Philosophers traditionally use the term "virtue" to designate morally good character traits such as benevolence, charity, honesty, wisdom, and honor. Although ethicists, past and present, do not commonly offer a definitive list of virtues, the number of virtues discussed is often short and their moral significance is clear. Aristotle, for example, focuses on 12 principal virtues. Hume's analysis of the virtues departs from this tradition both in terms of the quantity of virtues he discusses and their obvious moral significance. A conservative estimate of the various virtues Hume refers to in his moral writings would put the number at around seventy, with the more untraditional ones including wit, good manners, and dialog. Philippa Foot attacks Hume for making nonsense of the concept of virtue by construing it so widely:
 

...one does not find in Hume an account of the difference between skills or talents and virtues and he even says that there is no reason to consider virtue as something distinct. I suppose it is partly due to Hume's influence that this important topic, which was splendidly treated by Aristotle and Aquinas, is hardly discussed by modern moral philosophers.⁽¹⁾

Foot's critique is an echo of a more heated attack made by 18th century critics who also believed that Hume's notion of virtue was too broad. Hume was aware that his broad understanding of virtue was controversial, and we know from his letters that he read at least one of these critiques. He offered several defenses for his position, perhaps in response to some of these critics. Although Hume sees this as an issue worthy of defending, contemporary Hume scholarship has neither discussed the nature of these18th century criticisms nor given adequate attention to the success or failure of Hume's defense. After presenting the neglected attacks of his contemporaries, and Hume's response, I argue that a problem remains. By failing to distinguish between degrees of virtue, Hume also fails to distinguish between degrees of vice. But, some vices such as malevolence clearly deserve punishment whereas other alleged vices such as uncleanliness clearly do not. Thus, for adequate retribution, a distinction is needed between important and less important virtues and vices. I conclude that Hume could have used his own account of instinctive vengeance as a natural indicator distinguishing between important and unimportant vices. I begin with a discussion of Hume's position.

Hume's first account of the scope of the virtues appears in Book three, Part three of the Treatise, published in 1740.⁽²⁾ His main criterion in determining the virtue of an agent's character trait is whether the action it produces has consequences such that a spectator has sympathetic feelings of moral pleasure⁽³⁾ in response (T 574-575).

Hume recognized at the time that his test for virtue was broader than that proposed by traditional ethical systems which distinguished between natural abilities (such as wit) and moral virtues (such as benevolence). In Section four, titled "Of Natural Abilities" he argues that the distinction between natural abilities and moral virtues is unfounded since both groups of mental qualities elicit the same sympathetic moral feelings in the spectator (T 606-614). Anticipating objections, Hume suggests several possible points of distinction between natural abilities and moral virtues, but then rejects them all.

First he suggests that the precise feelings elicited in the spectator designate two separate classes. Hume's reply is that the precise feelings elicited by each token virtue (e.g. benevolence, charity, justice) are also distinct in minute ways. Nevertheless, we still classify all such elicited responses as feelings of moral approval. Second, he suggests that natural abilities are involuntary whereas moral virtues are the result of free will. Hume's reply is that many involuntary attributes, such as fortitude, have classically been catalogued as virtues. Also, there is no conceivable reason why an involuntary quality of an agent may not produce sympathetic pleasure in the spectator. Third, Hume suggests that habits relevant to moral virtues can be altered through reward and punishment, whereas habits relevant to natural abilities are more fixed and cannot be altered by these means. Hume's reply is that if we are not biased by such moral systems, then this distinction would not even arise.

Given the lack of success of the Treatise, it was several decades before a critic responded to Hume's account of the virtues as it appears in Book three of the Treatise. However, Hume's moral theory appeared with much greater success in his Enquiry Concerning the Principles of Morals (1751),⁽⁴⁾ and in this version, the broad nature of the virtues is presented with even greater emphasis. At the very outset of the Enquiry Hume establishes the following as his method of investigation:
 

... we shall consider every attribute of the mind, which renders a man an object either of esteem and affection, or of hatred and contempt; every habit or sentiment or faculty, which, if ascribed to any person, implies either praise or blame, and may enter into any panegyric or satire of his character and manners. (E 174)

Once having compiled his list of virtues, he argues in the conclusion to the Enquiry that they all fall into at least one of four categories according to the consequences that they produce. Hume writes, "personal merit consists altogether in the possession of mental qualities, useful or agreeable to the person himself or to others" (E 268; cf. T 590, 604, 611). Accordingly, any mental attribute of an agent that brings about such utility or agreeability may be deemed a virtue. Indeed, in the Enquiry this becomes the definitive litmus test for Hume, and he boldly applies it to as many mental attributes of an agent that he can. He argues that the only reason why we do not all recognize this is because more elaborate ethical systems have perverted our natural understanding.

The first attack on Hume's broad understanding of the virtues probably came in a private correspondence from Francis Hutcheson in 1739, from whom Hume solicited comments on Book III of the Treatise. Although Hutcheson's letter is no longer extant, Hume's reply is:
 

Whether natural Abilitys be Virtues is a Dispute of Words. I think I follow the common Use of Language. Virtus signify'd chiefly Courage among the Romans. I was just now reading this Character of Alexander the 6th in Guicciardin. In Alessandro Sesto fu solertia & sagacita singulare: consiglio eccellente, efficacia a persuadere maravigliosa, & a tutte le facende gravi, sollicitudine & destrezza incredible. Ma erano queste virtù avanzate di grande intervallo da vitii &c. Were Benevolence the only Virtue no Characters cou'd be mixt, but wou'd depend entirely on their Degrees of Benevolence. Upon the whole, I desire to take my Catalogue of Virtues from Cicero's Offices, not from the Whole Duty of Man. I had, indeed, the former Book in my Eye in all my Reasonings. [Hume to Francis Hutcheson, Sept. 17, 1739]

Apparently, drawing from Butler's and his own moral theory, Hutcheson maintained that benevolence is the source of all virtue. Since many of the attributes that Hume listed as virtues do not stem from benevolence, then they would be better classified as "natural abilities." In response, Hume argues that calling them "virtues" follows a common use of language, especially when considering the how classical writers such as Cicero used the term.

Almost immediately after the publication of the Enquiry, critics attacked the broad understanding of the virtues which it contained.⁽⁵⁾ The first of these appeared in the anonymous pamphlet Some late opinions concerning the foundations of morality examined (1753).⁽⁶⁾ The author begins noting that,
 

[Hume's] ...notion is, That whatever, in character or conduct, is approved as useful, is virtue; and thus, by reducing morality into the same class in which we place some very trifling qualities, he destroys moral distinctions altogether.

The author continues arguing that Hume himself distinguishes between a spectator's pleasure resulting from an agent's character traits and that resulting from inanimate objects. The former, for Hume, is a stronger feeling than the latter. The author criticizes that, by the same reasoning, we can distinguish between intellectual abilities, such as courage, and "qualities of an inferior nature" such as "wit; decency; and even cleanliness." The brief review of this pamphlet in the Monthly Review contains a concise summary of this part of the author's attack on Hume. This indicates both that the dispute is intrinsically interesting and that, from early on, this problem in Hume's theory was well publicized.⁽⁷⁾

In the same year, James Balfour made a similar attack in his anonymously published A delineation of the nature and obligation of morality (1753).⁽⁸⁾ In Section four of that work Balfour argues sarcastically,
 

[Hume] ...has paved the way to enrich mankind with the possession of a thousand virtues that were never once dremt of before. For every minister of pleasure, even of the lowest kind, may put in his claim for virtue, and rise in his demands in proportion as he can increase our sensual gratifications. Strange morality indeed! But it is not confined to those functions common to us with the brutes; it even extends itself to inanimate things; so that the beauty of a flower, and the useful qualities of a plant, may assume the name of moral virtue.

Balfour has misunderstood Hume by accusing him of extending virtue to inanimate objects. Clearly, for Hume, the term "virtue" can only apply to the mental quality of an agent. Nevertheless, Balfour's concern is that Hume "indeed crouds into his assemblage of virtues, every endowment of mind, every quality of the body, and every external ornament, and advantage of fortune." Balfour believes that Hume's key mistake is "confounding these qualities themselves, with the proper use we make of them," for, Balfour notes, a useful ship is relative to the pilot. Consequently, Hume is misguided in his appeal to the Greek and Roman moralists in their broad use of the term "virtue" since the ancients believed that misapplied virtues would be vices. Again, the review of this pamphlet in the Monthly Review contained an extended excerpt from this portion of Balfour's essay.⁽⁹⁾ Hume read Balfour's text and wrote him a warm letter appreciation. Hume states in the letter that, "If I have occasion to give a new edition of the work, which you have honoured with [an] answer, I shall make great advantage of your remarks, and hope to obviate some of your criticisms."

In 1755, John Leland continued the assault in Volume two of his A view of the principal deistical writers.⁽¹⁰⁾ After citing Hume's wide ranging list of virtues, Leland argues,
 

I cannot see what valuable end it can answer in a treatise of morals to extend the notion of virtue so far. It is of high importance to mankind rightly to distinguish things that are morally good and excellent from those which are not so; and great care should be taken, that both our ideas of these things, and the expressions designed to signify them, should be kept distinct.

Leland takes Hume to task not only for the unnecessary attributes he includes under the heading of virtue, but also for leaving out key Christian virtues, such as humility. Excerpts from Leland's discussion appeared in both the Monthly Review, and the newly formed Critical Review.⁽¹¹⁾

James Beattie presented the most detailed attack on Hume's broad view of the virtues in An essay on the nature and immutability of truth (1770).⁽¹²⁾ Unlike the above attacks, which focus on Hume's theory as it appears in the second Enquiry, Beattie examines Hume's account in the Treatise. For Beattie, Hume errs by founding his position "not on fact, but on theory, and [is] supported by ambiguous words and inaccurate experience." Beattie reconstructs Hume's argument as follows:
 

Justice, humanity, generosity, excite approbation; -- a handsome face excites approbation; -- great genius excites approbation: the effect or sentiment produced is the same in each instance: the object, or cause, must therefore, in each instance, be of the same kind.

Beattie then counters that,
 

I am conscious, that my approbation of a fine face is different in kind from my approbation of great genius; and that both are extremely different from my approbation of justice, humanity, and generosity: if I call these three different kinds of approbation by the same general name, I use that name in three different significations. Therefore moral, intellectual, and corporeal virtues, are not of the same, but of different kinds.

The wide construal of the virtues in Hume's moral theory was attacked not only in philosophical works such as the above, but also in the late appearing biographical sketches of Hume. For example, a 1773 essay in The Weekly Magazine or Edinburgh Amusement contains the following critique:
 

Our author's notion and definition of virtue is very singular. He says, that it is the possession of such qualities as are useful or agreeable to ourselves or others.... This picture of virtue, which our author has drawn, is an unnatural groupe of a strange variety of features, very inconsistent, and badly proportioned.... He makes virtue intirely dependent upon the capricious humours of mankind, and even to take its form the prevailing vices of the age.⁽¹³⁾

A similar attack appeared in the same publication following Hume's death in 1777.⁽¹⁴⁾

Although all of the above attacks demonstrate a dissatisfaction with Hume's failure to distinguish between natural abilities and moral virtues, none offers clear suggestions for how a distinction should be made: they merely note that the spectator's response is different for natural abilities versus moral virtues. Hume, of course, would disagree. In the absence of a more objective criterion than the differing nuances of a spectator's sentiments, Hume seems justified in holding his ground.

Perhaps as a response to some of the above attacks ­ especially Balfour's -- in 1764 Hume added a final appendix to his Enquiry titled "Of Some Verbal Disputes." In this appendix Hume offers two lines of defense for construing the virtues so widely. First, as in the Treatise, he again argues that one cannot find a natural basis of distinction between moral virtues and natural abilities. He suggests possible points of distinction, such as those between voluntary and involuntary, social and private, moral and intellectual, or heart and head, but argues that in each of these cases such distinctions would misclassify important moral virtues (E 313). He concludes that, in the absence of any clear distinction between moral virtues and natural abilities, moral virtues must include all useful and agreeable qualities that produce moral pleasure in the spectator.

His second argument (paralleling his earlier reply to Hutcheson) is based on our broad usage of ethical language, particularly the moral injunctions that are used in reference to moral virtues as well as natural abilities. According to Hume, theologians who employ their theological methods in ethics, advance the restricted and thereby distorted notion of virtue, which excludes natural abilities. This theological method "bends every branch of knowledge to its own purpose, without much regard to the phenomena of nature, or the unbiased sentiments of the mind, hence reasoning, and even language, have been warped from their natural course" (E 322, cf. T 609). This in some ways anticipates C.L. Stevenson's argument in this century that ethical discourse is somewhat vague, and that we must preserve this vagueness in our analysis of ethical terms.⁽¹⁵⁾ However, it would be misleading to see Hume as suggesting that ethical theory should be developed out of language analysis alone. Instead, Hume's point is that our broad use of ethical language confirms his earlier argument that there is no natural distinction between moral virtues and natural abilities.

Although 18^th century critics found Hume's wide construal of the virtues so controversial, contemporary Hume scholars typically do not address the issue. Two exceptions are Ardal and Mackie. Ardal, though, is happy to accept Hume's final word on the issue:
 

... [Hume] is perhaps right in thinking that no clear-cut criterion to distinguish the specifically moral is in use by ordinary people. We do not, for example, enquire whether a man's courage is native to him before we call it a virtue.⁽¹⁶⁾

Mackie, by contrast, does suggest a possible means of distinction. Following Aristotle's contention that virtues are "dispositions of choice" Mackie proposes that virtues are dispositions which we choose to cultivate, often for reasons of social pressure. Abilities, by contrast, "are not similarly responsive to social pressure or cultivation."⁽¹⁷⁾ However, considering that the problematic natural abilities in question include those of wit, cleanliness, good manners, and dialog, Mackie's suggestion does not make sense; these are indeed dispositions which we may cultivate by choice because of social pressure or many other motives.

In spite of Hume's apparent success at deflecting the charges of critics, one might still contend that the two groups of character traits do differ significantly in kind. The problem is less evident when grouping together benevolence and, say, cleanliness as virtues; it becomes more pronounced, however, when grouping malevolence and uncleanliness as vices. A spectator's pleasurable response to an agent's benevolent act and clean appearance may not differ significantly (cf. T 608), but we would expect the spectator's painful response to differ significantly regarding a malevolent act in contrast to an unclean appearance. This raises a further problem about punishment, for without a natural basis to distinguish vices such as malevolence from personal defects such as uncleanliness, punishing one and not the other would be at worst capricious and at best artificially imposed. Thus, by blurring the distinction between vices such as malevolence and uncleanliness, Hume creates problems for a natural theory of retribution.

Interestingly, within Hume's writings we can find a natural basis for punishment insofar as Hume groups resentment and the desire to punish one's enemies as instinctive desires (T 417, 439, E 201). This inclination to punish is specific to an agent's provoking action, and not merely a reaction we have against enemies in general:
 

When I receive any injury from another, I often feel a violent passion of resentment, which makes me desire his evil and punishment, independent of all considerations of pleasure and advantage to myself. (T 418)

From this instinctive desire a natural distinction between moral vices and personal defects can be established: an agent's character trait is a moral vice only if, (a) it produces feelings of moral pain in the spectator and, (b) incites the spectator's desire for punishment. By contrast, a character trait would be a personal defect only when (a) obtains but (b) does not. It would then be a small step to establish a natural distinction between moral virtues and natural abilities. A moral virtue is a character trait such that its corresponding vice is a "moral vice" (as defined above). By contrast, a natural ability is a motive such that its corresponding defect is a "personal defect" (as defined above).

Thus, this possibility of distinguishing moral virtues from natural abilities was open to Hume. Had he adopted it, his moral theory could have stood much as it is now, with only a few classificatory alterations. Given the fact that he held firm on his position for at least 36 years in the face of numerous criticisms, we may speculate about why this was so important to him. First, in the Natural History of Religion, Hume contends it is very easy for us to be virtuous and indeed requires little effort: "all virtue, when men are reconciled to it by ever so little practice, is agreeable."⁽¹⁸⁾ From that perspective, all of the virtues that Hume lists fall into a single group with nothing to differentiate them with respect to their difficulty. Second, and more importantly, in the conclusion to the Enquiry, Hume insists that the sole criteria of moral virtue, or "personal merit," consist of utility and agreeability:
 

It may justly appear surprising, that any man, in so late an age, should find it requisite to prove, by elaborate reasoning, that personal merit consists altogether in the possession of mental qualities, useful or agreeable to the person himself, or to others.

If these are the only criteria of virtue, then any character trait that has these effects must be deemed virtuous. To the extent that these are the only criteria of virtue, then there is no basis for making further distinctions between "natural abilities" and "virtues proper." Perhaps it is Hume's drive for a simple criterion of moral virtue that inclined him to hold his ground on the broad nature of the virtues.

1. Philippa Foot, "Hume on Moral Judgment," in Virtues and Vices, (Berkeley: University of California Press, 1978), p. 75.

2. David Hume, A Treatise of Human Nature, ed. L.A. Selby-Bigge, second edition, (Oxford: Clarendon Press, 1978), hereafter abbreviated T.

3. For Hume, a spectator's sympathetic feelings of pleasure is an approving sentiment of moral pleasure just in case it (a) produces further feelings of love, hate, pride or humility in the spectator (T 473), (b) has an agent's mental quality as its object (T 472), and (c) arises impartially, without consideration of the agent's relation to the spectator (E 272, cf. T 472).

4. David Hume, Enquiries concerning Human Understanding and concerning the Principles of Morals, ed. L.A. Selby-Bigge, third edition, (Oxford: Clarendon Press, 1975), hereafter abbreviated E.

5. The first printed response to Hume's moral Enquiry was William Rose's review of that work in the Monthly Review, 1752, Vol. 6, pp. 1-19. Rose's review consists mainly of extended excerpts and impartial summaries. Although he does not criticize Hume for his wide construal of the virtues, Rose's choice of excerpts and summaries clearly emphasizes Hume's broad understanding of the virtues.

6. Some late opinions concerning the foundations of morality examined. In a letter to a friend. London, R. Dodsley, 1753, pp. 24-31.

7. Review of Some late opinions... in the Monthly Review, 1753, Vol. 8, p. 400. The reviewer's summary is as follows: "After this he proceeds to consider that idea of approbation, which mr. Hume makes to include the whole of the moral feeling, and upon which he founds the distinction betwixt virtue and vice; and here he takes a short view of that variety of different objects and qualities, which the author of the Enquiry ranks under the same class of moral approbation; and endeavours to shew, that the sentiments they produce in us differ as widely as the approbation of inanimate from that of rational objects."

8. [James Balfour], A delineation of the nature and obligation of morality with reflexions upon Mr. Hume's book, intitled, An inquiry concerning the principles of morals. Edinburgh, Hamilton, Balfour, and Neill, 1753, pp. 105-109, 116-123.

9. Review of James Balfour's A delineation... in the Monthly Review, 1753, Vol. 8, pp. 364-372.

10. John Leland, A view of the principal deistical writers of the last and present century, 1755, Vol. II, London, B. Dodd. (Letter 21 in Volume I of the 1757 edition.)

11. Review of John Leland's A view of the principal deistical writers..., in the Monthly Review, 1757, Vol. 14, pp. 465-477, and in the Critical Review, 1756, Vol. 1, pp. 193-208.

12. James Beattie, Essay on the nature and immutability of truth in opposition to sophistry and scepticism. 1770, Edinburgh, A. Kincaid and J. Bell, pp. 421-448.

13. "Character of the Works of David Hume Esq," in The Weekly Magazine or Edinburgh Amusement, 1773, Vol. 22, pp. 233-234.

14. "Tobias Simple," "Strictures on the account of the life and writings of David Hume," in Weekly Magazine, or Edinburgh Review, 1777, Vol. 38, pp. 289-292.

15. C.L. Stevenson, Ethics and Language, (New Haven: 1944), pp. 34-35.

16. Pall S. Ardal, Passion and Value, (Edinburgh: Edinburgh University Press, 1966), pp. 160-161.

17. J.L. Mackie, Hume's Moral Theory, (London: Routledge, 1980), p. 129.

18.  David Hume, The Natural History of Religion, ed. James Fieser (New York: MacMillan, 1992), p. 70.