Hume’s Early Memoranda, 1729-1740: The Complete Text”, Journal of the History of Ideas 11 (1948): 492-518; but see also J. P. Pittion, ‘Hume's reading of Bayle: an inquiry into the source and role of the Memoranda’, Journal of the History of Philosophy 15 (1977): 373-386). 4 Stroud, Hume, pp. 27-9. 5 The ‘missing shade of blue’ is one of the classic teasers in Hume scholarship, esp. for positivist readers. For a recent and compelling discussion see D. Garrett, Cognition and Commitment in Hume’s Thought (Oxford: Oxford University Press, 1997), ch. 2, 'The copy principle', pp. 43-57, esp. pp. ??? 3 26/1/15 It is clear that Hume is well aware of all or most of this. But, he insists, it doesn’t really matter—for we know full well the difference, as he puts it, between feeling and thinking; and the so-called copy principle simply says that feeling always comes before, is more compelling than, and provides the material for the corresponding thinking—or conversely, that thinking is always after, less vivid than, and about the corresponding feeling. Hume is a famously artful writer. What he is saying here is not all that clear, and far from uncontroversial; but his rhetoric is reassuring. The message conveyed by his confident manner is that he is saying nothing challenging or less than obvious; we are all agreed, really, that all our ideas derive from experience.6 So if I want to know the taste of a pineapple I don’t conduct an interview of pineapple-eaters: the only way for me to acquire this piece of information is to try it myself. If I want to give a child the idea of a colour or a taste, I present him with the relevant impression. And we are in no doubt that a blind person, who cannot access colour impressions, cannot have ideas of colours, indeed cannot have any notion at all of what colours are (T5). For a post-Lockean all this seems pretty straightforward. In fact, the implications of what Hume is saying are far-reaching. Impressions are prior in time to the corresponding ideas; they are constantly conjoined with ideas; we are unable to form ideas without the corresponding impressions: in short we can conclude, in perfect conformity with Hume’s discussion of causation later in the book, that impressions cause ideas. In this sense, the so-called copy principle is Hume’s version of Locke’s theory of ideas: but while Lockean ideas copy or represent objects (e.g. E2.31.3, 4.4.3, 4.4.5), and are caused by a physical process through which objects affect our nerves and brain (e.g. E2.8.12),7 Humean ideas copy impressions, and are caused by processes that Hume declares he does not intend to discuss because ‘the examination of our sensations belongs more to anatomists and natural philosophers than to moral’ (T8).8 The pattern of systematic differences within the structural similarity of the two theories could be followed up in detail 6 See the splendid J.J. Richetti, Philosophical Writing: Locke, Berkeley, Hume (Cambridge, MA: Harvard University Press, 1983), ch. 4, pp. 183-263, for a detailed and illuminating discussion of rhetoric and philosophy in Hume’s writing. 7 See for example the discussions in J. L. Mackie, Problems from Locke (Oxford: Clarendon Press, 1976), ch. 2, pp. 37-71; and J.W. Yolton, A Locke Dictionary (Oxford: Blackwell, 1993), entries ‘Idea’, ‘Knowledge’, and ‘Representation’. 8 See also the introductory section of Book 2, T275-6. What he means here is evident from a handful of occasional and ultra-cautious references he makes to the workings of the brain—the most extended example is in T60-61 (when discussing why we tend to confuse resembling ideas), but see also T98-9, T123, or T185. On the reception of this Malebranchean version of Cartesian physiology in England and on Hume’s use of it see John Wright, The Sceptical Realism of David Hume (Manchester: Manchester University Press, 1983), esp. pp. 68-71, 204-14. 4 26/1/15 on a number of issues and problems. To mention just one of them as an example, Hume is explicit in pointing out that his distinction and correlation of impressions and ideas involves a version of Locke’s famous rejection of innate ideas. All arguments against innate ideas, he says, boil down to proving ‘nothing but that ideas are preceded by other more lively perceptions, from which they are derived, and which they represent’ (T7); or indeed, as he puts it in the Enquiry: ‘by admitting these terms, impressions and ideas, in the sense above explained, and understanding by innate, what is original or copied from no precedent perception, then may we assert that all our impressions are innate, and our ideas not innate’ (E17n). In other words, sense impressions ‘arise from the soul originally, from unknown causes’ (T7); and this is why by ‘impressions’ he intends to refer not to ‘the manner, in which our lively perceptions are produced in the soul, but merely the perceptions themselves’ (T2n).9 Let me try to put all this slightly differently. A Lockean account of knowledge duplicates the world: roughly speaking, the real world, the world of external objects is reproduced in the mind in a world of ideas which represent those objects. Hume's copy principle is a version of this duplication of the world—all perceptions, he says, come in pairs: ‘All perceptions of the mind are double, and appear both as impressions and as ideas’ (T2-3). But, unlike Locke, Hume regards both worlds as composed of perceptions. The single most revealing feature of Humean impressions is that by this term he refers not only to our sensations, but also, as we have seen, to pains and pleasures. And this is not all: the term also applies to yet another variety of original perceptions, those ‘passions and emotions’ (T1) or ‘impressions of reflection’ (thus labelled for the element of reflexivity they contain, being our feelings about our former sense impressions and ideas) which are the subject of Book 2 of Hume’s Treatise and of his Dissertation on the Passions. This is how he explains what sense impressions, bodily pains and pleasures, and impressions of reflection have in common: ... every impressions, external and internal, passions, affections, sensations, pains and pleasures, are originally on the same footing; and [...] whatever other differences we may observe among them, they appear, all of them, in their true colours, as impressions or perceptions (T190). So my visual impressions of the black marks on the white page of Hume’s Treatise are on a par with, say, the pain I feel if the spine of the book strikes me on the head, and with my 9 This is in spite of some of the most obvious eighteenth-century uses of the terms: in Johnson’s Dictionary, for example, the first definition for ‘impression’ is ‘the act of pressing one body upon another’. See Space and the Self, p. 19 n.13. 5 26/1/15 feeling of confidence in reading the familiar first lines of the book, and so on, in that all these kinds of perceptions are immediate, and immediately transparent to my mind. In this my visual sensations of patches of colour and my feeling of the bulk of the book are crucially similar to the bodily pain it gives me if adequately applied to my head, and to my reflexive reactions to it and to the text it contains and the ‘discourse’ conveyed by it. In Hume’s own words: ‘Every thing that enters the mind, being in reality a perception, ‘tis impossible any thing shou’d to feeling appear different’ (T190). Of course, this complete transparency of impressions is more evident in the cases of the pains and pleasures and of the passions than in the case of sense impressions: for there can be no doubt that pleasures and pains, like the desire or disgust and hope or fear to which our impressions and ideas give rise, are just what they are and do not represent anything else. Pain, desire or loathing are certainly caused or sparked off, or at least accompanied, by sense impressions or ideas, but there is no sense in which they represent these impressions or ideas, or indeed anything else—nor is there, therefore, any sense in which they could be deceitful. But in fact it is sense impressions Hume is primarily interested in here, and the conclusion he is angling at has to do specifically with sense impressions: this immediacy is the reason why, he writes, our senses simply cannot be prone to the ‘kind of fallacy and illusion’ necessary for them to suggest that their own objects, that is, sense impressions, exist externally and independently of ourselves. Some readers have found that Hume’s grouping together of sense impressions, bodily pains and pleasures, and passions is odd.10 This may be so. But then, as we have seen, the oddity appears to be deliberate: the way Hume restored ‘ideas’ to its original, pre-Lockean meaning is by grouping together the non-representational mental contents—all sensations, including bodily pains and pleasures, and passions—under the label ‘impressions’, and all the representational ones under the label ‘ideas’.11 What this discussion of the immediacy of impressions suggests is that Hume’s sense impressions too are meant, like bodily pains and pleasures and like the impressions of reflexion, to be regarded as non-representational: rather than being representations of objects, they themselves are the objects of our common experience.12 10 See for example Weller. [Who else?] 11 Otherwise we should think that impressions of objects give rise to ideas of impressions of objects, and together they give rise to passions which are not impressions-of anything (see Weller)—messy, as well as at odds with the text discussed in this section. Indeed, the whole discussion of the immediacy of impressions becomes a puzzle if we think that the difference between impressions and ideas has to do, as suggested by Stroud and by Weller, with the presence of the perceived objects. See above, p.1. 12 Peter Lipton pointed out to me that this reading would provide a neat solution to the missing shade of blue puzzle: the idea of the missing shade of blue is still about the impression, even though the impression has not actually taken place. To put it slightly differently, we can have ideas of possible impressions; the reason why we have no idea of an empty space is not that we haven’t ever seen one, but that it is in principle impossible for us ever to have any impression corresponding to it. 6 26/1/15 Hume’s spot of ink: or, Humean reflections on Kant’s first analogy I have suggested that Hume’s emphasis on the originality, immediacy, and innateness of sense impressions may be regarded as showing that Humean sense impressions are nonrepresentational. Of course Hume is not saying that there isn't anything in the world but our minds—in fact, often he seems to assume just the opposite. Let us consider a concrete case. Take Hume’s apparently innocent reference to a visual object, a spot of ink: Put a spot of ink upon paper, fix your eye upon that spot, and retire to such a distance, that, at last you lose sight of it; 'tis plain, that the moment before it vanish'd the image or impression was perfectly indivisible. ‘Tis not for want of rays of light striking on our eyes, that the minute parts of distant bodies convey not any sensible impression; but because they are remov'd beyond that distance, at which their impressions were reduc'd to a minimum, and were incapable of any farther diminution. A microscope or telescope, which renders them visible, produces not any new rays of light, but only spreads those, which always flow'd from them; and by that means both gives parts to impressions, which to the naked eye appear simple and uncompounded, and advances to a minimum, what was formerly imperceptible. (T27-28). Here Hume is discussing the conceivability of indivisibles of space, and arguing that, far from being counterintuitive and paradoxical, minima are familiar objects of our perception. In particular, he is showing that we are actually acquainted with them through vision: the spot of ink, just before disappearing, is a unit of sight, that is, an indivisible of space immediately present to our eyes. There are several difficulties in this passage. I have treated extensively elsewhere the various and thorny problems specifically connected with the idea of space and with the nature, conceivability etc. of visual and spatial indivisibles.13 Here I would like to mention only three questions. First, what comes across most immediately is that Hume is using a (broadly defined) Lockean approach to descrive the process of sensation: the whole ‘minute parts’ story, with the addition of the natural philosophical talk of rays of light and optical instruments is in apparent tension with Hume’s open and repeated refusal I cited earlier to engage with either natural philosophy, or anatomical matters.14 I shall come back to this tension later, at the end of this section. A second equally obvious feature of this passage is that this supposed phenomenology of a visual experience seems to 13 See Space and the Self, ch.1, passim, for a full discussion of the difficulties connected with Hume’s notion of visual unit, and the relation between that notion and his talk about the nature and properties of ‘real’ space. 14 See Space and the Self, pp. 52-4, for detailed discussions of Hume’s natural philosophical talk and its sources. To put Hume’s discussion in historical perspective I found very useful G. Cantor, Optics after Newton. Theories of Light in Britain and Ireland 1704-1840, Manchester 1983, especially pp. 32 ff. 7 26/1/15 be based on the commonsensical assumption of the existence out there of such things as spots of ink, rays of light and optical instruments.15 But if impressions are all we have as the material of our knowledge of the world, how can we ever conceive and talk about anything like ‘a spot of ink’? Especially so, since the spot of ink cannot be easily identified with any one visual impression, being in fact constituted by a series of impressions. Now, as John Laird noted, ‘What is the “spot” if the “impressions” form a series?’16 And this is not all, for as we are going to see, Hume entitles us to wonder, even more radically: how can impressions ever ‘form series’ at all? I intend to focus specifically on this last question; but it may be useful to introduce it by elaborating on the more general question of Hume’s use of objective language. Hume’s terminological ambiguity is notorious. In connection with our present problem, in a stimulating recent study Marjorie Greene presents a systematic examination of Hume’s uses of the term ‘object’ in the Treatise. These can be classified, she suggests, under three main headings. The first is ‘intentional objects’ or ‘targets of attention’—that is, what ideas, passions and perhaps occasionally even sense impressions are about. The second use of ‘object’ is as synonym of ‘sense impression’. This is, according to Greene, the canonical sense of Hume’s ‘object’ and the very core of his empiricist project; and from what I have said so far it is clear that on the whole I agree with her on this point. The third use of ‘objects’ is in the ‘objective sense’, meaning external objects—and she notices that, rather surprisingly, this third category appears to be the best represented, with over 120 instances!17 Hume’s use of objective terminology in connection with the question of external objects is indeed a locus classicus of Hume scholarship.18 And Hume’s scanty, but unmistakable references to ‘unknown efficacies’, ‘secret force and energy of causes’, ‘conceal’d causes’, ‘secret operations’, ‘unknown and mysterious connexions’, and so on, has been for the last 15 years at the centre of a lively scholarly debate relative to his analysis of the idea of cause and effect. Let me briefly outline the main point of this debate, which is of some direct relevance to my present purposes because its focus is on a problem analogous to mine. 15 Ibid., pp. 46-55. 16 Laird, Hume’s Philosophy, p. 68. 17 M. Greene, ’The objects of Hume’s Treatise’, Hume Studies 20/2 (1994): 163-77. 18 The classic analytic treatment is H. H. Price, Hume's theory of the external world (Oxford: Oxford University Press, 1940); very interesting material in R. F. Anderson, Hume’s First Principles (Lincoln: University of Nebraska Press, 1966); excellent discussions in J. W. Yolton, Perceptual Acquaintance (Oxford: Blackwell, 1984), ch. 8: ‘Hume on single and double existence’, pp. 147-64, and A Baier, A Progress of Sentiments: Reflections on Hume’s ‘Treatise’ (Cambridge, MA: Harvard University Press, 1991), ch. 5, pp. 101-128. See also the recent debate between F. Wilson, ‘Is Hume a sceptic with regard to the senses?’, Journal of the History of Philosophy 27:1 (1989): 49-73, and D. Livingston, ‘A Sellarsian Hume?’, Journal of the History of Philosophy 29:2 (1991): 281-96. 8 26/1/15 Hume famously pointed out that our idea of a necessary connexion between cause and effect is not copied from a sense impression of power in the cause, or of interaction, or of necessary connexion between the cause and the effect—it is a fact, Hume maintains, that we have no such sense impression. Rather, the idea of necessary connexion is copied from the feeling of easy and inescapable transition from the sense impression of the ‘cause’ to a strong and lively idea of the ‘effect’, once we have acquired the habit of experiencing this particular kind of ‘cause’ as constantly conjoyned with this particular kind of ‘effect’. The alleged unknown power, or secret spring, or necessary connexion between the two are, however, totally unavailable to us. And yet, there are various references to them in Hume’s text. Don’t they seem to indicate that he did believe in the existence, out there, of necessary connexions, however inaccessible to us, between causes and effects? And exactly how inaccessible are they—can we at least meaningfully talk about them? Etc.19 As I said, the question of how impressions can ever ‘form series’ poses a problem which is at the same time very similar to this, and more basic—so, not with our notions of causal links among objects, but rather with the links in virtue of which we can think of the objects themselves as such. All our particular perceptions, Hume says, ‘are different, and distinguishable, and separable from each other, and may be separately consider’d, and may exist separately, and have no need of any thing to support their existence…’ (T252). In other words, we take it for granted that impressions are grouped together to make up objects, while each of them is, in fact, completely self-contained. So it is not just the connection, necessary or otherwise, between distinct and causally linked objects which is on the line of fire here, but also the connection between sense impressions which we should regard as independent existences, and of which, on the contrary, we think as different appearances of the same thing. If each of our sightings of the retiring spot of ink is a new and independent existence, how do we come to see them as forming the series which we further unify under the term ‘a spot of ink’? Hume’s answer to this question is to be found in the section of the Treatise devoted to ‘the scepticism with regard to the senses’. Here Hume considers the origin of our idea of both the continued, and the distinct existence of objects, establishing that these are separate, if strictly linked questions. What is the faculty responsible for our idea of the continued existence of objects? Certainly not our senses, for ‘they cannot operate beyond the extent, in which they really operate’ (T191); nor, as we have seen, can they mislead us in this, given the immediacy of their imput. Reason cannot be responsible either: for whatever argument 19 See K. Winkler, ‘The New Hume’, Philosophical Review 100 (1991): 541-579, and the more recent elegant paper by P. Kail, ‘New Humes and old: Sceptical realism and the AP property’ (to be presented at the 1999 Hume Society Conference), for excellent overviews of the discussion from opposite standpoints. 9 26/1/15 philosophers can come up with, since it has not been produced so far, it would have to be very far-fetched and sophisticated. But then, by definition any such argument would be an unlikely candidate for the reason why ‘children, peasants, and the greatest part of mankind’ think of continuous objects in spite of the discontinuities of sensation. The construction of continuous objects turns out to be the result of a synthesis operated by the imagination, which attributes continuity when it finds habitual constancy and coherence—the constancy of ‘my bed and table, my books and papers’, which ‘present themselves in the same uniform manner, and change not upon account of any interruptions of my seeing or perceiving them’ (T194-5), and the coherence of the fire in the fireplace: ‘when I return to my chamber after an hour’s absence, I find not my fire in the same situation, in which I left it: But then I am accustom’d in other instances to see a like alteration produc’d in a like time, whether I am present or absent, near or remote’ (T195). As Hume puts it, the imagination fills the gaps through a kind of mental inertia: ‘when set into any train of thinking, [it] is apt to continue, even when its object fails it, and like a galley put in motion by the oars, carries on its course without any new impulse’ (T198). The supposed continuity is the result of the imagination’s inertial movement started by the observed partial uniformity and coherence. Here there is another parallelism with the case of the idea of cause and effect: for the inertial movement of the mind which constructs a continuous object to underpin the constancy and/or coherence of fleeting sense impressions is of a sort similar to the expectation created by causation, that is, to the easy passage of the mind from the impression of the cause to the lively idea of the effect which usually accompanies it. Such sense impressions as the creaking of a door while I am not looking are unreflectively connected by the imagination to the other sense impressions usually associated with it, regardless of their actual presence in this particular occasion—in this case, say, the sight of the door opening. The result is that, in short, ‘I suppose that the door still remains, and that it was open’d without my perceiving it’ (T196-7). Without this, a creaking door heard but not seen would be in contradiction with my common experience, in which doors creaking and being seen moving in certain ways always go together. Indeed, Hume suggests that such a door could put under strain the whole system of maxims on which we base our causal reasonings, by confusing our habitual associations of constantly conjoined events. So, to summarise with the words of two recent readers of Hume, the whole point of Hume’s discussion here is that ‘the notion of physical objects and processes is built up from sense 10 26/1/15 impressions by means of the imagination’.20 While thus making our experience an experience of objects, our imagination also gives rise to the notion that the objects of our experience are distinct and independent existences—in the manner that we only know to be the case, at least in principle, for our individual perceptions. This is how our experience is an experience of objects, and of apparently external objects. Each of these objects consists of a number of self-contained sense impressions clustered together by our imagination on the basis of a pattern arising from habit. They are fictions of the imagination. This is not, of course, the end of the story; and the part played by philosophy in its next instalment is of some interest. The question of the existence of continuous and separately existent objects opens up a fatal dialectic between vulgar thought and philosophy: for philosophy informs us, that every thing, which appears to the mind, is nothing but a perception, and is interrupted, and dependent on the mind; whereas the vulgar confound perceptions and objects, and attribute a distinct continu’d existence to the very things they feel or see (T193). The vulgar naturally and unreflexively identify perceptions and objects: ‘this pen or paper, which is immediately perceiv’d’, is all there is. To describe their point of view, Hume suggests, the words ‘perception’ and ‘object’ may be used as synonyms, as both meaning ‘what any common man means by a hat, or shoe, or stone, or any other impression, convey’d to him by his senses’ (T202). Note that this is what we all do most of the time, regardless of how philosophically sophisticated we are when we are wearing our philosophical hat: even philosophers themselves are common people when the intellectual tension of their pursuits relaxes, making them snap back to nature so that they ‘take their perceptions to be their only objects, and suppose, that the very being, which is intimately present to the mind, is the real body or material existence’ (T206; see also T205, 216, 218). But how can our sensations continue to exist when we are not having them? This contradiction is exposed by a ‘philosophy’ which Hume seems to consider as the result of the most basic, and entirely natural use of reason. It is again a fact of human nature that our mind feels ill at ease with contradictions, so this one gives rise to a ‘system of the double existence’, something of a natural, and hence timeless, simplified version of Lockeanism: our perceptions are not all there is; underlying them there is a world of continuous and independent objects. While the objects of the vulgar are unreflectively identified with perceptions, within this philosophical system objects are 20 A. Hausman and D. Hausman, ‘Idealising Hume’, Hume Studies 18/2 (1992): 209-18, esp. p. 211. Also Hausman and Hausman, ‘Hume’s use of illicit substances’, Hume Studies 15/1 (1989): 1-38. AND CHECK the discussion in Stroud, Hume, pp. 96 ff. esp. p. 106. 11 26/1/15 split into perceptions, with which we are familiar and which are dependent on the mind and therefore interrupted and fleeting, and something else, which we can only suppose to be similar to those perceptions, except that we endow it with continuity and independence. In other words we try to pacify our mind by satisfying at the same time our reason—with the attribution of the interruptions to perceptions—and our imagination—with the attribution of the continuity to independent objects (T215). This is as much relief as we can get in this case: clearly not a solution of the contradiction, but an attempt to defuse the mind’s unease. In this sense, to a true philosopher both the vulgar system and the philosophical system of the double existence are unacceptable, the one for being based on an illusion and riddled with contradictions, and the other for being in the same position with the addition of a touch of absurdity.21 The conflict can only be reconciled at a metaphilosophical level, and the solution is the position of the ‘true philosopher’, that is, Hume’s own famous ‘moderate scepticism’. To the eyes of a ‘moderate sceptic’ the spot of ink is both the obvious object of common sense (this is when he is not philosophising, that is, most of the time), or a devilishly complicated set of interactions involving impressions, ideas, habits and operations of the imagination, even acts of judgment.22 And as far as the existence out there of such continuous and independent things as ‘spots of ink’ is concerned, ‘moderate scepticism’ means learning how to do philosophy ‘in a careless manner’—that is, how to live with the fact that our belief in it is at the same time natural, hence absolutely inevitable, and philosophically incoherent (T218, 273).23 Hume Prussian and non-Prussian Let us now go back to Humean objects and their origin. As I have said, these ‘objects’ are constructed by our imagination out of sense impressions which are, in principle, selfcontained existences—that is, before we start philosophising our imagination quietly and routinely synthesises the manifold of our scattered sense impressions, organising and coordinating them into things such as ‘spots of ink’, pens and sheets of paper, hats and shoes. These are the objects with which we are familiar when we are living our life without thinking about difficult philosophical issues; when we start wandering about them, we get disastrously entangled in a web of metaphysical inconsistencies and miseries. This situation can only be 21 Hume’s actual position in this section, in particular with respect to ‘the system of the double existence’, is controversial: see the works cited above, in fn. 18, esp. Yolton and the discussion between Wilson and Livingston. My presentation follows Yolton’s and Livingston’s lead, vs Wilson. CHECK!!! 22 See Space and the Self, pp. 47-50. 23 Connect with Winkler’s idea that to Hume causal powers are unthinkable? 12 26/1/15 sorted out by adopting a metaphilosophical stance, that is, by considering it as the inevitable result of the features and limitations of human nature and reason. This way of expressing the situation is deliberately suggestive of a very Prussian Hume. My suggestion is indeed that Humean sense impressions may be regarded, in some important respects, as similar to Kantian phenomena; and that while he is clearly very far from denying the existence of an external world, Hume does, like Kant, work at tracing the limits of our knowledge of it. In this sense I think that to Hume, as to Kant, epistemology is preliminary to metaphysics, because it inevitably determines the scope of metaphysics—accounting for human perception is the first step towards any discussion of what there is, for, as we have seen, before worrying about whether objects are causally connected to each other, or indeed whether they exist out there, we are to decide in what way we come to think of our experience as being experience of ‘objects’. There are also, however, big differences; and I would like briefly to mention one. In his classic ‘A Prussian Hume and a Scottish Kant’ Lewis White Beck noticed that in treating first causation, then objects, the order Hume is following is the reverse of what Kant was to do in the Critique of Pure Reason—as if Kant had put the second analogy of experience, on the irreversibility of time and the objectivity of causal succession, before the first, on duration and the objectivity of permanence.24 I can think of two reasons for this apparently perverse inversion. The first is Hume’s famous commitment to ‘carelessness and inattention’ as a form of wisdom. Unlike Kant, who, perhaps paradoxically, wanted metaphysics to get somewhere, if not to progress just as if it were a science, Hume, convinced as he was that vulgar thought, false metaphysics and true metaphysics are all equally insuppressible expressions of human nature, adopted a drastically non-cumulative approach: every new question is tackled against the background of the whole system of commonly held beliefs left unscathed by former discussions, and, in turn, leaves that system entirely unshaken. With this we are at the first and most immediate problem I signalled above in Hume’s passage on the spot of ink, his adoption of Lockean-cum-natural philosophical and medical terminology and approach which is so evidently at odds with much of his stated convictions. And this is why in spite of all that has happened in the course of Book 1 of the Treatise at the beginning of Book 2, devoted to the passions, he feels that, after repeating that sense impressions are the ones ‘which without any introduction make their appearance to the soul’, he can explain: 24 ‘A Prussian Hume and a Scottish Kant’, Essays on Kant and Hume, Yale University Press, 1989; repr. in B. Logan (ed.), Immanuel Kant’s Prolegomena in Focus, London: Routledge, 1996, pp. 139-55, esp. n. 39, pp. 154-5. 13 26/1/15 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 (T275). Here the ‘application of objects to the external organs’, again so reminiscent of Locke’s causal theory of perception, figures among the possible ‘unknown causes’ of our sensations, together with the ‘constitution of the body’ and the ‘animal spirits’ of contemporary medical and physiological accounts of perception; and they are all referred to in the typically flat, matter-of-fact commonsensical manner of a Hume who is focussing his sharp mind somewhere else, a Hume who is preparing to tackle a different question.25 My second possible explanation is, in fact, the same, but put in more complex and philosophically attractive terms. Here I can only offer it in the form of a brief suggestion. It has to do with a remark by Hume to the effect that even in the case of the identity of objects the synthesis can only be operated through the relation of cause and effect: we readily suppose an object may continue individually the same, tho’ several times absent from and present to the senses; and ascribe to it identity, notwithstanding the interruption of the perception, whenever we conclude, that if we had kept our eye or hand constantly upon it, it wou’d have convey’d an invariable and interrupted perception. But this conclusion beyond the impressions of the senses can be founded only on the connexion of cause and effect; nor can we otherwise have any security, that the object is not chang’d upon us, however much the new object may resemble that which was formerly present to the senses (T74). This reference to a wider role of causation as the connective tissue of our experience is tantalisingly short, and, I find, not entirely clear. It certainly seems to suggest that there is no way we can first consider and guarantee the existence of objects, then tackle the issue of how they may be connected to each other by causal links—or, as Hume himself puts it, duration is only conceivable on the basis of succession and coexistence: for when we think of an unchangeable object in time this is ‘only by a fiction of the imagination, by which the unchangeable object is suppos’d to participate of the changes of the co-existent objects’ (T200-201).26 The very way we organise our scattered sense data into those clusters that we 25 Greene, ‘Objects...’, pp. 171-2, produces a similar argument as her main explanation for Hume’s objective use of ‘objects’: ‘Once we have gone through our sceptical cleansing—she writes—being creatures of habit, we not only can, but must, indulge our propensity to feign. We do talk about objects other than our perceptions, about the “objects” we can’t help believing constant in the intervals of our perceptions. So Hume, like any other person, is bound to talk in the same way’. 26 See also T37, on the origin of the idea of time, and T65, on how we come to apply the notion of duration to unchangeable objects. 14 26/1/15 call ‘objects’ already shows the natural relation of cause-and-effect at work. And this means that there is no natural order for us to follow in our attempt to uncover the foundations of human knowledge: duration, causation and coexistence are so interdependent, that we can only tackle them, in whatever order we wish, one at a time and against the background of the other two firmly taken for granted in the commonsensical way. Coda on strength and liveliness I would now like, before concluding, to try to sketch an answer to the question: if the difference between impressions and ideas is that impressions are originals while ideas are copies, and impressions are non-representationals while ideas represent impressions, why does Hume give liveliness and strength such prominence when introducing the distinction, and why does he keep going back to it at crucial argumentative junctures?27 In a very interesting study, Stephen Everson suggests that the best way to make sense of Hume’s insistence on liveliness, force, vivacity, strength, etc. is, for a start, to note that this range of qualities, being immediately available to introspection, do not depend on unwarranted references to anything external to the mind.28 Secondly, according to Everson the difference between impressions and ideas, like the difference between an idea which is believed and one which is merely entertained, is functional: the lively, vivacious, strong, etc. perceptions, such as impressions and believed ideas, are similar in their ‘tendency to affect the mind and influence behaviour’—i.e., they are the ones that can serve, under appropriate circumstances, as stimuli to action.29 This is a very suggestive reading. Both its attraction, and its weakness lie in the fact that it takes the question of the difference between feeling and thinking resolutely out of the cognitive sphere, and into the practical. Is it possible to articulate a more Humean reading, while preserving something of Everson’s intuition? In order to try to do so, let me go back to the starting point of this paper. Hume’s distinction between impressions and ideas can be described in three different ways: impressions are lively and strong, while ideas are faint; impressions are innate and are originals, while ideas are copies of former impressions; impressions are non-representational, while ideas are representations of impressions. If we take it that the foundation of the distinction is 27 In fact, sometimes when the idea under discussion is of a particularly strong and lively variety he talks about it as ‘the idea or rather impression’—in particular this occurs with very strange and difficult ideas, such as that of self. A memorable example in connection with the present discussion is in T208-10; see Greene for other examples. 28 Everson, ‘The difference...’, p. 402. 29 Ibid., pp. 406-8. Everson cites from the Enquiry: ‘a man in a fit of anger is actuated in a very different manner from one who only thinks of that emotion’ (E11). 15 26/1/15 guaranteed by the original vs copy, and non-representational vs representational contrasts, then perhaps strength and liveliness should be regarded simply as the practical criterion for it. If so, the fact that this criterion is not watertight—that it is possible to construct counterexamples, and that it only really applies in the average cases, leaving a large area of untidy borderline cases—may create practical problems, but certainly does not in any way threaten the distinction. Of course, one could object that if there was no other independent criterion for the distiction we would be back at square one: what is the point of having a wellestablished distinction whose only practical criterion is hopelessly vague and only applicable to average cases?30 My answer to this question is a cautious no: we are not exactly back at square one. Strength and liveliness are not, in fact, the only practical criterion. To see this we need to consider more closely Hume’s uses of the strength and liveliness terminology. In particular, it is essential to remember that this terminology is not only used to illustrate the difference between impressions and ideas; the whole Treatise seems to be peppered with it. For example, it is referred to to mark the distinction between ideas of memory and ideas of the imagination: ‘the ideas of the memory are much more lively and strong than those of the imagination’, the former ‘flow[ing] in upon the mind in a forcible manner’, while the latter are ‘faint and languid’, etc. (T9). And it also makes its appearance, typically and crucially, in Hume’s attempted descriptions of belief, which is called ‘... a superior force, or vivacity, or solidity, or firmness, or steadiness’ (T629). This is not just a superficial similarity, but rather an explicit connection: the belief or assent, which always attends the memory and senses, is nothing but the vivacity of those perceptions they present [...] To believe is in this case to feel an immediate impression of the senses, or a repetition of that impression in the memory. ‘Tis merely the force and liveliness of the perception, which constitutes the first act of judgment, and lays the foundation of that reasoning, which we build upon it, when we trace the relation of cause and effect (T86). Now, that Hume himself was not happy with this vague terminology is a known fact: in the additions printed in the Appendix to the Treatise he goes back to it again and again, acknowledging that ‘this variety of terms’ may well seem very unphilosophical (T629), and even admitting that reference to ‘degrees of force and vivacity’ to describe all the possible differences between two ideas of the same object is an ‘error’ (T636). But in both cases this is only to suggest that perhaps talk in terms of ‘different feeling’ would be more appropriate. So I think it is evident that he is trying to pin down something having to do with the 30 My treatment of this point owes much to discussions with Peter Lipton and Nick Jardine. 16 26/1/15 emotional or affective raw material of knowledge; and that the connection he finds between sense impressions, memories, and beliefs is in terms of this raw material. In this sense, in my opinion Hume’s talk of the strength and liveliness of impressions and beliefs should be considered as a sketch of a taxonomy of cognitive states of mind. That belief needs such raw material is clear. Consider Hume’s discussion of belief in connection with causation. When our inferences from cause to effect carry us beyond what is actually present to either our senses or memory, our mind still refers to a direct perception, that is, to an actual or remembered sense impression: Tho’ the mind in its reasonings from causes or effects carries its view beyond those objects, which it sees or remembers, it must never lose sight of them entirely, nor reason merely upon its own ideas, without some mixture of impressions, or at least of ideas of the memory, which are equivalent to impressions (T82). Our belief in the link between cause and effect is based on the transference of belief from the present sense impression of the cause to the idea of the as yet absent effect—similarly, our belief in the what we read in a book is fed by the impressions of the black marks on the white page which lend ‘the authority either of the memory or senses’ to our whole reasoning: Here are certain characters and letters present either to our memory or senses; which characters we likewise remember to have been us’d as the signs of certain ideas; and these ideas were either in the minds of such as were immediately present at that action, and receiv’d the ideas directly from its existence; or they were deriv’d from the testimony of others, and that again from another testimony, by a visible gradation, ‘till we arrive at those who were eye-witnesses and spectators of the event (T83). Here we are again looking hard at the page of a book: and our belief in what we are reading—in the case Hume is considering it is Julius Caesar’s assassination—is based on the transference of affective vivacity from the present impression of ‘those character or letters which are seen or remember’d’, through a long ‘chain of argument or connexion of causes and effects’, to the idea of knives in the tyrant’s flesh. A full discussion of this passage would bring me much too far afield.31 Here I intend to mention only one very obvious problem—namely, that the sense impressions of the black characters should equally well lend their affective vivacity to our conception of the contents of whatever book we read, 31 On the various difficulties of this passage see E. G. Anscombe, “Hume and Julius Caesar”, From Parmenides to Wittgenstein: Collected Philosophical Papers, (Oxford: Basil Blackwell, 1981), vol. 1, pp. 86-92; D. Livingston, “Anscombe, Hume and Julius Caesar”, Analysis 35 (1974): 13-19; and M. Sprevak, “A reason for belief, or a reasonable way of believing: Hume, Anscombe, and Julius Caesar” (forthcoming). 17 26/1/15 regardless of whether it is Roman history, philosophy, or, say, science fiction. But then, the question is still: what is the difference between believing in what we are reading, and thinking that it is a fiction—that is, what is the difference between a believed and a merely entertained idea? And this is, after all, a variant of the more basic question with which we started: how can we tell the difference between an impression and an idea? Of course this problem did not escape Hume. His text shows clearly that he is aware of the difficulty, on the basis of his approach to belief, in accounting for the difference between, for example, reading boring history and reading very exciting fiction: ‘a poetical description may have a more sensible effect on the fancy, than an historical narration’ and ‘may seem to set the object before us in more lively colours’; and yet, ‘the ideas that it presents are different to the feeling from those, which arise from memory and the judgment’. Hume states clearly enough that belief is irreducible, for example when he says that ‘an opinion or belief is nothing but an idea, that is different from fiction, not in the nature, or the order of its parts, but in the manner of its being conceiv’d’, and goes on to admit that if he tries to explain this ‘manner’ he can ‘scarce find any word that fully answers the case’ (T628-9). Nevertheless, his language oscillates between this manner-talk, with lots of italicised feelings, lively, manner, etc., and something entirely different: the use of what Hume calls ‘reflexion and general rules’ (T631), that is, roughly speaking, the various criteria, including the rules of traditional Baconian inductive logic, by means of which we ‘vet’ the affective raw materials of belief and decide which of them are to be assented to, thus coming to constitute beliefs proper.32 This interaction between the affective raw materials of belief, and the combination of reflexion and general rules, is complex and most promising, and, even though Hume chose, for some reason, to leave it unclear and underdeveloped, it evidently complements the criterion of liveliness and strength. So ‘reflexion and general rules’ seem to provide the independent practical criterion for the distinction between impressions and ideas that we were looking for. Marina Frasca-Spada Department of History and Philosophy of Science Free School Lane, Cambridge CB2 3RH, U.K. <[email protected]> 32 On the ‘general rules’ in the Treatise see Laird, Hume’s Philosophy, pp. 91-2; J. Passmore, Hume's Intentions, 3rd edition (London: Duckworth, 1980), pp. 10, 63-4 and passim; L. W. Beck, “A Prussian Hume and a Scottish Kant”, pp. 146-7; T. K. Hearn, Jr., ‘“General rules” in Hume's Treatise’, Journal of the History of Philosophy 8, 1970, pp. 405-22; W. H. Walsh, ‘Hume's concept of truth’, in G. Vesey (ed.), Reason and Reality, Royal Institute of Philosophy Lectures, vol. 5, 1970-1971, London: Macmillan, 1972, pp. 99-116; and A. Baier, A Progress of Sentiments, pp. 56-9. 18