1

David Hume's most brilliant and ambitious work is entitled A Treatise of Human Nature, and it, together with his other writings, has left an indelible mark on philosophical conceptions of human nature. So it is not merely the title of Hume's work that makes discussion of it an appropriate inclusion to this volume, but the fact of its sheer influence. However, its pattern of influence – including, of course, the formulations of ideas consciously antithetical Hume's own – is an immensely complex one, subtle and incredibly difficult to decode. In all probability ‘Hume's’ presence in contemporary thinking of human nature is to likened to the end product of a historiographical game of Chinese whispers, whereby ‘Hume's’ view on x and y is now inflected with interpretations his work – or, more accurately, selected parts of it – that are in turn filtered by thinkers and traditions with different focuses and interest from Hume's own. I am not equipped even to begin to trace this line of influence, a lack compounded by my relative ignorance of the present state of the debate on human nature. Nevetheless various ‘humean’ doctrines still orient debate (even if they aren't labelled as such) and I guess these claims include the idea that causation is a matter of instantiating a universal regularity, that normativity can understood causally, that motivation is a matter of belief plus some independently intelligibly ‘attitude’, that a self is best conceived as a collection of independent states that (somehow) combine to yield a self and so on.

It may or may not come as a surprise that, within the narrow confines of Hume scholarship, much of what is taken to be central ‘humean’ doctrine has been probed, debated, questioned and sometimes rejected. I don't, however, propose to become embroiled in the nitty-gritty of these exegetical issues (though inevitably what I say will be informed by my own views). Instead I shall make some remarks about how the author of A Treatise of Human Nature conceives of human nature. I shall do this by considering the relation between aspects of Hume's philosophy and the age-old sceptical trope of drawing comparisons human thought and behaviour with animal thought and behaviour. This trope has both descriptive and evaluative dimensions. The descriptive dimension is that human beings and animals are not, cognitively speaking, radically different in kind and so human nature and animal nature, are, in the relevant respects, the same. The evaluative aspect concerns the presumptuousness, as it were, of the idea that humans and the beast are in fact different in kind. The cognitive superiority of humans is then presumed to show a difference in kind that partly places us in a morally superior position. But this is a pretension the sceptic seeks prick. Human beings are actually in an inferior position in key respects and treated less well by the hand of nature than dumb beasts. Our misguided view of our own superiority expresses only a vanity that is to be exposed by the sceptic. Hume at one participates in this tradition and transforms it, signalling his alignment with the descriptive claim but rejecting the evaluative view of human nature.

2

We begin with Hume's essay ‘Of the Dignity or Meanness of Human Nature’, published in 1741.Footnote ¹ In this essay Hume discusses two sects, which, for want of a better pair of terms, I shall call the ‘optimists’ and the ‘pessimists’. The former ‘exalt our species to the skies, and represent man as a kind of human demigod, who derives his origin from heaven, and retains evident marks of his lineage and descent’, whilst the pessimists ‘insist upon the blind sides of human nature, and can discover nothing, except vanity, in which man surpasses the other animals, whom he affects so much to despise’ (EMPL 80–81). The contrast echoes what Hume identifies as the common but false view of the relation of reason and passion, where the ‘pre-eminence’ of the former owes to its ‘divine origin’, in contrast to the ‘blindness, unconstancy, and deceitfulness of the latter’ (T 2.3.3.1; SBN 413),Footnote ² and a careless thought would be that, since Hume sides with the passions, he sides with the ‘mean’ view of human nature. But matters are far more subtle, as we shall see.

‘Of the Dignity or Meanness of Human Nature’ begins to navigate these two camps by noting that in ‘forming our notions of human nature, we are apt make a comparison between men and animals, the only creatures endowed with thought that fall under our senses’ (EMPL 82). The vast differences in cognitive achievement between them and us show that there ‘comparison is favourable to mankind’ (EMPL 82). Animals are far more limited, cognitively speaking, than human beings. Those stressing human nature's meanness seek to ‘destroy this conclusion’ by a) ‘insisting only upon the weakness of human nature’ and b) by ‘forming a new and secret comparison between man and beings of the most perfect wisdom’ (EMPL 82–83). The optimist is mistaken.

It is not difficult to suppose that Hume is here thinking of Michel de Montaigne. In his longest essay, ‘An Apology for Raymond Sebond’, one finds a sustained deployment of the sceptical trope I mentioned in section I, namely a comparison of human and animal achievements. Montaigne thinks that humans are in many ways inferior and appearances suggest that we differ from beasts only in vanity. The work itself is ostensibly a defence of the eponymous Spanish ‘theologians’ claim that human reason is impotent without divine illumination, though what its real aim is a vexed issue (rather like what Bayle's real purposes are in his uses of scepticism). It is easily read as a pessimist text, however, where the comparison is unfavourable to humans. First, Hume's allusion to a ‘secret comparison between man and beings of the most perfect wisdom’ is not hard to find. The vanity of man ‘makes him equal himself to God; attribute to himself God's mode of being; pick himself out and set himself apart from the mass of other creatures’ (505). Montaigne's sceptical piety means that whilst he will officially reject the idea that we are like God, his attitude of humility is fuelled by the presumption of an infinitely greater being that the human. We are then offered pages and pages cataloguing claims, anecdotes, ranging from dancing elephants (519)Footnote ³ and a merciful tiger (535), and (occasionally) arguments to show that the human ‘is the most blighted and frail of all creatures and, moreover, the most given to pride’ (505). Animals instantiate virtues and show more fidelity to them than humans do (516).Footnote ⁴ They have instincts that constitute a greater sensitivity to features of the world than mere human reason. Thracians, reports Montaigne, use foxes to determine whether it is safe to walk on ice, showing the animal's superior detective capacity (515). We can also credit animals with reason. Since their behaviour resembles ours in key respects, then from ‘similar effects we should conclude that there are similar faculties. Consequently, we should admit that animals employ the same method and the same reasoning as ourselves when we do anything’ (514). Chrysippus's apparently reasoning dog makes an appearance in this connection.Footnote ⁵ On a road that branches into three forks, the dog, hunting his prey, sniffs the first two forks, and, having failed to pick up the scent on the first two, moves immediately down the third road. Surely Fido is reasoning disjunctively (517). In line with his Pyrrhonist temper,Footnote ⁶ Montaigne does not draw a final conclusion that it is true that human and animal cognitive capacities are of the same kind, but only that appearances do not support the presumed superiority.

Where there are apparent differences, they support a view of humans as inferior to animals. Human beings have a reflective capacity, ‘the freedom to think’, but this provides ‘little cause to boast about it, since it is the chief source of the woes which beset’ humanity (514). The freedom to think breaks us away from the rest of sentient creatures whom ‘Nature clasps….in a universal embrace; there is not one of them which she has not plainly furnished with all means necessary to the conservation of its being’ (509). At best, the freedom to think helps to show that humanity is ‘lodged down here, among the mire and shit of the world, bound and nailed to the deadest, most stagnant part of the universe…the lowest category of animate creatures’ (505).

3

A lot of water passed under many bridges between Montaigne and Hume of course, and the differences between the two emerge from a confluence of many streams. In the intervening period Descartes became notorious for his view that animals are mere machines. He thought they lacked feeling and reason. One reason for this was what Descartes saw as their lack of language. But this is not a novel claim, and indeed Montaigne devoted some of his essay to showing that animals do have a form of language.Footnote ⁷ Second, Descartes refused to see the analogies between human and animal behaviour as holding any epistemic weight.Footnote ⁸ But Descartes's views were not, to put it mildly, met with universal approbation. John Locke and Henry More resisted them and this resistance was to continue well into the eighteenth century in British thought. Descartes' refusal to take analogies between human and animal behaviours seriously was lampooned. Thus Mandeville wrote in the Fable of the Bees

Ironically, Descartes' mechanical programme for animals was taken up by others and applied to human beings. His claims in part encouraged a closer look at the structure of animal anatomy and physiology, and the discovery of deeper similarity between humans and animals pushed against Descartes' cleavage between man and beast. La Mettrie's notorious Man A Machine was published in 1747, and in the year Hume published the first two books of the Treatise Jacques de Vacuason's mechanical duck took its first bite of grain and ‘defecated’, pressing home the idea that the ‘inner workings’ and as well as ‘external workings’ of all creature could be given a mechanical gloss.Footnote ¹⁰

Hume embraces analogical argument and its application to animal cognition. His confidence in analogy is expressed in his approval of those anatomists who ‘join their observations and experiments on human bodies to those on beasts, and from the agreement of these experiments…derive an additional argument for any particular hypothesis.’ (T 2.1.12.2; SBN 325) In key places in the Treatise (and elsewhere) Hume draws explicit comparisons between humans and animals in order to emphasize commonalities. Both the Treatise and the first Enquiry include sections entitled ‘Of the reason of animals’ arguing for the conclusion (which he takes to be evident anyway) that ‘beasts are endow'd with thought and reason, as well as men’ (T 1.3.16.1; SBN 176). The long Treatise discussion of pride and humility (the indirect passions central to humean moral psychology) is capped with a section entitled ‘Of the pride and humility of animals’ (T 2.1.12), making much of various examples of animal comportment which suggest pride, such as ‘the port and gait of a swan’ and the ‘vanity and emulation of nightingales’ (T 2.1.12; SBN 326). Correlatively, Hume concludes his long discussion of love and hate in the second part of book II with ‘Of the love and hatred of animals’ (T 2.2.12). Note, first, the significant placement of these sections: each comes at the end of long discussions on reason, pride and humility and love and hatred. They function, in effect, as conclusions for those discussions that show that we differ from animals in degree and not kind.Footnote ¹¹ Second, Hume takes the continuities between humans and animals in these sections to be ‘evident’. Third, his deep-seated commitment to naturalism is expressed in his claim that respecting such continuities is a ‘touchstone’ against which any philosophical system is to be tried (T 1.3.16.2; SBN 176). Fourth, Hume takes these commonalities to express the view that in the ‘whole sensitive creation…[e]very thing is conducted by springs and principles, which are not peculiar to man, or any one species of animals’ (T 2.2.12.1; SBN 397). This amounts to a large-scale view that the mechanisms (in the non-technical sense of ‘mechanism’) underlying human thought and behaviour are no different in kind from those underlying animal behaviour. So if we take ‘human nature’ to mean the fundamental ‘spring and principles’ that guide thought and behaviour then ‘human nature’ is no different in kind from ‘animal nature’.

Hume's philosophy is explanatory in its aspirations – he seeks to explain human cognitive processes and the presence of certain distinct areas of thought – but his account is shot through and through with sceptical layers which display affinities with Montaigne.Footnote ¹² The first thing to note is that basic materials Hume uses in his explanation – impressions, ideas, association, force, vivacity-are not his own invention (though he puts his own peculiar stamp on them). All this vocabulary stretches at least as far back at Hobbes and what is significant for our concerns is such vocabulary was employed in the explanation animal behaviour, and whilst some human behaviour owes itself to these origins, our cognitive lives are not exhausted by such materials and processes. What is different about Hume is that he seeks to extend this vocabulary to explain all our mental live in its terms.

Pressed into the service of this task is a sceptical argument. Hume's positive conclusion that causal reasoning is to be identified fundamentally with associational mechanisms that drive the beasts is effected by an argument that alternative accounts of what constitutes such reasoning are faulty. Consider this claim of Leibniz's.

For Hume, ‘knowledge of some reason for perceptions’ being linked together' amounts either to awareness of necessary connections, the powers and forces that maintain the course of nature, or the grasp of some reason to think the future will resemble the past. Hume however thinks that neither is possible. His scepticism about whether we are ‘determin'd by reason’ to draw an inference from cause to effect issues in a ‘naturalism’ whereby our reasoning is ‘brute’ both in the sense that it is not caused by our capacity to grasp reasons in its favour and that the causal mechanism is the same as that which guides animals. In effect, Hume argues that the narrow sense of ‘reasoning’, which Leibniz claims as the accepted usage does not exist and so it is the very broad sense of ‘reasoning’ which Leibniz extends to animals that governs us too. So Hume deploys the sceptical trope that seeks to undermine the differences between animals and humans on reasoning used by from Sextus through Montaigne, Charron and others by arguing against particular inflationary accounts of the nature of human inference and modelling our practice on that which is taken to be operative in the beasts.

Hume of course recognizes differences between humans and beasts. He recognizes that many animals are governed by peculiar behavioural routines (like nest building) that he calls ‘original instincts’. With respect to causal inference, the differences between us and the animals resolve themselves into higher-order association relations operating on lower-order ones. Hence in ‘Of the dignity or meanness of human nature’ Hume notes the superiority of human reason over animals but doesn't imply any difference in kind, and in a long footnote to Enquiry 9 ‘Of the reason of animals’ he lists the different ways in which reasoning capacities can be extensive or limited.

So far so good. However, there are two puzzles about Hume's overall project of explaining human cognition with animal materials. One is that he doesn't discuss the ‘demonstrative reasoning’ (roughly, inferences based on conceptual relations) when talking about the reason of animals. So it seems left out of the account. The second puzzle is altogether nastier. In a notorious footnote to an essay entitled ‘Of National Characters’, Hume wrote ‘I am apt to suspect the Negroes to be naturally inferior to the whites. There scarcely ever was a civilized nation of that complexion, nor even any individual eminent either in action or speculation…In JAMAICA, indeed, they talk of one negroe as a man of parts and learning; but it is likely he is admired for slender accomplishments, like a parrot, who speaks a few words plainly.’ (EMPL 208n) Hume's claim is quite incredible, and not merely for its evident offensiveness. It is further incredible because Hume feels that the alleged ‘uniform and constant difference’ between the cognitive achievements of whites and non-whites marks ‘an original distinction between breeds of men’. But the whole tenor of Hume's naturalizing philosophy and his emphasis on continuity with animals is set against the invocation of original distinctions regarding cognitive achievements.Footnote ¹⁴ It seems to me that not only is Hume is guilty of being racially offensive but also in a way that seems to go entirely against the thrust of his philosophy.

4

Hume rejects the pessimists' conclusion of the ‘meanness’ of human nature. For one thing, he is set against the ‘secret comparison between man and beings of the most perfect wisdom’ (EMPL 82–83). The kind of despairing asceticism that he takes to infect conceptions of morality stems from this religious presupposition that Hume thinks we should dispense with. The pessimists also held that humans are those that are least adapted to the environment. Now, there is much to suggest that Hume takes animal nature to be adaptive (though he refuses to give any religious interpretation to it). All the interlocutors in the Dialogues Concerning Natural Religion agree that there is a ‘curious adapting of means to ends, throughout all nature’Footnote ¹⁵, but disagree about its implications. This adaptive strand appears in Hume's account of human nature too, and it is one of the strands upon which Norman Kemp Smith's naturalist reading alights: he suggests, for example, that the so-called natural beliefs are ‘wonderfully adapted as any of the animal instincts’.Footnote ¹⁶ There is certainly textual evidence for this reading too. In both the Enquiry Concerning Human Understanding and the Enquiry Concerning the Principles of Morals there are statements that seem to go just that way. Thus in connection with causal inference he writes of a

This is followed in the next paragraph with a reference to ‘the ordinary wisdom of nature’ (EHU 5.22; SBN 55). In the Enquiry Concerning the Principle of Morals he writes of the standard of moral sentiment ‘arising from the internal frame and constitution of animals’, each of which has its ‘peculiar nature’ (EPM Appendix 1, 21; SBN 294).Footnote ¹⁸

I don't want pursue this issue any further here. Let me instead note that one reason that Montaigne held the human condition to be wretched is that human beings are less well adapted to their environment than our dumb friends. This is something with which that other pessimist, Mandeville, agrees. In the ‘wild State of Nature, those creatures are fittest to live peaceably in great numbers [are those] that discover the least of Understanding and have the fewest Appetites to gratify’ (41). Hume echoes this view of the pre-social human condition. He writes

Both Mandeville and Hume think that the human animal can transcend its ill-suited natural state by the imposition of society. Both thinkers also offer genealogical accounts of how the human animal establishes society. These accounts differ in a number of ways, but two are relevant for our concerns. The first is that whereas Mandeville holds that human nature is constant both within and without society, Hume takes society to be transformative of that nature. The second is that Mandeville's view of human nature echoes the pessimist conception of it that Montaigne espoused and Hume rejects.

Mandeville, like Montaigne, maintains that the only way in which humans are relevantly different from animals is in terms of our susceptibility to flattering self-conceptions. Just as Montaigne wrote that the human ‘is the most blighted and frail of all creatures and, moreover, the most given to pride’ (505), Mandeville's Fable of the Bees maintains that of man ‘the most perfect of Animals’, pride is ‘inseparable from his very Essence (however cunningly soever some may learn to hide or disguise it)’ (44–5). It is this feature of human beings that is central to the establishment of society. It is through this ‘bewitching engine’ (43) that others can be manipulated by an ‘artful form of flattery’ (43) to conceive selfless actions as expressions of estimable virtue. Persons act in line with the interest of others because such flattery helps to ‘buoy them up in mortifying what was dearest to them’. Their natural disposition to be ‘led by the sensual dictates of nature’ is kept in check because they are ‘asham'd of confessing themselves to be those despicable wretches…so little remov'd from Brutes’ (44–5). So for Mandeville, like Montaigne, a false pride is what divides us from the rest of the brutes and though it helps to establish a conception of ourselves that motivates actions that are beneficial to society, it leads to a false conception that alienates us from our first-order nature. Human nature is mean all along.

Hume's solution is different. He thinks that our characters can be changed by the social environment. His clearest statement of this idea is in the essay ‘Of National Characters’ where Hume enters in the eighteenth-century dispute regarding the extent to which differing manners and characters of nations are determined by the climate, and more generally, physical environment in which human creatures are placed. Such ‘physical causes’ Hume defines as ‘qualities of air and climate…supposed to work insensibly on the temper, by altering the tone and habit of the body’ (EMPL 198), which probably picks up on the theories of John Arbuthnot and Abbé Du Bos. Hume rejects such accounts,Footnote ¹⁹ siding with those who hold that such differences are accounted for by the action of ‘moral causes’, though adding to this theory an account of cultural transmission which makes much of sympathy and imitation, the latter he takes to be a marked feature of the human animal. What is important for our concerns is how he characterizes a ‘moral cause’ and what its effects can be. Moral causes are ‘those circumstances which are fitted to work on the mind as motives or reasons and which render a particular set of manners habitual to us’ (op. cit.). Moral causes then are circumstances (and our conception of them) that we perceive as having practical salience, which in turn change our character dispositions. He adds a given moral cause ‘alter[s] even the [character] disposition that …[we] receive from the hand of nature’ (op. cit). He illustrates this by arguing that the peculiar circumstances attaching to the professions of the soldier and the priest does not merely constrain some fixed character of the persons occupying that role but changes their character.

With this in mind, let us now consider how Hume thinks humans transcend their given nature. What he seeks to explain in his account of the ‘artificial virtues’ are, roughly speaking, moral norms that govern impartial interpersonal relations, including respect for property and fidelity to promises. There are a number of senses in which these are ‘artificial’, and a key idea that Hume thinks there is no natural psychological motive linked to the relevant behaviour. It makes sense to credit humans with other-regarding behaviour – such as care for their children – that is not motivated by a grasp of its being morally required. But being honest or keeping one's promises are motivations that are ipso facto moral motivations. So Hume proposes an account of how conventions that produce and govern such behaviour emerge. His account begins with a conjecture that, prior to the institution of property, human animals inhabit family groups and this makes them sensible of the advantages of co-operative behaviour. Awareness of co-operative behaviour leads to a convention to establish property rights whose normative force derives from awareness of the fact that they serve one's own non-moral interests. At this stage, Hume's picture of co-operative behaviour produces an artifice – a convention – but an artifice that does not involve the imposition of a false view of human nature. It is simply enlightened self-interest, and so in ‘so sagacious an animal, what necessarily arises from the exertion of his intellectual faculties may justly be esteemed natural’ (EPM Appendix 3.9; SBN 307).

For many commentators this is the whole of Hume's account of the emergence of co-operative behaviour. Justice is a complex convention the normative force of which rests on long-term self-interest (long self-interest being turned to check more short-term self-interest). There are two reasons to think that this is incorrect. The first is that Hume thinks that our hardwired disposition to approve morally of certain dispositional features becomes extended, via sympathy to the convention itself. Second, there is every reason to think that the establishment of the convention feeds back into the motivational dispositions of human nature.Footnote ²⁰ Through education respect for the conventions becomes part of the motivational dispositions of human creatures (‘changeableness is essential’ to ‘human nature’ Hume writes earlier in the Treatise (T 2.1.4.3; SBN 283). Thus sentiments of justice take on such ‘firmness and solidity, that they may fall little short of those principles which are the most essential to our natures, and the most deeply radicated in our internal constitution.’ (T 3.2.2.26; SBN 501) Our motivational patterns are changed and, unlike Mandeville's view, human beings become sincerely motivated and integrated creatures. Human nature might be animal but its capacity for second-nature does not make for its meanness.