Humanity, sympathy and the puzzle of Hume's second enquiry

British Journal for the History of Philosophy 15(1) 2007: 27 – 57 ARTICLE HUMANITY, SYMPATHY AND THE PUZZLE OF HUME’S SECOND ENQUIRY Remy Debes Here is a well-known and curious puzzle of Hume’s second Enquiry: After going to great lengths to develop his associationist account of sympathy in Book II of the Treatise, and then using this account to argue for extensive sympathy as the source of our moral sentiments in Book III of the same work, Hume not only seems to replace extensive sympathy with the principle of humanity as the source of our moral sentiments in the second Enquiry, but he appears to drop the Treatise account of sympathy altogether.1 Almost of equal interest is the variety of proposed solutions to this puzzle. Some simply do not address it or have ignored it by virtue of ignoring the Enquiries altogether.2 Others, such as Selby-Bigge, have thought that Hume merely decided to abandon the associationist account of sympathy and consequently replaced it with a different principle.3 Similarly, John Rawls suggests Hume might have realized that he had made too much of the associationist mechanism and that such an account of sympathy yielded only ‘imparted feeling,’ a result that did not satisfy Hume’s philosophical needs, for it could not by itself explain the ‘peculiar’ moral sentiment that Hume thought underwrote the possibility of general moral agreement. Rawls adds that the account of sympathy in the Treatise depended on a ‘dubious’ idea of the self, which Hume himself later rejected, and which thus might also have led him to reject the Treatise account of sympathy altogether.4 1 This particular puzzle of the second Enquiry was discussed at least as early as 1893, when Selby-Bigge made note of it in his introduction to the Enquiries (Enquiries Concerning Human Understanding and Concerning the Principles of Morals, edited by P. H. Nidditch (Oxford, 1975) xxvi–xxix). Most recently, this puzzle is the focus of Kate Abramson’s ‘Sympathy and the Project of Hume’s Second Inquiry’ (Archiv für Geschichte der Philosophie, 83 (2001) No. 1: 45– 80), and Rico Vitz’s ‘Sympathy and Benevolence in Hume’s Moral Psychology’ (Journal of the History of Philosophy 42 (2004) No. 3: 261–75). 2 As to the former, see, for example, John Bricke, Mind and Morality: An Examination of Hume’s Moral Psychology (Oxford, 1996) and Barry Stroud, Hume (London, 1977). As to the latter, see Abramson’s discussion of the point, p. 45. 3 See n1. 4 John Rawls, Lectures on the History of Moral Philosophy (Cambridge, 2000) 101–2. Actually, at the point where he makes these remarks Rawls does not clearly explain why he thinks imparted feeling was philosophically insufficient for Hume. But given what Rawls says in British Journal for the History of Philosophy ISSN 0960-8788 print/ISSN 1469-3526 online ª 2007 BSHP http://www.tandf.co.uk/journals DOI: 10.1080/09608780601087954 28 REMY DEBES More moderately, Norman Kemp Smith argues that there was a general cooling of Hume’s interest in associationism, which explains the change of doctrine between the two works.5 And in contrast to all these, Kate Abramson has recently argued that Hume did not change his doctrine at all. Rather, the ‘principle of humanity’ is shorthand for ‘extensive sympathy’ and performs all the theoretical work in the second Enquiry that extensive sympathy did in the Treatise. On her account, Hume retained the associationist account of sympathy in the Enquiry, but left out the overt technical discussion of the Treatise in order to fulfil particular rhetorical goals that he thought this technical discussion might jeopardize.6 All of these critical commentaries, however, share one important feature: They are each asking, ‘What happened to sympathy?’ In contrast, I suggest that we would be well served by working in the opposite direction and asking, ‘What is humanity?’ Once we understand what humanity is for Hume, and the role it is supposed to play in our moral judgments, we can make better sense of the seeming change of positions from the Treatise to the second Enquiry. Indeed, investigating Hume’s notion of humanity will, I think, reveal the initial puzzle itself to be somewhat misconstrued. For, the assumption that humanity has replaced sympathy as the source of our moral sentiments results from a critical misunderstanding of the relationship between Hume’s notions of humanity and sympathy. Humanity is fundamentally dependent on sympathy, thereby implicating the latter at the deepest level of Hume’s sentimentalist ethics as presented in the second Enquiry. In this light, the puzzle of Hume’s second Enquiry is best recast in two critical pieces: First, what is humanity in the second Enquiry and what are we to make of its seeming replacement of extensive sympathy as the source of our moral sentiments? And second, what happened to the associationist account of sympathy emphasized so keenly in the Treatise? My primary task in this paper will be to answer the first of these two questions. To do this, I will conduct a careful analysis of Hume’s notion of humanity with heavy emphasis on the text of the second Enquiry (hereafter simply, the Enquiry). This is not to say I will skip over sympathy. On the Section 5 of the same chapter, this appears to be what he has in mind (see especially, 93–4). The dubious idea of the self that Rawls refers to in the connected point is, of course, the so-called bundle theory of identity. 5 Norman Kemp Smith, The Philosophy of David Hume (London, 1966) 530–6. Nicholas Capaldi might also be included here (David Hume: The Newtonian Philosopher (Boston, 1975) 180–7, especially 185–6). But I have two reasons for not mentioning Capaldi directly: (1) Including Capaldi’s interpretation would muddle my presentation, because his approach shares some similarities with those of Selby-Bigge and Rawls; (2) While Capaldi does have one unique solution to the puzzle of sympathy in the second Enquiry, like Terence Penelhum I worry that Capaldi’s interpretation of the text is confused; See Terence Penelhum, David Hume: An Introduction to His Philosophical System (West Lafayette, 1992) 155–6. For more response to Capaldi, see Vitz, 268–70. 6 Abramson, 45–80. Along with Abramson we can also place Rico Vitz, who essentially works towards the same ends (See especially, 274–5). HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 29 contrary, if I am right that humanity is dependent on sympathy (or at a minimum, the process of sympathizing), then we must also take a close look at the Enquiry account of sympathy. I will not in this paper, however, directly discuss the mysterious disappearance of the associationist account of sympathy from the Enquiry. That part of the puzzle I address fully in a second, related article.7 What we will find in that related analysis, however, is that the associationist explanation of sympathy never really disappeared from the Enquiry to begin with. Taken together, then, the arguments of these two papers will go some distance towards dissolving the initial puzzle altogether, and confirming that there is no real inconsistency between Hume’s position in the Enquiry and the Treatise regarding humanity, sympathy, and the source of our moral sentiments.8 I. THE BASIC ACCOUNT OF HUMANITY IN THE ENQUIRY Hume can be frustrating to read, for he often does not take the time to explain his own terminology, or he mixes several different descriptions of the same phenomenon into a single argument. Nevertheless, this much can be confidently demonstrated in the Enquiry: Humanity just is benevolence. This is an oversimplification of course. Indeed, as we shall see humanity (as a principle of human nature) is more like a disposition towards benevolence, at least insofar as we conceptualize benevolence as the desire for another’s good. But given Hume’s tendency towards terminological imprecision – in particular, his tendency to use both ‘humanity’ and ‘benevolence’ to describe a range of concepts from character traits and motives to principles of human nature – it is important to recognize his willingness to equate these two words. This equivalence is apparent even from the earliest discussion of benevolence in Section 2 of the Enquiry, 7 Remy Debes, ‘Has Anything Changed? Hume’s Theory of Association and Sympathy after the Treatise,’ forthcoming in the British Journal for the History of Philosophy (hereafter simply, ‘Has Anything Changed?’). 8 This conclusion thus partially aligns me with Abramson and Vitz. However, although Abramson, Vitz, and I agree on the essential point of consistency between Hume’s Treatise and second Enquiry, I have some important philosophical differences from both. In particular, as will be seen, I do not think we can understand the principle of humanity as mere shorthand for extensive sympathy, as Abramson claims. Such a reading would oversimplify both Hume’s account of humanity and its connections to sympathy. And while my arguments overlap more closely with Vitz’s, we differ considerably and importantly in method and emphasis. Although Vitz also wants to bring some attention to Hume’s concept of humanity, humanity remains a secondary focus for him. Moreover, as regards the second piece of the puzzle, my discussion of associationism and its connection to Hume’s notion of sympathy in ‘Has Anything Changed?’, is more elaborate and explicit than anything in Vitz. 30 REMY DEBES where the two words are used in conjunction to describe the same quality of character (see, e.g. EPM 2.1; SBN 176 and 2.6; 178).9,10 And that Hume does not intend a general distinction between humanity and benevolence is made clear by the implied synonymy between the two words throughout the Enquiry, shown especially by their obvious interchangeability.11 Most important for our purposes, however, and what cements our certainty of this equivalence, is that both benevolence and humanity are described as the same innate, original motive of good will towards our fellow human beings which grounds the moral sense. That is, both are described as innate moral motives or desires for the good of others (loosely, moral motivation), that in turn serve as the basis of moral approbation and blame (loosely, moral judgment).12 Two passages from the Enquiry illustrate this point particularly well and are worth quoting at length. Here is the first: It is sufficient for our present purpose, if it be allowed, what surely, without the greatest absurdity cannot be disputed, that there is some benevolence, however small, infused into our bosom; some spark of friendship for human kind; some particle of the dove kneaded into our frame, along with the elements of the wolf and serpent. Let these generous sentiments be supposed ever so weak; let 9 All citations to Hume’s Enquiries are to the Clarendon Critical Editions of An Enquiry Concerning Human Understanding and An Enquiry Concerning the Principles of Morals, edited by Tom Beauchamp (Oxford Clarendon Press, 2000 and 1998, respectively) with ‘EHU’ and ‘EPM’ respectively, followed by section and paragraph numbers (and sometimes line numbers as ‘ln’), and to the Selby-Bigge/Nidditch edition (Oxford Clarendon Press, 1975) by ‘SBN’, followed by page numbers. (Please note: I occasionally make reference to the Oxford Philosophical Texts Student Editions of these same works in contrast to the Scholarly Critical Editions, cited here. Bibliographic information for the former appears at the first citation.) All citations to Hume’s Treatise are to A Treatise of Human Nature, eds. David and Mary Norton (Oxford University Press, 2000) with ’T’, followed by book, part, section and paragraph numbers, and to the Selby-Bigge/Nidditch edition (Oxford Clarendon Press, 1978) by ‘SBN’, followed by page numbers. Multiple references to the same text are separated by a semicolon with ‘&’. 10 The location of the discussion on benevolence where these examples occur, so close to the start of the book, and its bold section title, are also suggestive. Given that the Enquiry leans so heavily on the principle of humanity, it makes sense, if humanity is benevolence, for Hume to introduce the discussion of benevolence early on. 11 It has been pointed out to me that, if humanity and benevolence are equivalent, using the terms in conjunction suggests a redundancy. This may be right, but it seems equally plausible that, if the equivalence is as important as I will argue it to be, using the terms in conjunction is a reasonable rhetorical method of showing this equivalence, in lieu of an explicit definition. Regardless, that Hume is making this equivalence seems amply borne out by the text: See in particular, EPM 5.18; SBN 220; also 7.19; 257; & 9.2; 269 (here ‘humanity’ is used to describe the virtue of benevolence); & 9.20; 281; & 9.21 (ln. 17–21); 282; & Appx. 1.3; 286. 12 Hume’s distinction between motivation and judgment and exactly what Hume means by ‘judgment’ are both notoriously tricky questions that I will only partly address in the course of this paper. I am grateful to an anonymous reviewer, however, for urging a more careful analysis and presentation of this distinction than I had originally intended. Of course, whatever faults remain are my own. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 31 them be insufficient to move even a hand or finger of our body, they must still direct the determinations of our mind, and where everything else is equal, produce a cool preference of what is useful and serviceable to mankind, above what is pernicious and dangerous. A moral distinction, therefore, immediately arises; a general sentiment of blame and approbation; a tendency, however faint, to the objects of the one, and a proportionable aversion to those of the other. (EPM 9.4; SBN 271, first emphasis added) The last sentence is key. It suggests that we are to understand Hume’s crediting of a ‘cool preference of what is useful and serviceable to mankind’ to innate benevolence (or a disposition to feel benevolent sentiments), as the basis of his claim that benevolence is the source of our moral distinctions, and blame and approbation in particular. This claim has two parts. First, it is clear that benevolence is essentially motivational: it is a ‘generous sentiment’ that, though it may not ‘move even a hand or finger’, does cause a ‘cool preference’ or ‘tendency’ towards what is useful and an ‘aversion’ towards what is not. Second, this benevolent sentiment raises approbation. That is, according to Hume, it is because of an innate benevolence (or benevolent sentiment) that what is serviceable to others produces or is attended by an approving or disapproving sentiment that grounds a ‘moral distinction.’ Thus, benevolence is here credited as grounding moral judgment. In light of this last Enquiry passage, then, it is especially remarkable that, in a passage almost immediately following it, Hume goes on to rework the same claim about the basis of moral judgment, now substituting humanity for benevolence: The notion of morals implies some sentiment common to all mankind, which recommends the same object to general approbation, and makes every man, or most men, agree in the same opinion or decision concerning it. It also implies some sentiment, so universal and comprehensive as to extend to all mankind, and render the actions and conduct, even of the persons the most remote, an object of applause or censure, according as they agree or disagree with that rule of right which is established. These two requisite circumstances belong alone to the sentiment of humanity here insisted on. (EPM 9.5; SBN 272)13 Hume is admittedly up to more than just a little rewording in this passage, but it is enough for our purposes to note that Hume is explicitly attributing 13 There are many more passages which credit humanity, as well as humanity and sympathy, as the source of our moral sentiments. I have only tried to point out passages which help illustrate humanity’s equivalence to benevolence. For a complete list of the humanity and sympathy passages in the Enquiry, see Abramson, 50, footnotes 16, 17, 18. 32 REMY DEBES the basis of moral judgment to a universal sentiment of humanity. It is by this sentiment that objects are ‘recommended’ for approbation and we are able to form similar ‘opinions’ or reach similar ‘decisions’ concerning these objects. Motivation is obviously not in the fore in this particular passage, but when taken together, these two passages make clear that the benevolent sentiment, which is ‘infused into our bosom,’ is identified with the sentiment of humanity, which is ‘common to all mankind,’ and that it is this (one and the same) sentiment that is the basis of moral judgment.14 This is, however, a crude description of Hume’s sense of humanity and so we will need to refine it. A few of these refinements can be made quickly, while others will have to be developed more slowly in the next section. First, in this paper I am not concerned with the virtue of humanity. Although Hume is keen to highlight humanity as a virtue, and when doing so maintains its equivalence with benevolence, it is the innate principle of humanity and the sentiments it produces that play the fundamental role in determining our moral judgments and with which I am ultimately concerned. Second, we must be careful to distinguish the sentiment of humanity (the sentiment of benevolence) from the principle of humanity (the principle of benevolence). The sentiment of humanity is essentially motivational. It is what Hume alternatively describes as a tender and agreeable feeling of kindness, generosity, or good-will towards others.15 It is the ‘cool preference of what is useful and serviceable to mankind’ (EPM 9.4; SBN 271), or ‘the benevolent concern for others’ that all human beings naturally feel for all other human beings (EPM 9.9; SBN 275).16 And it is to this humane concern or sentiment that Hume attributes the basis of moral judgment. This sentiment is the foundation of the ‘moral distinctions’ that ground our moral judgments (or our ‘opinions’ and ‘decisions’) of praise and blame. The principle of humanity, on the other hand, is the original or innate disposition of human nature towards benevolence (as the concern for the good of others), from which the sentiment of humanity springs. In light of the foregoing, in this sense humanity is an original disposition to be concerned for others that via the benevolent sentiments it raises 14 If any doubt remains about this point, note the two passages following the one just quoted in which Hume refers to the principle of humanity as ‘the universal principle of the human frame’ (EPM 9.7; SBN 272), and his assertion that ‘the humanity of one man is the humanity of every man’ (9.7; 273). Finally, see also EPM 3.6; SBN 185 (where Hume refers to our ‘innate humanity’); & 5.45; 231 (where the ‘benevolent principles’ are equated to ‘principles of humanity’); & 5.46; 231–2 (where the principle of humanity is equated to ‘the benevolent principle’ which is ‘implanted in our nature’ as the source of ‘moral blame and approbation’); & 9.8; 273–4 (where conduct with a beneficial tendency is agreeable to our humanity, and this is the basis of our moral distinctions). 15 For example, EPM 9.21; SBN 282. 16 See also, EPM 5.46; SBN 231 (‘humanity or a concern for others’). HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 33 leads us to approve, all else being equal, whatever promotes their happiness.17 As such, the principle of humanity is the source of both moral motivation and moral judgment. Thus Hume remarks in the heart of the Enquiry, ‘[I]f the principles of humanity are capable, in many instances, of influencing our actions, they must, at all times, have some authority over our sentiments, and give us a general approbation of what is useful to society, and blame of what is dangerous or pernicious’ (EPM 5.39; SBN 226).18 I will interchangeably refer to this as the principle of humanity or disposition towards humanity.19 Of course, the foregoing comments somewhat gloss over the connection between the (motivating) sentiments of benevolence and the moral sentiments. Why, we might wonder, do the benevolent sentiments produce or lead us to feel approbation or disapprobation? The passages considered so far suggest that the moral sentiments simply follow or attend benevolence (or its negative counterpart) naturally. Hume reinforces this impression in the Enquiry by alternatively arguing that benevolence (as a sentiment) ‘attaches’ approbation to its objects, that approbation is ‘derived from’ or ‘arises from’ benevolence, or that benevolence ‘gives us’ or is ‘attended with’ approbation.20 Indeed, from the start Hume explicitly expresses his opinion that the tight connection between the benevolent and moral sentiments is so obvious that it is arguably a ‘superfluous task’ to prove it (EPM 2.1; SBN 176). Still, this explanation of their connection might seem facile. We know 17 Thus Hume writes: If we consider the principles of the human make, such as they appear to daily experience and observation, we must, a priori, conclude it impossible for such a creature as man to be totally indifferent to the well or ill-being of his fellow-creatures, and not readily, of himself, to pronounce, where nothing gives him any particular bias, that what promotes their happiness is good, what tends to their misery is evil, without any farther regard or consideration. (EPM 5.43; SBN 230). 18 All emphases in this last quote are mine. Hume’s own emphasis is on ‘some,’ but I have altered this to bring out the important dual connection of humanity to motivation and judgment. See also 5.45; SBN 231, where useful tendencies not only ‘engage’ us on the side of virtue by ‘affecting the benevolent principles of our frame,’ but also these ‘principles of humanity’ ‘enter so deeply’ into our sentiments that they ‘excite the strongest censure and applause.’ 19 Hume has other ways of referring to this principle, most notably ‘instinct’ (especially in the Treatise). For the sake of economy, however, I’ll avoid these other terms. 20 Hume remarks in Section 5 of the Enquiry: ‘[I]t is always found, that a warm concern for the interests of our species is attended with a delicate feeling of all moral distinctions; a strong resentment of injury done to men; a lively approbation of their welfare’ (EPM 5.39; SBN 225, emphasis added). See also, EPM 5.42; SBN 186–7 (the benevolent sentiments make a moral ‘distinction’ for they ‘attach the notion good to a beneficent conduct’); & 5.44; 230–1 (the merit of social virtue is ‘derived from the feelings of humanity’); & 5.43; 230 (approbation ‘arises from’ the ‘natural sentiment of benevolence); & 5.39; SBN 226 (humanity ‘gives us’ approbation). 34 REMY DEBES that for Hume approbation and disapprobation are based on pleasure and pain. This is forcefully argued for in the Treatise, but is also carried through to the Enquiry, even if more quietly.21 For example, in Section 5 of the Enquiry, Hume argues that utility is a source of moral approbation, which results from the pleasurable sentiments we feel upon the sympathetic contemplation of the useful tendency of the traits or actions of others.22 So if we are to be fully satisfied with Hume’s claim that moral judgment is based on the benevolent sentiments, then we need an explanation connecting the benevolent sentiments to pleasure. The passages considered so far, however, are obscure in this regard, and Hume is arguably elusive about this connection throughout his work. But ultimately, I think the answer is simple: the sentiment of humanity is itself pleasurable. Sometimes this is subtle: at several important points in the Enquiry Hume credits humanity as the source of our pleasurable reaction to virtue.23 However, Hume also explicitly describes the benevolent sentiment as ‘warm’ (EPM 5.39; SBN 225), or as the ‘pleasing sentiment of humanity’ (EPM 6.3; SBN 234), or as sentiments ‘delightful in themselves’ (EPM 7.19; SBN 257).24 This is nowhere more clear than in the conclusion to the Enquiry, where Hume reminds us that while every passion, when gratified, conveys pleasure, benevolence is already pleasurable: ‘the immediate feeling of benevolence and friendship, humanity and kindness, is sweet, smooth, tender, and agreeable, independent of all fortune and accidents’ (EPM 9.21; SBN 282). We should thus read Hume as identifying the pleasurable sentiments that produce (or are attended by) approbation as species of 21 In the Treatise see especially T. 3.1.2.3; SBN 471, but the point is echoed in many places; see, for example, T.3.1.2.11; SBN 510 & 3.3.5.1; SBN 614 (hereafter I speak only of pleasure and approbation; the inverse is entailed at every point). 22 As early as Section 2 Part 2, Hume introduces the argument for utility as the source of moral approbation. The idea is then well elaborated in Sections 3 and 5, but of course reappears regularly throughout EPM. As for the connection between approbation and pleasure in the Enquiry, see especially EPM 5.4; SBN 214–15 (utility must please or be agreeable to be approved); & 5.4; 217 (virtue pleases and we give it ‘the just tribute of praise and approbation’); & 5.15; 218 (the agreeableness of usefulness engages approbation). See also EPM 7.2; 250–1 (‘affection and approbation’ attend cheerful dispositions because they ‘please’ and are ‘agreeable’); & 8.3; 262 (wit is approved of because it causes immediate enjoyment). Finally, as a point of interest, the connection between approbation and pleasure is also stated baldly in A Dissertation on the Passions (edited by T. H. Green and T. H. Grosse (London, 1875) Vol. II, p. 146–7), which from 1758–77 was seamlessly bonded alongside EPM in Essays and Treatises on Several Subjects. 23 See, for example, the conclusion of Section 5, the import of which I highlight later in this paper (EPM 5.46; SBN 231–232); & the conclusion of Appendix 1, which echoes a major argument from the Treatise that sentiment and not reason is the basis of moral judgment (EPM Appx. 1.20; 293–4). 24 This last description comes in the context of arguing that the benevolent sentiment is itself the object of approbation because it is immediately agreeable apart from a consideration of its utility. The possible wrinkle this presents to my argument I consider in n47. The important point here is that benevolence has this effect because it is a pleasurable sentiment. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 35 benevolence. This is why Hume describes the sentiments of benevolence or humanity as the basis of our moral distinctions and judgment. At least, this reading is borne out by the evidence so far.25 But, strictly speaking, my own argument for reconciling the Enquiry with the Treatise only requires that we recognize Hume’s willingness to credit the sentiment of humanity as the basis of approbation or disapprobation; it does not require an explanation of this connection per se. What remaining evidence exists for this reading, then, I will present only in passing. This said, on Hume’s account the principle of humanity only explains why the happiness of others concerns or interests us, and why we consequently prefer what conduces to their happiness, assuming those interests or that happiness is the object of our consideration. Humanity is effectively impotent without a means for this original principle of human nature to be engaged – or as Hume sometimes styles it, ‘affected’ or ‘touched’.26 That is, our disposition towards humanity is inert unless the interests or happiness of others can somehow be made to activate it. We must, then, have some further means by which we can represent the interests and feelings of others to ourselves so as to raise our natural sentiment of humanity and provoke our moral sentiments. ‘No passion,’ Hume writes, ‘when well represented, can be entirely indifferent to us’ (EPM 5.30; SBN 222). Here enters sympathy; at least, so I shall argue. Sympathy provides the means by which we represent the passions and interests of others, and by which our innate humanity is affected. The sentiment of humanity, then, which actually engages us in the interests of mankind and society, must be understood as arising from the original disposition of humanity, via sympathy. It is only when the happiness of others is represented to us by sympathy, that this natural disposition towards benevolence raises a desire for their happiness and approval for what promotes it.27 It is, in other words, a desire activated through sympathy. All this, however, still requires some proof. 25 I think this reading leaves open the possibility that approbation and the benevolent sentiment are in fact identical. I won’t try to decide this issue here, and in the end I don’t think this is the correct way to read Hume. It is true that according to Hume desires, tendencies, and aversions are themselves passions and thus akin to moral sentiments. But generally speaking, Hume doesn’t seem to think we have any trouble distinguishing between these impressions, as evidenced by his own sustained terminological distinction between motives, desires, and preferences, on the one hand, and approval and disapproval on the other (not to mention his famous discussion of the ‘peculiarity’ of moral approbation in the Treatise (T. 3.1.2; SBN 470– 6)). Thus, we should first consider explanations which respect Hume’s prima facie distinction between motivational passions and moral sentiments. 26 For example, EPM 5.44; SBN 230–1 (the merit of virtue is derived from the feelings of humanity when it affects the spectators); & 9.8; 274 (conduct gains my approbation by touching my humanity); & 9.6; 272–3. 27 See also, EPM 5.40; SBN 227 (from our ‘natural philanthropy’ we have a desire for the happiness of society); & 6.3 (starting at ln. 27); 234; & 6.4 (ln. 7–10); 235 and 6.5; 235, taken in contrast to each other; & Appx. 2.13; 302.3–9. 36 REMY DEBES II. TWO FEATURES OF HUMANITY Thus far I have made little mention of the Treatise, and have not explicitly attempted to show how understanding humanity in the Enquiry helps reconcile the two works. However, making clear the distinction between the ‘principle’ and the ‘sentiment’ of humanity, as we have just done, provides the starting point for analyzing two further features of Hume’s account of humanity – features that will aid us in demonstrating that no inconsistency between the two works exists, while simultaneously helping to prove that our humane disposition is indeed activated through sympathy. II.1 The first feature of humanity: extensive versus narrow humanity Although it might be a somewhat obvious inference in light of the Treatise, the following feature of Hume’s account of humanity needs to be distinguished in the Enquiry: humanity has both a narrow and expanded, or extensive form. Proof of this comes both directly from Hume’s remarks about the scope of humanity and indirectly from the account of sympathy in the Enquiry. I will consider the two lines of evidence in turn. To start, it will help to explain what Hume means by ‘extensive’ (and by contrast, ‘narrow’), and then to identify a few manifestations of such extension in the broader moral context. This will provide some initial evidence for my argument. Extension is sometimes discussed in proximity terms, or how close or closely related to me and my interests the object of my consideration is. It may also be in terms of temporal distance, or how distant in the past (or future) the object of consideration is. Let us call these the dimensions of scope.28 Extended humanity, then, has a wide scope: it is a concern for others even when they are far away or unrelated to me (and my interests), or when they are in the distant past. Hume discusses scope frequently in the Enquiry, but usually refers to these two dimensions in conjunction: for example, ‘[I]f you represent a tyrannical, insolent, or barbarous behaviour, in any country or in any age of the world, I soon carry my eye to the pernicious tendency of such a conduct, and feel the sentiment of repugnance and displeasure towards it. No character can be so remote as to be, in this light, wholly indifferent to me. What is beneficial to society or to the person himself must still be preferred’ (EPM 9.7; SBN 273, emphases added).29 Strictly speaking, humanity is here affected by the consequences of 28 Resemblance might also be distinguished as a dimension of scope, but I pass over it as it doesn’t seem to play a large role in the account of extensive humanity in the Enquiry. 29 Some of Hume’s usage pertains to the moral sentiments that attend the benevolent sentiments. A fortiori such examples thus prove extensive humanity; but I discuss these later (see n43). For examples that seem to have extensive benevolence per se in mind (albeit sometimes mixed with moral sentiment), see EPM 5.41–3; 227–30; & 9.5, 7–8; 272, 273–4 (I discuss the significance of 9.8 in section II.2.2 of this paper). HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 37 the tyrant’s actions, and is thus an assessment based on utility, a matter I discuss shortly (section II.2.2). The present point, however, is only to demonstrate that humanity has the capacity for extensive scope, which is made plain by this passage. Also, faintly in the background of this passage is one of the moral manifestations of this expanded scope, and perhaps its primary one. Extended scope remedies the prejudices of self-interest. In the case of humanity, it allows our humane concern to prevail over selfinterested bias, caused both by our own immediate interests and the interests of our private community, so that this concern can ‘comprehend all human creatures’ (EPM 9.7; SBN 273).30 That is, extending the scope of humanity makes the interests (and the joys and miseries) of others interest me, and even prevail over my own, despite considerations of self-interest. As Hume himself sometimes dubs it, let us call this public humanity.31 Part and parcel of this is a second manifestation of extensiveness: we can have extensive sentiments. That is, we do not just expand the scope of our disposition but also the consequent concern or desire for the welfare of others that this disposition raises. This is partly evidenced by the contrast Hume draws between public or ‘social’ humane passions and more ‘narrow’ and ‘private’ ones. Usually, this contrast is to self-interested or selfish passion (i.e. passions that result from the satisfaction or frustration of selfinterest).32 But there is also evidence that Hume thinks the sentiment of humanity per se can have a narrow or private manifestation as a result of a narrow (or narrower) scope. This seems to be true in the case of the benevolent man who by accident has the scope of his considerations constrained to his ‘private life’ and a ‘narrower’ sphere of activity, but who if exalted into a ‘higher station’ we can expect to have a corresponding extension of humane sentiments (EPM 2.7; SBN 178). Also relevant is Hume’s explanation of party zeal or sedition. For, on the one hand, Hume explicitly credits such zeal and sedition to the interpersonal concern and feelings of loyalty generated by ‘social sympathy’ (which again, I am claiming humanity depends on), in contrast to private interest; but on the other had zeal and sedition suggest an expanded scope but not a fully extensive humanity, for the interpersonal concern and loyalty is still limited by self-interest to a particular circle of people – that is why it is party zeal.33 We should thus understand narrow or private humanity as that concern for others that is to some degree limited by self-interest. That is, it is partial and biased to some degree, directing itself at those in whom I have, or with whom I share, some private self-interest. This may be benign as in the accidentally limited scope of the benevolent man, or more dangerous as in 30 See also EPM 5.8–12; SBN 216–17; & 5.17; 219; & 5.42; 228–9. See especially EPM 5.17; SBN 219. 32 E.g. EPM 5.35; SBN 224; & 9.9; 275–6 (‘selfish and private emotions’); & 9.20; 281 (‘social’ sentiments contrasted to ‘selfish’); & Appx. 2.7; 299 (generous sentiments contrasted to ‘narrow and interested’ considerations). 33 The principle passage is at EPM 5.35; SBN 224. 31 38 REMY DEBES sedition. At its worse, though it is not Hume’s focus, we can imagine that narrow humanity involves the utter subjugation of one’s scope to the limits of self-interest. In any event, we can see that this conception of narrow humanity is entailed by Hume’s understanding of the dimensions (scope) and manifestations (public disposition and sentiments) of extensive humanity. And although much more could be said in regard to the dimensions and manifestation of extension, this sketch will suffice for my purposes.34 With this description in hand, now note that in the Enquiry, Hume obviously thinks it a truism that we have an extensive sense of humanity.35 This was already implied in the passage about the tyrant, but also many passages in the Enquiry regarding benevolence and humanity speak only of this more expanded concern with society’s interests, seemingly shortcutting the intermediate step of sympathy that I have suggested is required. This is perhaps one source of confusion underlying the basic puzzle of the Enquiry. It is precisely because Hume often speaks as if we have an original and extensive concern for society’s happiness, without the need for sympathetic intervention, which likely has led many readers to think that there is a fundamental shift in doctrines from the Treatise to the Enquiry.36 This is especially likely because Hume seems to deny the possibility of an original extensive concern for others in the Treatise, when he says that we have no concern for others ‘merely as such’ (T. 3.2.1.12; SBN 481). In fact, I think this is only an apparent contradiction between the texts. But let us table this particular discrepancy for the moment; I will return to it when we have the necessary resources to settle it (in II.2.2–4). The immediate point is simply that the Enquiry clearly assumes this extensive sense of humanity. But besides noting Hume’s empirical conviction that we simply do have (at least the capacity for) extensive humanity, Hume’s discussion of justice in the Enquiry provides a further explicit argument for extensive humanity and its narrow counterpart. In his description of the fictional land of abundance (which notably parallels the Treatise), Hume argues that it is the possibility of enlarged generosity and ‘extensive benevolence’ which would make justice unnecessary in such a place.37 More important, that Hume 34 I have here somewhat glossed over the merit (or demerit) of extensive and narrow humanity. But as my focus is not the virtue of humanity I will continue to pass over this aspect in an effort to limit length. 35 For example, ‘It appears also, that, in our general approbation of character and manners, the useful tendency of the social virtues moves us not by any regards to self-interest, but has an influence much more universal and extensive’ (EPM 5.45; SBN 231). See also EPM 5.46; SBN 231 (we see in ‘numberless instances’ that whatever has a tendency to promote ‘the interests of society’ we approve). 36 For example, Selby-Bigge suggests as much in his introduction to the Enquiries (p. xxv). A good example of the kind of remark that might mislead readers is Hume’s comment in Section 5 that people act from a ‘natural philanthropy’ (EPM 5.40; SBN 227). 37 The land of abundance example opens Section 3 on Justice. The comments on enlarged generosity and extensive benevolence occur at EPM 3.6; SBN 184–5. Hume also remarks that HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 39 even discusses the fictional land of abundance in connection with justice is instructive. For, if it were not the case that humanity is naturally narrow but has the capacity to be extensive, the discussion would be pointless. Were humanity innately extensive, then justice would never be needed in the first place. We would already live in a land of abundance. This is the whole point of the example. So already this begins to counter the misperception that the Enquiry posits an original, extensive humanity in human nature, as opposed to a narrow humanity that becomes extended. On the other hand, were humanity unalterably narrowly fixed, we could never be affected by a system of justice.38 That is, if our sense of humanity was narrowly fixed, justice could not engage our sentiments. For as Hume states, ‘The good of mankind is the only object of all these laws and regulations’ (EPM 3.22; SBN 192).39 But the good of mankind is clearly an extensive concern that in turn requires an expanded public scope, not a narrow one. So again, Hume must think that humanity can take an extensive form. Beyond these direct textual appeals, there is also indirect discussion of extensive humanity in the Enquiry. To make the indirect case, however, I need to say a few more words about sympathy in the Enquiry. First, as Abramson has helped make clear, the account of sympathy in the Enquiry has retained the fundamental deliberative aspects of the Treatise account.40 In both works, sympathetic deliberation requires both that we consider the tendency of traits to benefit the possessor or others, and that we consider only the typical effects of that trait. More relevant to our purposes, Abramson demonstrates that in both works Hume stresses the necessity of restricting ourselves to a general point of view in sympathetic deliberation, in order to correct the natural biases of our private (subjective) situation. But this is, in effect, to expand the scope of sympathy – to make sympathy public. And although in the Enquiry Hume never calls this the viewpoint of the ‘judicious spectator,’ as he does in the Treatise, referring instead to the ‘common point of view,’ there can be no doubt that the function of the viewpoint remains the same – to act as a safeguard for the intersubjectivity of our moral sentiments.41 Thus, in prose nearly identical to the Treatise, Hume argues in the Enquiry, ‘Sympathy, we shall allow, is much fainter than our concern for ourselves, and sympathy with persons remote from us much although such ‘enlarged affections’ are hard to find, we see some evidence of this more extensive humanity in families (EPM 3.7; SBN 185). If I am correct and humanity is activated through sympathy, then this is unsurprising given that it is in families that sympathy is usually found to be most intensive. 38 This is not to say that extensive benevolence is the original source of justice. It only accounts for our moral approbation of it, and our moral obligation to it. 39 See also, EPM Appx. 3, which argues against private humanity as the source of justice. 40 Abramson, 53–4. 41 Compare, for example, EPM 9.6 (especially ln. 4–13); SBN 272, to T. 3.3.1.14; SBN 580 (‘judicious spectator’); & both 3.3.1.15; SBN 581 and T. 3.3.1.30; 591 (where ‘common point of view’ is used). 40 REMY DEBES fainter than that with persons near and contiguous; but for this very reason it is necessary for us, in our calm judgments and discourse concerning the characters of men, to neglect all these differences, and render our sentiments more public and social’ (EPM 5.42; SBN 229).42 We must, in other words, take up the viewpoint of the judicious spectator in order to extend the scope of sympathy, from which we are to consider only the effects of traits on those who have a close connection to the agent. In short, the kinds of corrections to our moral sentiments which are needed in the Enquiry are structurally analogous to those in the Treatise, and they depend on extensive sympathy. The implication of this parallel is that a more extensive sympathy renders our moral sentiments ‘more public and social’. And isn’t this just to say that expanding sympathy correspondingly expands our sense of humanity? That is, given that we already know that Hume has assigned humanity as the basis of our moral sentiments in the Enquiry, the implication of Abramson’s demonstration seems to be that if we are able to expand our sympathy to take in the interests of society, then this somehow sparks our innate humanity and raises our approbation towards that which conduces to society’s happiness. At least, it suggests that to the extent that we extend sympathy, we extend the scope of our moral sentiments, and a fortiori, our humanity. Or to put the point one other way, because sympathetic deliberation in the Enquiry is unchanged from the Treatise, and in particular because Hume is still committed to the idea that we must take up a viewpoint of extensive sympathy to correct for natural bias, then the implication is that our innate humanity can be raised from this expanded and public viewpoint, just as it can from our private one. Extensive humanity thus seems to follow from extensive sympathy. Or better, sympathy prepares the way for humanity: once (and insofar as) sympathy is extended, so too can we extend our humane concern for others and our moral sentiments that attend this concern. And so, this is further evidence that Hume makes room for extensive humanity in the Enquiry. Two final points: First, the foregoing argument reveals a second manifestation of extensive sentiment: we can have extensive moral sentiments. That is, not only does our humane concern or desire get extended (the motivational aspect of humanity), but also approbation and disapprobation get extended (the judgment aspect of humanity). And while I won’t discuss this consequence in any detail, we must acknowledge its significance.43 Extended moral sentiment or judgment is at the heart of resolving moral disagreement for Hume, a problem that arguably bookmarks the discussion of the Enquiry. For, extended moral sentiments are 42 Compare to T. 3.3.3.2; SBN 602. There are many examples of extensive moral sentiments. See especially, EPM 5.7; SBN 215– 216 (we praise virtue in ‘very distant ages and remote countries’ where there is no ‘appearance of self interest’). See also, EPM 5.10; SBN 216; & 5.41–3; 227–30; & 9.5, 7–8; 272, 273–4. 43 HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 41 precisely what others are able to share in or ‘concur with’ because it is only these impartial sentiments that others unrelated to us easily sympathize with, and thus which provoke their humanity and in turn their moral approbation (EPM 9.6; SBN 272–3).44 Second, and more relevant, the foregoing argument also already begins to advance my next interpretive claim. For, it suggests not only that extensive humanity is possible (and extensive judgment with it), but also that humanity is dependent on sympathy, at least for this very extension. This sort of dependence on sympathy, however, just is the second feature of humanity. But I will have much more to say about this second feature in a moment. For now, I must enter a cautionary note. In making the preceding appeal to Hume’s arguments on sympathetic deliberation, and to the corrective use of the common viewpoint in particular, we seem to have crossed over to a consideration of Hume’s normative claims regarding sympathy and humanity, which might strike some as muddying the waters. At least, up to this point the focus has been tacitly Hume’s moral psychology, not his normative ethics. Unfortunately, Hume himself is systematically unclear when it comes to distinguishing his normative arguments from the moral psychology that saturates them.45 It is an ambiguity that seems to be a deep, perhaps inescapable, feature of Hume’s moral philosophy. But thankfully, it may not be an ambiguity we need to sort out here. My direct goal is to provide an overall reconciling account in response to the initial puzzle of the Enquiry, not an account of Humean moral psychology or Humean 44 Moral ‘dispute’ opens the Enquiry (EPM 1.1–2; SBN 169–70) and is on centre-stage in the conclusion (see especially EPM 9.6–8; SBN 272–4). See also EPM 2.17; SBN 180 & especially 5.9; 216. 45 The problem is general enough for Tom Beauchamp to include it in his introduction to the Student Edition of EPM. Beauchamp writes, ‘It often seems unclear whether Hume has a normative theory of obligation or merely an account of human nature that explains how we come to feel obligated’ (David Hume, An Enquiry Concerning the Principles of Morals, edited by Tom L. Beauchamp (Oxford Clarendon Press, 1998) 40). Indeed, whether or not Humean moral psychology is able to support a normative theory of obligations, and just what this normative ethics amounts to if viable, has proven to be its own longstanding and complicated question in Hume scholarship. The problem, as it has been historically conceived, is a potential paradox endangering the ability of any sentimentalist account to establish a legitimate normative ethics: If our moral distinctions and/or judgments are based on subjective feelings and thus seemingly beyond normative standards of correctness, how could there ever be anything like the objective standard of good or right which morality intuitively seems to require (assuming, that is, that morality contains universal obligations: similar people in similar situations must be similarly obligated)? I shall not, however, presume to be providing a solution to this problem. Its historical legacy, Hume’s strategy of response, and the possible kind of obligation Hume might be developing are ably discussed by Mendel F. Cohen, who argues persuasively that Hume’s moral psychology and normative ethics are intentionally and necessarily mixed (‘Obligation and Human Nature in Hume’s Philosophy’ (Philosophical Quarterly, 40 (1990): 316–41)). And while I am more sanguine than Cohen about Hume’s efforts to justify his normative ethics, I am generally in agreement with Cohen about how Hume tries to establish this justification. See also Stephen Darwall, ‘Motive and Obligation in Hume’s Ethics’ (Nous, 27 (1993): 415–48). 42 REMY DEBES normative ethics per se. Thus, although it is true that the primary case against those who postulate a schism between the Treatise and Enquiry regarding sympathy rests in the psychological details, if my overall defense requires citing evidence from both Hume’s moral psychology and his (possible) normative ethics, this seems benign so long as the two are not recklessly conflated in such a way as to jeopardize the essential interpretation of Hume’s moral psychology on which my defense rests. And I don’t think anything I have said or will say violates this condition.46 This worry aside, however, the conclusions of this section still leave much unexplained. In particular, it is unclear exactly why extensive humanity should follow from extensive sympathy and how our humane disposition is raised at all. To answer these questions, we must turn to the second feature of humanity. II.2 The second feature of humanity: humanity’s dependence on sympathy Consider again my proposal from the end of Section I: Humanity is activated through sympathy. Now, exactly what sympathy amounts to in the Enquiry is, of course, part of the puzzle we are trying to solve. And, as is well known, the Enquiry does not give an explicit account of sympathy. Sympathy makes many appearances, for sure, with sympathy-related concepts appearing 30 times in the second Enquiry.47 But, then again, the 46 I am grateful to an anonymous reviewer at BJHP for urging the foregoing clarifications. It has been suggested to me that sympathy might make more appearances in EPM under different guises, in particular as ‘fellow-feeling’ or ‘compassion’. And although by and large, I think this is right, at least in the case of ‘fellow-feeling’ I believe we must be more cautious. Indeed, while I have benefited from Tom Beauchamp’s enriching annotations and other scholarly apparatus in the new Clarendon Editions of the Enquiries, his annotations indicating just this sort of identification pose some risks. Thus, in the Student Edition Beauchamp remarks on EPM 7.29, ‘In this conclusion to Section 7, Hume is maintaining that sympathy, or fellowfeeling, is the source of both types of ‘sentiments of approbation’ – namely, the sentiments that arise from the agreeableness and the utility of qualities of persons’ (EPM ann. 7.29). Now, there are two problems here. First, I think Beauchamp’s identification of sympathy with fellow-feeling is too quick. So, I agree that in this passage (7.29), fellow-feeling is equated with sympathy. But elsewhere at EPM n19, Hume arguably identifies humanity with fellowfeeling (see my n63 for analysis). This alternative reading combined with the scarcity of references to ‘fellow-feeling’ – and to my knowledge, these are the only two in EPM – lead me to believe that there is insufficient evidence for drawing a general equivalence between the terms. And perhaps Beauchamp himself is sensitive to this. For, in the scholarly Critical Edition the annotation of the same passage is altered to read: ‘The close connection of sympathy and fellowfeeling . . .’ (EPM ann. 66.8, emphasis added). I endorse this more cautious reading. But second, I also think Beauchamp is too quick in labeling sympathy as the source of the sentiments that arise from both the agreeableness and utility of qualities of persons. To understand my objection, however, we first need to register a distinction. Some qualities of objects or persons we approve for their utility. Other qualities, Hume argues, are immediately agreeable to us and we thus approve them without any regard to their useful tendency. Now, I think Hume is systematically unclear about the relationship between these two types of 47 HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 43 approbation. Nevertheless, this much is certain: Despite his labeling of one type of approbation as ‘immediate’, Hume thinks both types of approbation depend in some way on sympathy. This shared dependence is the point of the EPM passage Beauchamp is commenting on, and the same point is also stressed in the Treatise (T. 3.3.1.29; SBN 590; & 3.3.5.5; 616). However, it is far from obvious that Hume’s insistence on this shared dependence is equivalent to sympathy’s being the source of both types of sentiments, as Beauchamp seems to be suggesting. And in fact, in the EPM passage in question, Hume doesn’t actually say that sympathy is the source of both types of sentiment. He writes, ‘The same social sympathy, we may observe, or fellow-feeling with human happiness or misery, gives rise to both’ (emphasis added). Now, right away this puts a burden on Beauchamp’s position because ‘gives rise to’ is ambiguous and can be interpreted in more than one way while still preserving Hume’s insistence on the ‘shared dependence’ relationship: (a) (What fits Beauchamp’s claim) sympathy is the originating source of both types of approbation; (b) The two types have different originating sources – i.e. are activated by different principles, but sympathy is a necessary mediator to both principles (and perhaps either entirely accounts for the immediately agreeable or is not strictly necessary for the immediately agreeable but rather, as Hume says in the Treatise, there is only ‘considerable’ dependence of the immediately agreeable on sympathy, either of which would preserve his label of ‘immediate’); (c) The two types appear to have different sources, but ultimately humanity is the source of both, and sympathy the necessary mediator to humanity. Now, I am not sure if there is conclusive evidence that can decide between (b) and (c) but I do think (a) must be mistaken. First and foremost, all the evidence and argument I present in this paper for understanding humanity as the source of at least the approbation we have with regard to utility tells against option (a). Indeed, articulating a distinction in Hume between source qua originating source of our moral sentiments, and source qua the necessary mediator that ‘gives rise to’ or ‘activates’ this originating source is an important result of this paper. And thus, option (a) requires one to first explain away all my arguments and evidence to this end. Furthermore, despite my sceptical admission about the ultimate indeterminacy between (b) and (c), there are some independent reasons to think something like (b) actually is the case. So note this remark by Hume that appears just before the contentious passage under consideration, concerning the immediate agreeableness of some passions: ‘Yet those more elevated softer affections have a peculiar influence, and please from more than one cause or principle’ (EPM 7.26; SBN 259, emphasis added). This is suggestive on its own, but it also brings to mind a corresponding discussion from the Treatise, where Hume argues, ‘Some of these [immediately agreeable] qualities produce satisfaction in others by particular original principles of human nature, which cannot be accounted for: Others may be resolv’d into principles, which are more general’ (T. 3.3.1.27; SBN 590, second emphasis added). This statement along with the rest of T. 3.3.1 strikes me as strong evidence in favour of option (b). Indeed, wouldn’t it be particularly strange to make this remark, and then go immediately on to assert at T. 3.3.1.29 the shared dependence of both types of approbation on the ‘more general’ principle of sympathy unless something like (b) is what Hume intends? That is, unless sympathy is a shared (albeit necessary) mediator to multiple originating sources? Granted, if this is right, then I must admit a wrinkle in my own argument. For humanity would not be the only source of approbation. But this does not worry me, for it seems pretty clear in both EPM and the Treatise that Hume thinks approbation based on utility is primary and predominant (See again T. 3.3.1.27; In EPM see, for example, 6.2; SBN 233, but more important note that the focus of the concluding Section 9 is utility-based approbation). This said, perhaps Beauchamp is using ‘source’ more loosely, as Hume himself seems to. Indeed, for all I have said, even if EPM does not make sympathy the ‘source’ of our moral sentiments, the Treatise does credit sympathy in this way, which is why we have a puzzle in the first place. But if the interpretation I am presenting is correct regarding the essential dependence of humanity on sympathy, we will now have cause to be more careful. For, making the distinctions I am proposing allows us to make better sense both of the Treatise and EPM claims regarding sympathy, utility, and the immediately agreeable, not to mention the 44 REMY DEBES mere presence of sympathy in the Enquiry was never in question. What is at stake is the function of sympathy and its psychological explanation. Moreover, even when sympathy does appear in the Enquiry it is never given the sort of detailed, explicit analysis found in the Treatise (where, as a point of curiosity, sympathy-related concepts appear over 80 times in each of Books 2 and 3 of the Treatise).48 So what can be said about sympathy in the Enquiry, and in turn, humanity’s dependence on sympathy, requires a more subtle approach to the text. II.2.1 Sympathy qua sympathizing: First, it will help to clarify that when I refer to sympathy I am referring to the process of sympathy (or sympathizing) and sometimes to the principle of sympathy, but not to a feeling of sympathy if this means anything like a concern for the person or persons with whom I sympathize. A concern for others in this latter sense just is humanity activated. That is, it is already the sentiment of humanity. There is, however, a definite distinction between this sense of sympathy (as humanity activated) and sympathizing – the process by which feelings are shared or imparted between people, and what we might today call empathy, though this term is foreign to Hume. Most important, sympathizing is clearly a means for representing the feelings of others to ourselves. It is this sense of sympathy qua sympathizing, or as I will sometimes call it, sympathetic representation, with which I am immediately concerned. Whether or not sympathy is a distinct principle can be decided later. Of course, the Enquiry also does not give an explicit account of how sympathetic representation works. But this is hardly the last word. At a minimum there can be no denying that, whatever the psychological mechanism behind sympathy turns out to be, the notion of sympathy qua sympathizing is woven throughout the Enquiry. A short illustration of the fact will suffice. Note first that sympathy only really becomes the focus of the Enquiry starting in Section 5 Part 2, Why Utility Pleases. In fact, while it is true that sympathy-related concepts appear outside of this section and that some of these appearances are early and significant – not to mention expected, if the interpretation I am offering is right – as in the discussion of benevolence in EPM Section 2 (EPM 2.5; SBN 178), never again in the Enquiry does sympathy get so much direct attention or mention as in Section 5 Part 2. I think it fair to say, then, that if there is a place in the Enquiry where Hume is directly discussing sympathy, this is it. In any event, here begins Hume’s steady attack on theories deriving the basis of moral distinctions clarity it brings to the ethical arguments of EPM in general, and to the account of humanity in particular. 48 See the comparative table of Substantive Changes in Beauchamp’s introduction to EPM, lxiii. Interestingly, though unsurprising by this point, ‘humanity’ increases from a mere eleven appearances in Book 3 of the Treatise (and no appearances in Book 2) to sixty appearances in EPM. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 45 from self-love as opposed to humanity. Horace had it right, Hume opines near the start of this section, that the human countenance ‘borrows smiles or tears from the human countenance.’ Hume then adds, ‘Reduce a person to solitude, and he loses all enjoyment, except either of the sensual or speculative kind; and that because the movements of his heart are not forwarded by correspondent movements in his fellow-creatures’ (EPM 5.18; SBN 220). This certainly sounds like the kind of sympathetic representation I have described, and indeed, this whole stretch of the Enquiry consists of illustrations of such sympathy. Hume himself tips us off to this early on, writing, ‘In general, it is certain, that, wherever we go, whatever we reflect on or converse about, everything still presents us with the view of human happiness or misery, and excites in our breast a sympathetic representation of pleasure or uneasiness’ (EPM 5.23; SBN 221). And lest we worry that this last passage seems to gloss over the strict sympathetic sharing of feeling I am appealing to with some different, subsequent sentiment of pleasure or uneasiness – which here also seems to arise from humanity (and which is thus important for explaining the connection to approbation) – we need only read into the accompanying description of the theatergoer to assure ourselves that Hume is making definite room for sympathy qua sympathizing: A man who enters the theatre, is immediately struck with the view of so great a multitude, participating of one common amusement; and experiences, from their very aspect, a superior sensibility or disposition of being affected with every sentiment, which he shares with his fellow-creatures. . .Every movement of the theatre, by a skilful poet, is communicated, as it were by magic, to the spectators; who weep, tremble, resent, rejoice, and are inflamed with all the variety of passions, which actuate the several personages of the drama. (EPM 5.24; SBN 221–2) It is not a feeling of concern for the audience members or actors that strikes the theatergoer upon his sudden entrance. At least, that is not what Hume, with some deft imagery himself, is here emphasizing. Rather, the theatergoer and his fellow audience members are struck by a progression of distinct sentiments, each of which is the same in kind as the original being acted out on stage. The theatergoer feels the same sorrowful weeping, fearful trembling, bitter resentment, or merry rejoicing that the characters of the drama feel. Such transmission of sentiments is obvious sympathizing.49 II.2.2 Sympathy’s function: Enough said, then, about sympathy qua sympathizing; the rest of the Enquiry can easily be seen to preserve this notion. More important, we can now better discern sympathy’s function in the argument of the Enquiry and in particular its function in generating the moral sentiments. 49 There is also good evidence of Hume’s commitment to sympathetic representation before the Enquiry in his Essays: Moral, Political, and Literary, which I discuss briefly below (II.2.3). 46 REMY DEBES Consider: we know it is by a disposition towards humanity that moral judgments are made. That which I represent as promoting the happiness of others my humane disposition leads me both to desire and to approve. Hence I approve of joy and deplore misery. But I must also have some way of representing joy and misery, virtue and vice, if some innate disposition of mine is to raise these sentiments in me. Humanity requires some object or stimulus in order to become engaged – this was the conclusion from Section I. But, as we just saw, Hume clearly thinks that we do sympathize with joy and misery, virtue and vice. That is, it is clear that Hume thinks there is such a thing as sympathy qua sympathizing by which we represent the joy and misery of others, or any other passion that affects them. What other conclusion should we draw, then, except that humanity depends on sympathy? More precisely: the humane disposition depends for its engagement on the intermediate operation of sympathetic representation. We must first sympathetically enter into the passions and motives of others before our humanity can be aroused, and thus before we can make moral distinctions or judgments. When I sympathetically contemplate a quality or motive in an agent, I consider the advantage or happiness it will have a tendency to promote (to him or to others). That such representations of happiness can affect me is due to my innate humanity. But outside of my own interest, the only way the happiness of others actually does affect me, is by sympathy. It is by sympathetic contemplation of the well-being or happiness of others that our feelings of humanity are affected: ‘What sympathy then touches every human heart!’ (EPM 5.34; SBN 223). This might seem too fast. Humanity could simply be a standing desire for the good of others – a concern for persons as such, unmediated by any representation of their joy or misery, virtue or vice. This would not deny the phenomenon of sympathy. But it would allow room for the objection that the mere representation of a person would be sufficient to generate a care for another person’s welfare, and sympathy would play no essential role in generating such humane feelings. Perhaps this is the view of the Enquiry. This hypothesis, however, goes against the grain of textual evidence both in the Enquiry and the Treatise. It certainly stands in stark contrast to Hume’s claim in the Treatise, that ‘we have no extensive concern for society but from sympathy’ (T. 3.3.1.11; SBN 579). Or as he says with a bit more gusto: In general, it may be affirm’d, that there is no such passion in human minds, as the love of mankind, merely as such, independent of personal qualities, of services, or of relation to ourself. ‘Tis true, there is no human, and indeed no sensible, creature whose happiness or misery does not, in some measure, affect us, when brought near to us, and represented in lively colors: But this proceeds merely from sympathy, and is no proof of such an universal affection to mankind, since this concern extends itself beyond our own species. (T. 3.2.1.12; SBN 481) HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 47 Of course, this evidence from the Treatise cannot alone defeat the objection, as one premise of the broader puzzle under consideration in this paper is that Hume might have changed his view from the Treatise to the Enquiry.50 Indeed, because Hume does not make a similar denial of a concern for persons ‘merely as such’ in the Enquiry, we might hypothesize that there was a change of positions on this issue after all. Nonetheless, I’ll take it as a point in favour of my interpretation if it entails that the Enquiry did not change in this respect: Consistency between philosophical works, where not explicitly rejected by an author, or a contrary change endorsed, always tells in favor of an interpretation of those works. And while it is true that Hume does not repeat the Treatise denial of a concern for persons merely as such, Hume never explicitly rejects this position nor endorses a change from it, in the Enquiry. This said, my reading explains why we should expect exactly this; that is, why we should expect Hume to maintain the Treatise position. For, if I am correct about the relationship between humanity and sympathy, no notion of humanity as a concern for others merely as such could exist in the Enquiry. Any actual concern for another’s well-being is instead necessarily mediated by sympathetic representation. And this is exactly the direction in which we find Hume arguing. Thus, in the conclusion to the Enquiry, Hume argues, ‘There is no circumstance of conduct in any man, provided it have a beneficial tendency, that is not agreeable to my humanity, however remote the person’ (EPM 9.8; SBN 274, emphasis added). Right away this tells against the possibility of a standing concern for persons merely as such. I must consider a person’s circumstances in light of its beneficial tendencies if I am to care about her. In fact, when viewed in its larger context, Hume’s point in insisting that a representation of circumstances engages humanity ‘provided’ it has a beneficial tendency is just this: it is only because of the universality of our disposition to be concerned for others that universal approbation of beneficial actions and circumstances – which we know Hume endorses – is even possible. Thus just before this passage Hume argues, ‘Whatever conduct gains my approbation, by touching my humanity, procures also the applause of all mankind, by affecting the same principle in them’ (EPM 9.5; SBN 274). So we must have some way of representing these beneficial tendencies if we are to have any humane concern for others. Isn’t sympathy, then, the most plausible explanation for this representation? I think yes, and the text is supportive. First, we find in the last quote the familiar language ‘touching my humanity’, which, as I have already suggested (Section I), implies not just any kind of representation, but emotional representation, and hence the intermediate operation of sympathy. And we can begin to see why. Hume 50 Actually, this is already charitable. For, it could be rejoined that ours was a puzzle about sympathy, not humanity or benevolence, the latter of which is the real subject in the Treatise quote. 48 REMY DEBES is not laying a sufficient condition for humanity in the clause, ‘provided it have a beneficial tendency.’ Yes, it is true that humanity is activated, or triggers or motivates conduct, via the consideration of beneficial consequences to other human beings. But it is not the mere thought of such consequences that does so. The mere thought that someone faces injury does not make me rush to her aid. Rather, what moves me on Hume’s account is the possibility of misery or suffering that will attend her injury. That is, it is felt consequences that engage my humanity and raise some motivating concern for her welfare. Humanity, in other words, is not sparked by a belief about the probable effects of one’s character or action, or other object. It is sparked by imagining the effects themselves, which for Hume entails that we sympathetically share those effects to some degree. Indeed, this is the subtle thrust of EPM Section 5 Part 2, Why Utility Pleases, which I first discussed for its role in highlighting sympathy qua sympathizing, but which is now worth a second look in this new context. II.2.3 A closer look: EPM 5 Part 2: The discussion in this part of the Enquiry is framed by what I will refer to as the lead-in question: Hume asks, ‘Have we any difficulty to comprehend the force of humanity and benevolence? Or to conceive, that the very aspect of happiness, joy, prosperity, gives pleasure; that of pain, suffering, sorrow, communicates uneasiness?’ (EPM 5.18; SBN 220, emphasis added). In light of all that has been said so far, however, it should no longer seem like coincidence that this lead-in question is answered by way of Horace’s previously noted observation on imparted feeling, Hume’s sympathetic illustration of the theatregoer, or the many other obvious examples of sympathy qua sympathizing in this section. For while the explicit aim of this section is to illustrate the superior plausibility of humanity over self-love as the source of moral approbation, Hume is also implicitly pressing the idea that the objects of our disposition towards humanity are things felt – passions and motives, happiness and misery. This idea is reflected in the second part of the lead-in question (and reiterated at EPM 5.23; SBN 221, quoted earlier in this paper in II.2.3). Where the felt objects of our disposition are beneficial I feel pleasure; where harmful I feel pain or uneasiness. Once I represent the feelings of others, then I feel pleasure or pain. And here again is the connection between moral judgment and humanity discussed at the start. For we approve or disapprove based on whether the object of our contemplation is agreeable or uneasy, and it is humanity itself that raises pleasure or pain, depending on whether the felt objects of contemplation agree or frustrate this fundamental disposition to care about the welfare and happiness of others.51 For, where the objects of 51 We might compare my explanation here, of objects agreeing with or frustrating the humane disposition, to Hume’s occasional use of agreeable as meaning that something accords with or conforms to the sentiment or principle of humanity (e.g. EPM 6.4; SBN 235). I also think my HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 49 contemplation agree with this disposition (such as with happiness and joy) I feel the pleasurable sentiment of humanity; and where they disagree (such as with misery) I feel the painful counterpart of humanity, an aversion or resentment.52 Now, that I feel pleasure or pain is due to humanity – this is the explicit message of section 5.2. But, that I feel the original passion of happiness or joy, suffering or sorrow, is due to sympathy – this is the implicit and critical underlying message. This is also why Hume answers the lead-in question by immediate reference to our capacity to sympathize with the feelings of others: it is a key part of the answer to why utility pleases – the main question of the section. The force of humanity depends upon sympathy. And, as we should expect, this reading helps make sense of Hume’s subtle wording at many other points in this section and throughout the Enquiry. For example, consider the first illustration following the lead-in question: ‘We enter, I shall suppose, into a convenient, warm, well-contrived apartment: We necessarily receive a pleasure from its very survey; because it presents us with the pleasing ideas of ease, satisfaction, and enjoyment’ (EPM 5.19; SBN 220). Keeping in mind that the explicit goal of this section of the Enquiry is to demonstrate the workings of humanity, the obvious point to make is that the good arrangement of the apartment, by itself, can hardly explain why humanity might be raised, and hence why I feel any pleasure.53 I must instead be able to conceive of the benefit that good arrangement has for some person or persons, if humanity is to be relevant at all. Moreover, Hume’s language suggests that a belief about the possible pleasure or pain of the inhabitants is not what is doing the work. Instead, the passage suggests that what is entertained is the actual idea of ease or enjoyment directly (or perhaps better, vividly) – an idea which a probabilistic belief about the arrangement of the apartment only suggests as possible or likely. How else then might I directly entertain the ease or explanation fits with Hume’s language in the Treatise (see e.g. that some pleasurable impressions ‘by concurring with certain dormant principles of the human mind’ excite secondary passions (T. 2.3.9.3; SBN 438, emphasis added)), and with the Dissertation (see e.g. the opening of this work: ‘Some objects again, by being naturally conformable or contrary to passion, excite an agreeable or painful sensation; and are thence called Good or Evil’ (1.2, p. 139, first two emphases added)). But I must pass over a detailed exploration of these connections. 52 This reveals the full import of Hume’s early description of our praise for the benevolent person at the end of the Section 2 of the Enquiry: ‘We carry our view into the salutary consequences of such a character and disposition; and whatever has so benign an influence, and forwards so desirable an end, is beheld with complacency and pleasure’ (5.2; SBN 181). It is because of sympathy that we can ‘carry our view’ to the consequences and actually feel them; it is because of humanity that these ‘salutary consequences’ give us pleasure by raising the pleasurable sentiment of humanity; and it is based on this pleasure that we approve. 53 Keep in mind: Qualities immediately agreeable are discussed later in Sections 7 and 8 of EPM. For the wrinkle that the immediately agreeable may cause in my interpretation see n47. 50 REMY DEBES enjoyment from the survey of a well-contrived apartment, but by some imaginary sympathy with those who I suppose to live there?54 This interpretive line also illuminates the many literary examples Hume presents in Section 5 Part 2 as well as the persistent references to the imagination in these examples. Indeed, it is in the context of these literary examples that Hume makes the much earlier quoted claim that ‘no passion, when well represented, can be entirely indifferent to us’ (EPM 5.30; SBN 222). But now we can understand the import of it. Hume is at pains to show that in order to engage us – that is, to interest us – literature must sympathetically engage our affections. But why must it do so, unless these affections are the objects of humanity? Thus literature must ‘bring every affection near to us by lively imagery and representation’ (EPM 5.30; SBN 222–3).55 ‘The perusal of a history,’ Hume adds as if to make the point certain, ‘seems a calm entertainment; but would be no entertainment at all, did not our hearts beat with correspondent movements to those which are described by the historian’ (EPM 5.32; SBN 223). The point is that if we are to be interested in literary scenes, we must first be sympathetically engaged. Sympathy first, interest and humanity second. Coldly relating the facts, while it may convince us of a historian’s point, just doesn’t engage us. This is why we barely attend to the dry descriptions by Thucydides of the trivial affairs of small Greek cities: ‘The few persons interested and the small interest fill not the imagination, and engage not the affections.’ But, as Hume eagerly points out, ‘The deep distress of the numerous Athenian army before Syracuse; the danger which so nearly threatens Venice; these excite compassion; these move terror and anxiety’ (EPM 5.33; SBN 223). 54 One might balk here, and object that ‘directly’ entertaining the idea of ease, enjoyment, or any other emotion, need not entail sympathy. That is, even if belief about the likelihood of a particular emotion doesn’t suffice, the idea of the emotion per se can still be the content of a belief, without requiring or leading to any feeling, much less the feeling represented by the idea. Hence, belief alone could trigger humanity after all, without any intermediate operation of sympathy. There are two replies to this objection. First, even if this were true, this would not threaten my interpretation. Thus far, I am only trying to explain the text. And the text suggests that Hume believes humanity requires sympathetic representation. It remains to be seen if Hume is correct. But second, if I am right that the associationist explanation of sympathy has not disappeared from EPM, as I hinted at the start of this paper and argue for at length in ‘Has Anything Changed?’ (see n7), then Hume does have an answer: the idea of another’s emotion, considered merely as such, always results in my feeling some of that same emotion, through an association of ideas in which my own idea of another’s emotion is enlivened into an impression, or feeling. At least, that is roughly how Hume explains sympathy associationistally. Granted, this would not settle whether it is the original idea of another’s emotion or my own sympathetically enlivened impression of that idea that is the trigger for humanity. But it does give Hume the option of arguing that no real distinction can be made. If the idea of another’s emotion is always attended by an impression of that idea, and vice versa, then it is impossible to say which one is the real trigger of humanity. 55 This comment describes poetry, but it is clear from what follows that he intends the point for all literature. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 51 Likewise, although it casts the net of evidence beyond the Enquiry, this interpretation also fits nicely with what Hume writes elsewhere on related topics in his Essays: Moral, Political, and Literary (hereafter, Essays) – a work that was part of the multivolume collection Essays and Treatises on Several Subjects, in which the Enquiry appeared as one treatise from 1758–77.56 In the essay ‘Of Eloquence’, for example, Hume argues for the importance of arguing passionately (literally, with emotion) so as to engage the listener’s sentiments as well as her ears. What is striking in this essay, then, is the obvious appeal to sympathetic representation as a means for garnering interest or concern. Hume remarks at one point as he praises the ancient rhetors, ‘The orator, by the force of his own genius and eloquence, first inflamed himself with anger, indignation, pity, sorrow; and then communicated those impetuous movements to his audience’ (‘Of Eloquence’ 13).57 And we find this same sort of appeal in the background of ‘Of Tragedy’. Hume’s main goal in this essay, of course, is to explain the strange fact that we take pleasure in eloquently dressed up misfortune. According to Hume, the same artistic expression required to raise our emotions also impresses us with its beauty, hence leaving us with a mix of pleasure and sorrow. But for our purposes, what is important are the constant reminders in this essay of the need for the poet to use sympathetic representation to engage our interest and concern. ‘The whole art of the poet’, Hume states right at the start, ‘is employed in rousing and supporting the compassion and indignation, the anxiety and resentment, of his audience’ (‘Of Tragedy’ 1). And although my primary goal has been to demonstrate these features of humanity and sympathy in the Enquiry, noting some obvious parallels with the Essays, which were also published after the Treatise and alongside the Enquiry, can only fortify my conclusions. Finally, we would be missing something important if we failed to note the careful literary imagery Hume himself deploys throughout so much of his writing, but especially in the Enquiry and the Essays. It would also be a mistake to suppose it unintentional. Hume adds literary detail to his examples just as he asks us – his readers – to accept his claim that humanity is a universal psychological principle of human nature. In doing so, Hume must be understood as trying to get us to acknowledge our own humane feelings by actually feeling them. But – and this is the relevant point for my argument – in order to do this, Hume first tries invoking our sympathy for the distress or happiness of the characters in those very examples. For of course, Hume’s literary details are exactly those that help us imagine the 56 For a history of Hume’s publications including a helpful schematic of the evolution of the Essays and Treatises on Several Subjects, see Beauchamp’s introduction to EPM, xxvii. 57 All citations to individual essays from this work are to David Hume, Essays: Moral, Political, and Literary, edited by Eugene F. Miller, second edition (LibertyClassics, 1987) with the name of the essay followed by paragraph number. I am grateful to an anonymous reviewer for urging consideration of the Essays. 52 REMY DEBES passions of his characters. Once this is done, our heartstrings are already tugged, and hence the argument for humanity is made implicitly. This is, admittedly, a sort of meta-argument on Hume’s part, and one that is never explicitly mentioned. But it would be a shame to overlook this sophistication of Hume’s writing. And once recognized, this subtlety bolsters the interpretation I am offering. II.2.4 Humanity’s dependence on sympathy concluded: Given the analysis in the preceding section, it makes most sense to read the Enquiry as explaining humanity as a concern for others provided that certain conditions are met; mainly that we are considering a beneficial tendency but also that this consideration takes place via sympathetic deliberation. I must feel, if only imperfectly and weakly, some degree of that misery which might attend an injury or that joy that might attend some benefit, if I am to have any concern for the person or persons in question. Humanity is thus an original disposition to be concerned about others, but this concern for others is only activated when the passions or motives of others are the objects of sympathy. For, on Hume’s account, it is only in sympathetic deliberation that it even makes sense to say that the thought of beneficial tendencies can be ‘agreeable to my humanity.’ Thus, there is no room for a generalized concern for human beings ‘merely as such’ in the Enquiry, which explains why we do not find Hume rejecting the Treatise denial of such a sentiment. We do not care about others because we are aware of an innate dignity in them, nor is humanity a respect that is owed to human beings simply because they are human beings. Rather, we have a natural psychological readiness to be concerned about others. Hence, it is only insofar as other human beings are the objects of my sympathetic contemplation that my innate humanity raises a care for them. And so here again, looking through the lens of humanity helps show us that there is no inconsistency between the Treatise and the Enquiry. More generally, it is now evident why humanity is the source of our moral concern for the happiness of others, but also why it is the source of our moral distinctions. It is because I have a disposition to care about the welfare of others that whenever I perceive their happiness to be realized I feel the pleasurable sentiment of humanity, and whenever happiness is frustrated or destroyed I feel its painful counterparts (e.g. aversion). And it is upon these pleasures and pains that moral approbation or disapprobation is based or immediately linked. But because this concern requires sympathy to be affected, the only way I will be able to judge the merit of an agent is by means of sympathy. Hence Hume argues, ‘If any man from a cold insensibility, or narrow selfishness of temper, is unaffected with the images of human happiness or misery, he must be equally indifferent to the images of vice and virtue’ (EPM 5.39; SBN 225, emphasis added). And so we should refine our understanding of Hume’s account of humanity by formally noting this second important feature: humanity is dependent on sympathy. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 53 III. LOOSE ENDS Despite the dependence relationship of humanity on sympathy, I do not think humanity and sympathy are simply two functions of a single principle. At least, the Enquiry gives us good reason to read Hume as suggesting that humanity and sympathy are two distinct principles, which necessarily work in tandem as the source of our moral sentiments. Recall that it is, after all, An Enquiry Concerning the Principles of Morals, not the principle or a principle of morals. Obviously, this is only suggestive and not directly indicative of whether sympathy and humanity, per se, are distinct principles. But in fact, in the Enquiry Hume is often careful to speak of sympathy and humanity as two principles: ‘It appears that a tendency to public good, and to the promoting of peace, harmony, and order in society, does always, by affecting the benevolent principles of our frame, engage us on the side of the social virtues. And it appears, as an additional confirmation, that these principles of humanity and sympathy enter so deeply into all our sentiments, and have so powerful an influence, as may enable them to excite the strongest censure and applause’ (EPM 5.45; SBN 231, emphases added). This seems a particularly telling passage given that it occurs in the conclusion to Section 5 Part 2, which has already revealed itself as a crucial, and perhaps the crucial discussion of sympathy and sympathy’s function in the Enquiry. This one statement, however, is not the only evidence that Hume is thinking of sympathy and humanity as distinct principles. In the summary explanations of this same concluding part of Section 5, Hume actually uses the plural ‘principles’ four times. And this matches up well with the beginning of Section 5, where straight away Hume indicates that he is on the trail of multiple ‘principles.’58 These examples may seem minor, but they are explicit. And while it is true that outside of Section 5 the rest of the Enquiry emphasizes the principle of humanity, and not the several principles of sympathy and humanity, it would risk begging the question against me to insist that we should consequently downplay Hume’s wording in Section 5. For, if I am right, sympathy’s main function is to explain why utility pleases. And so outside of the discussion of why utility pleases, which to reiterate, is the main issue in Section 5 Part 2, Hume should have far less need for sympathy explanations, and thus for any reference to the principle of sympathy.59 And yet, I imagine that these purely terminological appeals may not fully satisfy. Against my interpretation, then, one might ask why humanity is needed as a separate principle in the first place? Why can’t its workings be 58 ‘And if we would employ a little thought on the present subject, we need be at no loss to account for the influence of utility, and to deduce it from principles, the most known and avowed in human nature’ (EPM 5.2; SBN 213, emphasis added). 59 The caveat to this claim is Hume’s explanation of qualities immediately agreeable; see n47. 54 REMY DEBES explained by the principle of sympathy itself?60 On this reading, humanity and sympathy are really just two sides of the same coin. Humanity is a disposition, in virtue of our being the sort of creatures that sympathize, to experience benevolent sentiments. Thus, humanity (as a principle) consists in our being the sort of creatures who sympathize. And to ‘affect’ us or to ‘touch’ us is just to trigger or activate sympathy and hence humanity. Moreover, wouldn’t this do justice to the many Treatise remarks crediting sympathy as the source of our moral sentiments, which partly sparked the initial puzzle of the Enquiry in the first place? In particular, consider what Hume says in the final stretch of the Treatise: Thus it appears, that sympathy is a very powerful principle in human nature, that it has a great influence on our taste of beauty, and that it produces our sentiments of morals in all the artificial virtues. From thence we may presume, that it also gives rise to many of the other virtues; and that qualities acquire our approbation, because of their tendency to the good of mankind. . .For having found, that such tendencies have force enough to produce the strongest sentiment of morals, we can never reasonably, in these cases, look for any other cause of approbation or blame; it being an inviolable maxim in philosophy, that where any particular cause is sufficient for an effect, we ought to rest satisfy’d with it, and ought not to multiply causes without necessity (T. 3.3.1.10; SBN 577–8)61 Here Hume stresses not only sympathy’s primary role as the source of our moral sentiments, but seems to stress its singularity as a principle. To this objection, I split my response. On the one hand, I’m not sure anything hangs on a reply. The central points of my interpretation, at least, are left untouched by the reading it proposes. For, even if we take the function of humanity to be subsumed by the principle of sympathy, this is not to say that it is an indistinguishable concept from sympathy. On the contrary, the argument thus far suggests a distinct and special conceptual role for humanity. In light of my analysis of the Enquiry, we can even see that the just quoted passage from the Treatise is infused with the same distinct conceptual reference to humanity: Sympathy, that passage tells us, produces our moral sentiments ‘because of their tendency to the good of mankind.’ This mirrors Hume’s wording in the Enquiry that virtue pleases provided it have a beneficial tendency (or more accurate: the Enquiry mirrors the Treatise), which proved a meaningful indicator of humanity’s dependence on sympathy. And in fact, the broader context of this crucial Treatise passage is largely dedicated to this purpose: to confirm that it is consideration of an action’s tendency towards utility that primarily 60 I am grateful to Steve Darwall for comments on an early draft of this paper in which he raised this objection, and to Louis Loeb who raised a similar concern and whose wording I have partly borrowed. 61 See also, T. 3.3.1.23; SBN 586; & 3.3.6.1–3; 618–19 (among others). HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 55 62 recommends the action for our approval. Granted, humanity takes a backseat to sympathy in the Treatise, and thus important work remains to bring out fully the parallels between the Enquiry and Treatise that my interpretation suggests, but which these Treatise remarks on virtue provide only a tantalizing glimpse. But that is a project in its own right. The immediate point is, these remarks present no threat. It was part of my original hypothesis that there is a change from Treatise to Enquiry – a change in emphasis. Humanity becomes the focus, and its role, which is subtle and secondary in the Treatise, is in the Enquiry stressed over sympathy. And whether or not humanity and sympathy are two manifestations of the same principle or two distinct principles doesn’t impact their conceptual roles for Hume, especially as regards humanity’s dependence on sympathy. So it is tempting simply to call this a verbal dispute and concede the objection. On the other hand, I have a hard time overlooking the explicit textual evidence from the Enquiry, regardless of how slim it is, where Hume uses the plural to talk about the principles of humanity and sympathy.63 Why, we must ask, would Hume ever talk of humanity and sympathy as multiple principles if he did not conceive the possibility of their being distinct? And even if we overlook this evidence, the objection also seems to overly downplay just how different the psychological functions of sympathy and 62 This is particularly clear starting at T. 3.3.1.11; SBN 578–9, and what follows. Indeed, what textual evidence specifically from EPM supports the opposite proposal of a single principle? Perhaps Hume’s footnote 19 in the crucial Section 5 Part 2, where he discusses ultimate principles generally: ‘It is needless to push our researches so far as to ask, why we have humanity or a fellow-feeling with others. It is sufficient, that this is experienced to be a principle in human nature’ (EPM n.19; SBN 219, emphasis added). But this stands as evidence only on the assumption that ‘fellow-feeling’ denotes sympathy, and I must confess that this identification is not obvious to me from this footnote or generally in Hume (see also my n47). On the contrary, it strikes me as equally if not more plausible to read Hume as identifying humanity with fellow-feeling in this footnote, especially given his attending explanation of these terms one sentence later: ‘No man is absolutely indifferent to the happiness and misery of others.’ This is closer to a description of humanity than sympathy. That is exactly the point of my argument. Humanity is what prevents me from being ‘indifferent’ once sympathy has represented that happiness or misery to me. So, of course, fellow-feeling connotes sympathy; this is just what we should expect given humanity’s dependence on sympathy. But nothing in this footnote makes me think fellow-feeling denotes sympathy. And lest there is worry that, at a minimum, my interpretation of humanity’s dependence on sympathy violates what fn. 19 stresses, mainly that humanity in Hume’s opinion is irreducible, recall that my argument only advocates a functional dependence. Humanity is not explained by sympathy. Indeed, this is part of my motivation for distinguishing them as two different principles in the first place, a distinction which, it seems to me, harmonizes well with the background sentiment of this passage; mainly, not to try and over-reduce the principles of human nature. As Hume elsewhere laments of philosophers, ‘There is one mistake to which they seem liable, almost without exception; they confine too much their principles, and make no account of that vast variety which nature has so much affected in all her operations’ (‘The Sceptic’ 1). It is a further, and interesting question, how exactly these remarks on scientific method mesh with the warning not to ‘multiply causes’ at Treatise 3.3.1.10, quoted above. 63 56 REMY DEBES humanity are from one another. To my mind, sympathy qua sympathizing is a remarkably different phenomenon from a disposition to care about those with whom I sympathize. So why conflate them? Moreover, the course of our analysis has revealed just how philosophically rich both humanity and sympathy are for Hume. Is it not more fruitful, then, to admit a clear distinction between these concepts when we interpret his work?64 And so, in all I remain hesitant to make the concession suggested above. However, as the concession reveals the tangential and ultimately benign nature of this objection to my central task, and as there are always some questions that must be sidelined in any inquiry constrained by space, I will set this one aside. Let us, then, take stock. We have seen that fleshing out Hume’s concept of humanity naturally begins to put together the supposed puzzle of the Enquiry. Hume is clear in the Enquiry that humanity has the capacity to take on an extensive form (wide scope and public), as does sympathy in the Treatise, in order to correct the natural biases in our moral sentiments and render these sentiments themselves more extensive. Second, humanity is so dependent on sympathy, the two principles function almost seamlessly together. From this we were able to conclude that the Enquiry, like the Treatise, rules out the possibility of an original extensive concern for others unmediated by sympathy. But, most importantly, it is now clear that the Enquiry has in no way eliminated what the Treatise insisted on so smartly, mainly the need for extensive sympathy in making moral distinctions. For on the Enquiry account, extensive humanity, which is necessary for proper moral deliberation, requires extensive sympathy. This is just what humanity’s dependence on sympathy entails. Humanity is extended by sympathy. Or perhaps more accurate: to the extent sympathy is extended, humanity can follow. And so in arguing that the principle of humanity is the basis of our moral distinctions in the Enquiry, Hume has not essentially changed his position from the Treatise, wherein he famously credited our moral distinctions to extensive sympathy. Rather, it seems we have good reason to take Hume at his word when he says in the Advertisement for the Enquiries that any change to be found in that work is more one of expression, than content. At least, based on my analysis in this paper, it would make more sense to read the Enquiry as simply stressing the role of humanity or benevolence over the role of sympathy, and not as replacing or rejecting sympathy. Of course, there is still the further problem regarding the Enquiry’s seeming abandonment of 64 Granted, in light of Treatise passages like T. 3.3.1.10 (just considered), this might entail admitting a second change from the Treatise: Hume might have decided by the time of the Enquiry that there was cause to think humanity its own principle, and that, contra the Treatise, it was an oversimplification to think of these two functions as a single principle. Of course, this assumes the Treatise really was advocating humanity and sympathy as a single principle. For some evidence and argument that tells against this, see my comments on the ‘immediately agreeable’ in n47. HUMANITY, SYMPATHY AND HUME’S SECOND ENQUIRY 57 the associationist account of sympathy from the Treatise, noted in the initial puzzle. And ultimately, my overall claim that Hume has not changed his doctrines between the two works requires consideration of the solution I offer elsewhere to this further piece of the puzzle.65 For now, though, it is remarkable that following the tracks of humanity has essentially aligned us with critics like Abramson, who argue that the two works are not only consistent but any apparent change is probably better explained by Hume’s pedagogical motives. The route we have taken, however, is quite different and should for that reason be seen as a strong support for such arguments; for although the direction of investigation was different the result was the same.66 University of Memphis 65 See n7. Special acknowledgment is owed to Louis Loeb. His expert advice and encouragement on this paper were invaluable, and I will forever be grateful to him for it. I am also especially grateful to an anonymous referee at BJHP, without whose extensive written comments this paper would surely have suffered, and to Steve Darwall for his comments on the earliest drafts of this paper. I also wish to thank John Rogers for his patience and diligence along the way, Anu Debes for her steady support, Rachana Kamtekar, Lori Watson, and Gabriel Zamosc-Regueros. Finally, I am indebted to Robert Gressis for his many helpful comments and stylistic suggestions on the penultimate draft. 66