Hume’s peculiar definition of justice James A. Harris Department of Philosophy, University of St. Andrews 1. Justice as Hume usually speaks of it consists in respect for rights of property. The first example given of an act of justice in Book 3 of A Treatise of Human Nature is the repayment of a loan: the return by a borrower of the property of an owner. Sometimes, Hume observes later, it is a requirement of justice that a poor man must repay what he owes to a rich man who has no need of it. Sometimes justice requires that the labour of the poor and industrious be bestowed on the rich and dissolute. Sometimes justice requires putting into the hands of the vicious the means of harming both themselves and others. The work of justice is no more (and no less) than the enforcement of rules which articulate the distinction between possession and property. Thus far Hume’s approach to justice appears to be consonant with what he terms the ‘vulgar’ definition of justice, the definition popular among the Roman lawyers and given classic formulation by Ulpian as ‘constans et perpetua voluntas ius suum cuique tribuendi’: in Hume’s translation, ‘a constant and perpetual will of giving every one his due’ (T 526). But, Hume is quick to point out, there is contained in this definition ‘such things as right and property, independent of justice, and antecedent to it’ (T 526). And Hume takes himself to have removed the basis for rational belief in those things. He has argued that there is no means of measuring rules of justice against ‘natural’ principles of right. Rules of justice, as Hume understands them, define the distinction between possession and rightful ownership. An act is just, therefore, to the extent that it is does not violate those rules. Justice as such cannot be distinguished from justice as it is defined in a particular system of law. Although Hume himself is not as clear about the matter as one might wish, it is an implication of such an approach to justice that, conceived of as a virtue, it is a virtue of abstention and omission. The justness of an individual agent lies in her not violating the rules of justice. It lies in not stealing or otherwise appropriating what is not hers. Sometimes Hume includes in his definition of justice the obligation to respect promises and contracts. Thus he talks in Treatise 3.2.6 of ‘the three fundamental laws of nature, that of the stability of possession, of its transference by consent, and of the performance of promises’ (T 526). But it is much more common for the scope of justice to be restricted to, as Hume puts it in the second Enquiry, ‘laws for the regulation of property’ (E 194). In the essay ‘Of the Original Contract’ Hume talks in terms ‘justice, or a regard to the property of others’ as distinct from ‘fidelity or the observance of promises’. David Hume, Essays Moral, Political and Literary, ed. Eugene F. Miller (Liberty Fund, 1987), p. 480. Fairly frequently Hume speaks of ‘equity’ as if it were part of justice as he understands it, but the context in almost all cases makes it clear that he is using ‘equity’ as a synonym of ‘justice’, rather than using it in its technical sense as principle of moral insight that may be used by judges, in Kames’s words, ‘to correct and mitigate the rigour, and what even in a proper sense may be termed the injustice of common law’. Henry Home, Lord Kames, Principles of Equity, second edition (Edinburgh, 1767), p. 44. In the Treatise Hume includes ‘equity’, presumably meant in the technical sense, among the natural virtues (cf. T 578). Justice is thus for Hume remarkably limited in its domain. This has been noted and regretted by a succession of commentators, beginning with some of Hume’s most notable philosophical contemporaries, including Adam Smith and Thomas Reid. Some more recent writers, such as John Mackie, Jonathan Harrison, and David Raphael, also remark on the oddness of Hume’s definition of justice See J. L. Mackie, Hume’s Moral Theory (Routledge and Kegan Paul, 1980); p. 94; Jonathan Harrison, Hume’s Theory of Justice (Clarendon Press, 1981), pp. 28 ff.; and David Raphael, Concepts of Justice (Clarendon Press, 2001), ch. 9. – though others proceed as if there were nothing remarkable at all in Hume’s definition of justice. Recent examples include David Miller, Philosophy and Ideology in Hume’s Political Thought (Clarendon Press, 1981); Stephen Buckle, Natural Law and the Theory of Property: Grotius to Hume (Clarendon Press, 1991); Russell Hardin, David Hume: Moral and Political Theorist (Oxford University Press, 2007); Rachel Cohon, Hume’s Morality: Feeling and Fabrication (Oxford University Press, 2008). My purpose in what follows is to use the context provided by the modern natural law tradition in order to show just how peculiar Hume’s definition is, and then to offer an explanation of why he restricts the scope of justice in this way. In Hume’s day it was usual, I shall show, to include in a definition of justice a much wider set of rights than rights of property. Hume focuses on property, I shall argue, because he believes, first, that prosecuting the case for a sympathy-based theory of moral judgment, and against a special moral sense, requires an account of the historical origins of justice; and, secondly, that rules concerning property must have been, historically speaking, the first rules to have been developed by human beings wanting to live in society with each other. In conclusion I shall make a brief comparison between Hume’s theory of justice and the rather broader theory developed by Adam Smith in The Theory of Moral Sentiments. Hume and Smith share a hostility to the idea of a special sense of justice, and analyse the moral sentiments in general in terms of the operations of sympathy, but Smith does not follow Hume in pursuing a historical examination of justice in the course of his explication of the moral sentiments. I shall suggest that this is to Smith’s credit. 2. The natural jurisprudence that had flourished in Europe since the publication of Grotius’s De Iure Belli et Pacis in 1625 was an attempt to give a scientific character to the teaching of morality by showing how moral and political duties could be derived from principles of reason. See Jean Barbeyrac’s An Historical and Critical Account of the Science of Morality, and the Progress it has made in the World, from the earliest Times down to the Publication of this Work, transl. Carew, ‘prefixed’ to to Samuel Pufendorf, Of the Law of Nature and Nations, transl. Basil Kennett, 4th edition (London, 1729), pp. 79ff. As in natural philosophy, the goal in modern moral philosophy was to do away with the jargon and obfuscation of scholasticism, and, as far as possible, to replace it with the clarity bestowed by the geometrical method of proof. Particularly influential in Scotland were the versions of natural law put forward by Samuel Pufendorf in De Iure Gentium et Naturae (1672) and by Johann Gottlieb Heineccius in Elementa Iuris Naturae et Gentium (1737). A translation of Pufendorf’s abbreviation of his system of natural law, De Officio Hominis et Civis (1673), together with copious ‘supplements and observations’, was published by Glasgow’s first professor of moral philosophy, Gershom Carmichael, in 1718, with a second edition six years later. George Turnbull, until 1727 regent at Marischal College, Aberdeen, translated Heineccius ‘with supplements and a discourse’ in 1741. These systems divided the duties of human beings into three kinds: duties to God, duties to others, and duties to self. Duties to others were in turn divided into two classes: perfect duties, and imperfect duties. Carmichael explains: There are some duties which are so absolutely necessary to social life that human society itself would be unsociable in their absence, and therefore they are rightly enforced even on those who do not want to do them. But there are other duties, which pertain to the comfort or ornament of social life, and are therefore left to the discretion and honor of each individual. James Moore and Michael Silverthorne (eds.), Natural Rights on the Threshold of the Scottish Enlightenment: The Writings of Gershom Carmichael (Liberty Fund, 2002), p. 44. The duties of beneficence and humanity are imperfect in this sense, while the duties of justice are perfect. Without beneficence and humanity, social life would be uncomfortable; without justice, it would be impossible. The perfect duties of justice, according to Heineccius, ‘may be reduced to not injuring any one, and rendering to every one his due’. Johann Gottlieb Heineccius, A Methodical System of Universal Law, with Supplements and a Discourse by George Turnbull, eds. Thomas Ahnert and Peter Schröder (Liberty Fund, 2008), p. 132. An explication of the particular perfect duties to others proceeded by means of an enumeration of the various kinds of harm that could be counted as injuries. An injury in this technical sense of the word was understood to be a violation of a right, and it was common for writers in the natural law tradition to explain duties in terms of rights, perfect and imperfect. Thus Hutcheson in his Philosophiae Moralis Institutio Compendaria (1745; translated into English as A Short Introduction to Moral Philosophy in 1747) asserts that ‘the several duties of life may be naturally explained by explaining the several rights belonging to men, and the corresponding obligations, in all the several states and relations they stand in to each other’. Francis Hutcheson, Philosophiae Moralis Institutio Compendaria, with A Short Introduction to Moral Philosophy, ed. Luigi Turco (Liberty Fund, 2007), p. 127. Hutcheson begins by distinguishing between natural rights and adventitious rights. Natural rights are then divided into three classes, ‘private, publick, and common to all’. Private rights are the rights of individuals, public rights are the rights of a particular society or body of people, and common rights are the rights of ‘mankind in general as a great community’. Hutcheson, Short Introduction, p. 128. The duties of justice as ordinarily understood protect private rights, and are divided into the perfect and imperfect. There are eight perfect private rights: (i) to life, and not to be injured in body; (ii) to preserve chastity; (iii) to reputation; (iv) right over life, ie, to be able sacrifice oneself to public good; (vi) right to private judgment; (vii) right to what is common to all, to acquire adventitious rights, and to equal treatment; (viii) to marriage. It is a matter of imperfect right that ‘each one may justly claim such offices as are profitable to him, and no burden or expense to the performer. Nay every innocent person has a right to such offices of others, as are of high advantage to him, and of small burden or expence to the performers’. Hutcheson, Short Introduction, p. 131. Then Hutcheson comes to adventitious rights, that is to say, to rights which are acquired, rather than which pertain to human beings as such. Adventitious rights are either ‘real’ or ‘personal’: ‘The real terminate upon some certain definite goods: the personal terminate upon some person, not peculiarly respecting one part of his goods more than any other’. Hutcheson, Short Introduction, p. 133. The principal real adventitious right is to property, a right which has its ground in a combination of the right to self-preservation and the necessity of labour. We see, then, that on this account, the rights of property make up only one part of a complex and variegated analysis of the concept of justice. It is very much in the spirit of Hutcheson’s analysis of private rights that Reid, in the course of his critique of Hume’s account of justice, notes that A man may be injured, first, in his person, by wounding, maiming or killing him; secondly, in his family, by robbing him of his children, or any way injuring those he is bound to protect; thirdly, in his liberty, by confinement; fourthly, in his reputation; fifthly, in his goods or property; and, lastly, in the violation of contracts or engagements made with him. Thomas Reid, Essays on the Active Powers of Man (M.I.T. Press, 1969), p. 415. The first four of the rights corresponding to these injuries are natural in the sense of being ‘innate’, ‘founded on the constitution of man, and antecedent to all deeds and conventions of society’. Reid, Essays on the Active Powers, p. 417. The last is acquired, ‘not grounded upon the constitution of man, but upon his actions’. Reid, Essays on the Active Powers, p. 419. But it still natural in the sense that it may be acquired even in a state of nature, prior to and independent of positive legislation. Reid expresses puzzlement, as well he might, as to why, in the light of this well-established way of analysing the rights and duties of justice, Hume in his account of justice ‘had in his eye only two particular branches of justice’, pertaining to rights to property and the fidelity to contracts. Reid, Essays on the Active Powers, p. 418. In the rest of this paper my intention is to explain why Hume limited his analysis of justice in this way. But first it is worth noting that there is yet more that appears to be missing from Hume’s treatment of justice. There is nothing, for example, about the notion of desert that is usually taken to underwrite the justice of punishment. And Reid points out that also absent is the distributive dimension to justice. Reid argues, as Locke had done in the second of the Two Treatises of Government, that the right to the acquisition of property of one individual can be restricted by the right to subsistence of another individual. He says that ‘justice, I think, as well as charity, requires, that the necessities of those who, by the providence of God, are disabled from supplying themselves, should be supplied from what might otherwise be stored for future wants’. Reid, Essays on the Active Powers, p. 424. The reason for Hume’s omission of this aspect of justice is given in the second Enquiry, where Hume, in the manner of Hobbes, argues that where human life is threatened, the rules of justice no longer apply. On the usual view, what happens in such circumstances is that one right trumps another, and Reid sharply criticizes Hume for diverging from this view. Our question, however, is why in his account of justice Hume gives almost exclusive attention rights of property. Buckle claims that ‘Hume’s account of justice is ... thoroughly in tune with the natural jurists’ (Natural law and the Theory of Property, p. 287). This is untrue. What is true is that there are important affinities between Hume’s account of property and the natural jurists – though there are important differences as well. I briefly consider the relation between Hume and Grotius as regards property in Section 4 below. 3. In Book 3 of the Treatise, though not in the second Enquiry, Hume’s analysis of morals is organized around the distinction between ‘artificial’ and ‘natural’ virtues. What distinguishes these two groups of virtues from each other is the fact that in the case of the former group, it is implausible, so Hume argues, to regard the practices in question as the result of innate principles of action. They are not expressions of human nature as it might manifest itself in a state of nature. We human beings are fitted with self-interest and a concern for those related to us, but neither of these principles of action prompts us to the virtues of justice, promise-keeping, and allegiance. These virtues are ‘artificial’, then, in the sense of being the work of human planning and invention. In order to solve problems set for them by a combination of human need, unfriendly circumstance, and limited generosity, human beings developed conventions regarding possessions, reciprocal exchange, and the subordination of most to the will of a chief, conventions that, as we will see, Hume regards as essential enabling conditions of social life as such. Now, it might be thought that Hume focuses on rights of property to the exclusion of other aspects of justice as usually defined because such rights are easier to portray as questionable in their naturalness than, for example, the rights not to be physically harmed and imprisoned. After all, as we have seen already, rights of property were generally accepted to be adventitious, and, in Reid’s words, ‘not grounded upon the constitution of man, but upon his actions’. So perhaps it was with a view to highlighting the artificiality of justice that Hume gives such importance to rights of property. I think this is implausible. There is no reason to think that showing the artificiality of justice was one of Hume’s primary intentions. That is indeed where his argument leads him, but it is not, I think, what he started out to show. The main agenda of Book 3 of the Treatise is, rather, the demonstration that, once rationalism in ethics is dispensed with, we are not forced to accept Hutcheson’s postulation of a special moral sense, but can explain the moral sentiments in a more parsimonious and (as we would say now) naturalistic manner, in terms of the operations of sympathy. The thesis of the artificiality of justice is simply part of the case for the sympathy theory, in the sense that it explains how and why justice is a virtue without recourse to some special instinct. Several of Hume’s early critics focused on the artificiality thesis because it seemed to them to be evidence of Hume’s ‘licentious’ intention to loosen the hold of morality upon his readers. But there is no reason whatsoever to think that Hume actually had such an intention. Hume wrote the Treatise in geographical and intellectual isolation in provincial France. Could it be that it is out of pure ignorance or carelessness that he fails to consider justice in its full extent? This too is implausible, most obviously because Hume does in fact give an argument for his prioritization of property. The argument is given in the course of the explication of the ‘interested’ obligation in Treatise 3.2.2, when Hume comes to explain the danger posed to the ‘new-establish’d’ social union by the fact that ‘each person loves himself better than any other single person, and in his love to others bears the greatest affection to his relations and acquaintance’ (T 487). This fact sets the passions of each of us against the passions of (almost) every one else; and this in turn creates a serious problem for peaceful social life, when we consider the general scarcity of many of the things we desire for ourselves and our relations and friends. Hume writes: There are three different species of goods, which we are posses’d of; the internal satisfaction of our minds, the external advantages of our body, and the enjoyment of such possessions as we have acquir’d by our industry and good fortune. We are perfectly secure in the enjoyment of the first. The second may be ravish’d from us, but can be of no advantage to him who deprives us of them. The last only are both expos’d to violence of others, and may be transferr’d without suffering any loss or alteration; while at the same time, there is not a sufficient quantity of them to supply every one’s desires and necessities. As the improvement, therefore, of these goods is the chief advantage of society, so the instability of their possession, along with their scarcity, is the chief impediment. (T 487-8) Hume’s claim, put crudely, is that that there is nothing to be gained from robbing people of their peace of mind or from physically abusing them. And so there is no need – at this early stage of social life, at least – to postulate prohibitions against such things, nor to postulate a ‘natural’ sense of rights that would be violated were people to be so harmed. There is surely an implicit response to Hobbes here. It is not true, Hume is saying, that in a ‘state of nature’ man is a wolf to man. We are not so naturally fearful of each other that without the safeguards provided by the state and its law-enforcement apparatus, we would pre-emptively strike against others to prevent them from striking against us. We are not naturally disposed, out of fear, to terrorise and maim and rape and kill. The reason for Hume’s rather more optimistic conception of pre-political human nature is surely to be found in his sense of the way our passions are shaped from the beginning by life in the (extended) family. This is an aspect of Hume’s theory of nature that has been vividly brought out by Annette Baier. As she says, Hume pictures us as social from our very first days, and creatures thus socialised will not generally be aggressive and bloodthirsty: ‘What Hume’s convenors of justice aim to eliminate is not a climate of violence against persons, but a climate of incommodious insecurity of possession of material goods’. Annette Baier, A Progress of Sentiments (Harvard University Press, 1991), p. 223. It might be thought that there is plenty of empirical evidence to suggest that we are in fact rather more pointlessly violent than Hume is prepared to admit here. Don’t we, after all, have whole bodies of law arising from cases of slander and libel (and now also ‘hate speech’), and from cases of assault, rape, and murder? And are these not obviously part of justice on any plausible understanding of it? Of course they are. But Hume’s concern in 3.2.2, and indeed throughout both Book 3 of the Treatise, is with justice considered as a condition of social life – specifically, as a condition of the possibility of the peaceful co-existence of families or tribes. As we have seen, natural ties of kinship and friendship bind small groups together. Problems arise that need to be solved by conventions when those groups have to interact with each other. And these problems have, in the first instance, to do with the instability and scarcity of possessions. Hume’s concern here, that is to say, is historical. He is tracing the development of human society as such. And the importance of the fixing of conventions to do with property for the possibility of human society is what explains his focus on property to the exclusion of other aspects of justice as we ordinarily conceive of it – and as his precursors in the tradition of natural jurisprudence ordinarily conceived of it. The fact that we have no natural instinct for justice sets off Hume on a historical journey that takes him back to the origins of human sociability, and he never returns back from that journey to engage in a full consideration of justice as it is understood in the kinds of societies that he and we actually live in. But, as Duncan Forbes pointed out in his book on Hume’s politics, this does not mean that Hume’s theory could not have been developed to encompass a larger set of rights: the point is that, given his conjectural-historical objectives in 3.2.2, ‘there was no need for Hume to go any further at that juncture’. Duncan Forbes, Hume’s Philosophical Politics (Cambridge University Press, 1975), p. 89. It is thus perfectly consonant with Hume’s approach to justice in the Treatise that he should in other works appear to be working with a much more commonsensical notion of what justice amounts to. Annette Baier has drawn attention to this more ordinary Humean conceptions of justice in recent work on The History of England. Baier points, for example, to Hume’s praise of various acts of ‘justice’ on the part of James I. James is referred to as having introduced ‘justice’ to Ireland, where that meant, in Baier’s words, ‘a fair return for one’s labor, as well as fixed tenure of one’s land’. Annette Baier, ‘Hume’s Enlargement of his Concept of Justice’, forthcoming in The Cautious Jealous Virtue: Hume on Justice. In another ‘laudable act of justice’ James insisted on the proper sentence (of death) for a nobleman convicted of murder. Baier remarks that ‘[o]nce wrongful death counts as an injustice ... English history becomes a long string of injustices, since many who ended on the scaffold or the gibbet can be seen as wrongly convicted, and of course many died at the hands of ordinary murderers’. Baier, ‘Hume’s Enlargement of his Concept of Justice’. James I was as guilty of injustice in this sense as any of his predecessors and sucessors. For reasons of geopolitical expediency, after all, James executed Sir Walter Raleigh. Hume comments: No measure of James’s reign was attended with more public dissatisfaction than the punishment of Sir Walter Raleigh. To execute a sentence, which had been originally so hard, which had been so long suspended, and which seemed to have been tacitly pardoned by conferring on him a new trust and commission, was deemed an instance of cruelty and injustice. Hume, The History of England (Liberty Fund, 1983), vol. 5, p. 79. And we move on in the last two volumes of the History through the unjust executions of the Earl of Strafford and Archbishop Laud to the execution of Charles I. As Baier notes, Hume refrains from condemning Charles's execution as unjust, but his long and careful discussion of the case of Charles Stuart is coloured by a scrupulous – excessively and wilfully scrupulous, Hume’s Whig critics would say – consideration of the demands of justice. For present purposes, all that requires to be said about Hume’s use of the language of justice in the History is that the society he is describing is at some unspecified (and unspecifiable) distance from the primal stages being analysed in Treatise 3.2.2. In developed societies, Hume can allow as components of justice other rights than rights of property. They are absent from the Treatise discussion because, to repeat, Hume’s concern there is exclusively with the necessary conditions of social life as such. 4. I have claimed that Hume’s main concern in Book 3 of the Treatise taken as a whole is with the demonstration that there is no explanatory need for a special moral sense as postulated by Hutcheson, and a vindication of an alternative, more ‘naturalistic’ hypothesis, in the form of the sympathy theory of moral judgment. By the time he comes to rewrite his moral philosophy in An Enquiry concerning the Principles of Morals, it has plainly come to seem to Hume that the case against Hutcheson did not require the kind of elaborate conjectural history that is such a prominent feature of Treatise 3.2. Concomitantly, it no longer mattered that the virtues be divided into the ‘natural’ and ‘artificial’. The agenda remains that of undermining the case for a special moral sense, but the means to that end are different. Hume describes his project in Section III, on justice, as that of showing ‘[t]hat public utility is the sole origin of justice, and that reflection on the beneficial consequences of this virtue are the sole foundation of its merit’ (E 183). And he moves towards a conclusion of his discussion of justice in Section III as follows: As justice evidently tends to promote public utility and to support civil society, the sentiment of justice is either derived from our reflecting on that tendency, or like hunger, thirst, and other appetites, resentment, love of life, attachment to offspring, and other passions, arises from a simple original instinct in the human breast, which nature has implanted for like salutary purposes. If the latter be the case, it follows, that property, which is the object of justice, is also distinguished by a simple original instinct, and is not ascertained by any argument or reflection. But who is there that ever heard of such an instinct? Or is this a subject in which new discoveries can be made? We may as well expect to discover, in the body, new senses, which had before escaped the observation of all mankind. (E 201) The supposed need of a ‘simple original instinct’ is now obviated by the possibility of explaining the origins of justice more simply, in terms of reflection upon its obvious utility. Note, though, that justice is still defined as it was in the Treatise, in terms of rules of property. This, I believe, is a by-product of Hume’s continuing tendency to think of the ‘origins’ of justice in historical terms. When he says that ‘public utility is the sole origin of justice’, he must surely be talking about a historical origin, the origin of justice in the development of human society, not the origin of the concept of justice in the experience and sentiments of each and every human individual. And when one considers the history of human society, it remains plausible to think of conventions regarding property as having come before other elements of justice. It is sometimes said that Hume toned down the radicalism of his moral philosophy for more general public consumption in the second Enquiry, and that is why, for example, there is no insistence on the artificiality of certain of the virtues. This is unconvincing for a number of reasons, one of which is the fact that Hume in the second Enquiry draws attention to a consequence of his understanding to justice that he did not mention in the Treatise and that he must have known would be shocking to his contemporaries. It is unsettling even now. In the course of a series of counterfactual considerations designed to make the case for his account of the origins of justice, Hume argues that there are obligations of justice only between those who are sufficiently equal in strength to be able make each other ‘feel the effects of their resentment’ (E 190). Where one party is ‘incapable of all resistance’ in the face of the superior strength of another, the former are reliant for gentle treatment solely on the checks provided by compassion and kindness. Hume’s point here is that it makes no sense to imagine the strong making conventions and compacts regarding property with the weak. The strong have no need to do so. They are not threatened by the weak, and can take whatever they want, when they want. Conventions of justice will not naturally develop between two such unequal parties because there is no need for them. ‘This is plainly the situation of men, with regard to animals’, Hume says; it has been thought to be the situation of ‘civilized Europeans’ with regard to ‘barbarous Indians’; and ‘[i]n many nations’ it is the situation of women with regard to men (E 191). Nothing could make it plainer that Hume rejects absolutely the idea of what we now call ‘human rights’, rights to equal and fair treatment possessed by human beings as such. Obviously, he rejects natural ‘animal rights’ as well. It is conceivable, from the Humean point of view, that rights might come to be accorded to animals, Indians, and women through positive legislation. What Hume will have no truck with is the notion that such rights can be thought of as natural as opposed to conventional. And even where the right kinds of convention do exist, they will be rather late historical developments. Reid comments that ‘If Mr. Hume had not owned this sentiment as a consequence of his Theory of Morals, I should have thought it very uncharitable to impute it to him’. Reid, Essays on the Active Powers, p. 427. There can be no better evidence of the falsity of a moral theory, Reid says, than the fact that it subverts the rules of practical morality. Hume would deny that he subverts practical morality, since he is quick to stress the work that benevolence does in the protection of the weak. But it is surely true that he is here working with a conception of justice that is at odds with ordinary moral conceptions. The reason why Hume denies that (‘civilized’) men are under duties of justice with respect to animals, non-Europeans, and women is that he continues to think of justice in terms of the development of the primary and enabling conventions of human social life. In fact, his hostility to the idea of a natural, innate, or instinctual regard for justice pushes him close to embracing the Hobbesian position that there can be no injustice where there has been no prior covenant. It is therefore not surprising that Hume’s first critics accused him of Hobbism. An anonymous French reviewer called Book 3 of the Treatise ‘le Système de Hobbes habillé dans un goût nouveau’. Bibliothèque Raisonnée des Ouvrages des Savans de l’Europe 26 (1741), p. 427. The compiler of the charges against Hume that helped to bar him from the Edinburgh chair in moral philosophy accused Hume of going even further than Hobbes had: after all, ‘Mr. Hobbs, who was at Pains to shake loose all other natural Obligations, yet found it necessary to leave, or pretended to leave, the Obligation of Promises or Pactions; but our Author strikes a bolder Stroke...’. A Letter from a Gentleman to his Friend in Edinburgh, in David Hume, A Treatise of Human Nature, ed. David Fate Norton and Mary Norton (Clarendon Press, 2007), p. 424. As regards the argument from the second Enquiry just described, Reid suggests that Hume is merely repeating Hobbes’s doctrine ‘that right has its origin from power’, and remarks that here, despite his official disavowal of the selfish hypothesis, Hume ‘founds justice solely upon utility to ourselves’: ‘Mr Hobbes could have said no more’. Reid, Essays on the Active Powers, p. 428. In Appendix 3 of the second Enquiry Hume tries to distance himself from Hobbes by explicitly rejecting the view of those who, like Hobbes, had portrayed justice as arising from a promise – while at the same reaffirming the Treatise view that it arises from unspoken conventions adopted by self-interested individuals (cf. E 306). Hume is being very careful here. He would have known that most of Hobbes’s critics had fixed in particular on the claim that the duties of justice might be resolved into the terms of explicit promises. Pufendorf, for example, had claimed that the opinion that justice is ‘nothing else but a keeping of Faith, and fulfilling of Contracts’ was something that Hobbes ‘borrow’d from Epicurus’. Pufendorf, Of the Law of Nature and Nations, I.vii.13, p. 84. ‘Indeed,’ Pufendorf had continued, ‘so far is it from being rational to resolve all of Justice into Performance of Covenants, that, on the contrary, before we can know whether any Covenant is to be perform’d, we ought to be certain that it was entred upon, either by the Command, or with the Permission of the Laws of Nature; that is, that it was justly made’. Pufendorf, Of the Law of Nature and Nations, I.vii.13, p. 85. Hume is attempting to navigate a way between Hobbes and Pufendorf: neither covenants nor divine commands are the basis of justice as he understands it. In a footnote to Appendix 3 of the second Enquiry, Hume claims that ‘This theory concerning the origin of property, and consequently of justice, is, in the main, the same with that hinted at and adopted by Grotius’ (E 307 fn). He then quotes (in Latin) the following passage from The Rights of War and Peace: 4. From hence we learn, upon what Account Men departed from the antient Community; first of moveable, and then of immoveable Things: Namely, because Men being no longer contented with what the Earth produced of itself for their Nourishment; being no longer willing to dwell in Caves, to go naked, or covered only with the Barks of Trees, or the Skins of Wild Beasts, wanted to live in a more commodious and more agreeable Manner; to which End Labour and Industry was necessary, which some employed for one Thing, and others for another. And there was no Possibility then of using Things in common; first, by Reason of the Distance of Places where each was settled; and afterwards because of the Defect of Equity and Love, whereby a just Equality would not have been observed, either in their Labour, or in the Consumption of their Fruits and Resources. 5. Thus also we see what was the Original of Property, which was derived not from a mere internal Act of the Mind, since one could not possibly guess what others designed to appropriate to themselves, that he might abstain from it; and besides, several might have had a Mind to the same Thing, at the same Time; but it resulted from a certain Compact and Agreement, either expressly, as by a Division; or else tacitly, as by a Seizure. Hugo Grotius, The Rights of War and Peace, ed. Richard Tuck (Liberty Fund, 2005), II.ii.4-5, pp. 306-7. In his notes to Grotius’s text, Jean Barbeyrac claims that by ‘a certain Compact and Agreement’ Grotius meant a contract, and that is probably right. However, for Hume, I think, the really important aspect of this passage was Grotius’s attack on the idea that property might have its origin in ‘a mere internal Act of Mind’. That is, it is the negative dimension of the argument that struck a chord with Hume, along with the way Grotius looks to the material circumstances of early humanity, rather than to a God-given moral sense, for the origins of property, and, as Hume says, ‘consequently of justice’. This was the decisive move made by Grotius – as far as both Hobbes and Hume were concerned. It followed that a historical story had to be told of how humankind moved from a primal pre-political state of nature to a state in which compacts and agreements were possible. Hume’s way of telling that story, even though purely conjectural, is intended to be more historically (and philosophically) plausible than Hobbes’s. Civil society does not come into existence all at once, with the universal surrender of rights to a sovereign power. The process is a gradual one, and its first stage is the establishment of conventions regarding property. 5. My suggestion is that there is a connection to be drawn between Hume’s definition of justice in terms of rights of property and his critique of the idea of special moral sense. Having rejected a Hutchesonian moral sense as the faculty of moral judgment, Hume finds himself faced with a puzzle as to the origins of justice, and he solves that puzzle with a conjectural history of human sociability which gives pride of place to property among the conditions of the possibility of social life for human beings. Hume defines justice as he does because of the historical character of his analysis. There was no need for him to go any further: no need, that is, to explore the history of the other rights that were generally brought under the heading of justice by his contemporaries. Once the origins of justice, in the form of respect for property rights, had been uncovered, sympathy could be put to work to explain the moral dimension of respect for property. And once that had been done, Hume thought he had completed the task he had set himself. At this point, a comparison with Adam Smith’s treatment of justice in The Theory of Moral Sentiments is interesting. Smith follows Hume in rejecting Hutcheson’s notion of a special moral sense, and he follows Hume also in believing that the faculty of sympathy provides a better, more parsimonious, account of the origin of the moral sentiments. But, even though Smith was keenly interested in the historical origins of property rights, he does not follow Hume down the path of conjectural history in The Theory of Moral Sentiments. And justice as it is analysed in that book is – I suggest, as a result – considerably broader in scope than justice as it is treated of in the Treatise and second Enquiry. Smith gives detailed attention to the justice of punishment and its basis in sympathy with proper resentment. Smith is explicit about justice being a negative virtue: he says that ‘We may often fulfil all the rules of justice by sitting still and doing nothing’. Smith, The Theory of Moral Sentiments, ed. D. D. Raphael and A. L. Macfie (Liberty Fund, 1982), II.ii.1.10, p. 82. But he has a more complete description to give than does Hume of the injuries that the unjust may do: The most sacred laws of justice, ..., those whose violation seems to call loudest for vengeance and punishment, are the laws which guard the life and person of our neighbour; the next are those which guard his property and possessions; and last of all come those which guard what are called his personal rights, or what is due to him from the promises of others. Smith, Theory of Moral Sentiments, II.ii.2.2, p. 84. As we have seen, there is nothing in the Treatise and second Enquiry about the laws which guard the life and person of our neighbour. Hume’s historical orientation distracts him from a proper examination of this very considerable aspect of justice as it is ordinarily understood. Smith sees rather more clearly than Hume the difference between two questions regarding the origins of justice. One of these questions concerns the historical origins of justice, and the other concerns how it is that human individuals at later times come to think in terms of the just and the unjust. Both are questions for one who rejects a special moral sense able to provide us with ideas of justice. But they are different questions nonetheless. For it is highly implausible to imagine that the conjectural history of property that Hume gives in Treatise 3.2.2 describes a process that each of us must go through in order to be able to have a concept of justice. We are given our ideas of justice by sympathy with the consequences of the conventions of justice – or, more precisely, by sympathy with those harmed as a result of the violation of those conventions. But if sympathy attuned to already existing conventions is the origin of our ideas of justice, and if we as individuals do not beforehand have to go through the process of establishing conventions with other individuals, then it is not clear why the idea of justice that is the object of moral philosophy is restricted in the way that Hume’s idea of justice is restricted. Moral philosophy concerns itself, or should concern itself, with morality as we understand it now, rather than with morality’s historical origins. In The Theory of Moral Sentiments Smith addresses the question of how we now think in terms of the just and the unjust. Smith saves the other question, concerning the historical origins of justice, for a different kind of philosophical enquiry, that pursued on the basis of a stadial model of the human development in his Lectures on Jurisprudence. And for the most part, the philosophers of the Scottish Enlightenment followed Smith, and respected the difference between the history of justice and the analysis of the moral faculty. Some, like Reid, concentrated on the latter and ignored the former. Others, like John Millar, concentrated on the former and ignored the latter. Still others, like Lord Kames and Adam Ferguson, sought to do both – but not at the same time. Hume asked questions that the later Scottish writers found they needed to answer. In his own writings on justice, however, Hume tried to answer too many questions at once. PAGE 15 DRAFT – PLEASE DO NOT CITE WITHOUT PERMISSION