Hume’s position in ethics, which is based on hisempiricist theory of the mind, is best known for asserting four theses: (1) Reason alonecannot be a motive to the will, but rather is the “slave of the passions” (seeSection 3) (2) Moral distinctions are not derived from reason (seeSection 4). (3) Moral distinctions are derived from the moral sentiments: feelings ofapproval (esteem, praise) and disapproval (blame) felt by spectatorswho contemplate a character trait or action (seeSection 7). (4) While some virtues and vices are natural (seeSection 13), others, including justice, are artificial (seeSection 9). There is heated debate about what Hume intends by each of thesetheses and how he argues for them. He articulates and defends them within the broader context of his metaethics and his ethic of virtue andvice.

Hume’s main ethical writings are Book 3 of hisTreatise of HumanNature, “Of Morals” (which builds on Book 2, “Ofthe Passions”), hisEnquiry concerning the Principles ofMorals, and some of hisEssays. In part the moralEnquiry simply recasts central ideas from the moral part oftheTreatise in a more accessible style; but there areimportant differences. The ethical positions and arguments of theTreatise are set out below, noting where the moralEnquiry agrees; differences between theEnquiry andtheTreatise are discussed afterwards.

• 1. Issues from Hume’s Predecessors
• 2. The Passions and the Will
• 3. The Influencing Motives of the Will
• 4. Ethical Anti-rationalism
• 5. Is and Ought
• 6. The Nature of Moral Judgment
• 7. Sympathy, and the Nature and Origin of the Moral Sentiments
• 8. The Common Point of View
• 9. Artificial and Natural Virtues
• 10. Honesty with Respect to Property
  • 10.1 The Circle
  • 10.2 The Origin of Material Honesty
  • 10.3 The Motive of Honest Actions
• 11. Fidelity to Promises
• 12. Allegiance to Government
• 13. The Natural Virtues
• 14. Differences between theTreatise and the MoralEnquiry
• Bibliography
• Academic Tools
• Other Internet Resources
• Related Entries

1. Issues from Hume’s Predecessors

Hume inherits from his predecessors several controversies aboutethics and political philosophy.

One is a question of moral epistemology: how do human beings becomeaware of, or acquire knowledge or belief about, moral good and evil,right and wrong, duty and obligation? Ethical theorists andtheologians of the day held, variously, that moral good and evil arediscovered: (a) by reason in some of its uses (Hobbes, Locke, Clarke),(b) by divine revelation (Filmer), (c) by conscience or reflection onone’s (other) impulses (Butler), or (d) by a moral sense: an emotionalresponsiveness manifesting itself in approval or disapproval(Shaftesbury, Hutcheson). Hume sides with the moral sense theorists:we gain awareness of moral good and evil by experiencing the pleasureof approval and the uneasiness of disapproval when we contemplate acharacter trait or action from an imaginatively sensitive and unbiasedpoint of view. Hume maintains against the rationalists that, althoughreason is needed to discover the facts of any concrete situation andthe general social impact of a trait of character or a practice overtime, reason alone is insufficient to yield a judgment that somethingis virtuous or vicious. In the last analysis, the facts as known musttrigger a response by sentiment or “taste.”

A related but more metaphysical controversy would be stated thustoday: what is the source or foundation of moral norms? In Hume’s daythis is the question what is the ground of moral obligation (asdistinct from what is the faculty for acquiring moral knowledge orbelief). Moral rationalists of the period such as Clarke (and in somemoods, Hobbes and Locke) argue that moral standards or principles arerequirements of reason — that is, that the very rationality ofright actions is the ground of our obligation to perform them. Divinevoluntarists of the seventeenth and eighteenth centuries such as SamuelPufendorf claim that moral obligation or requirement, if not every sortof moral standard, is the product of God’s will. The moral sensetheorists (Shaftesbury and Hutcheson) and Butler see all requirementsto pursue goodness and avoid evil as consequent upon human nature,which is so structured that a particular feature of our consciousness(whether moral sense or conscience) evaluates the rest. Hume sides withthe moral sense theorists on this question: it is because we are thekinds of creatures we are, with the dispositions we have for pain andpleasure, the kinds of familial and friendly interdependence that makeup our life together, and our approvals and disapprovals of these, thatwe are bound by moral requirements at all.

Closely connected with the issue of the foundations of moral norms isthe question whether moral requirements are natural or conventional.Hobbes and Mandeville see them as conventional, and Shaftesbury,Hutcheson, Locke, and others see them as natural. Hume mocksMandeville’s contention that the very concepts of vice and virtue arefoisted on us by scheming politicians trying to manage usmore easily. If there were nothing in our experience and no sentimentsin our minds to produce the concept of virtue, Hume says, nolavish praise of heroes could generate it. So to a degree moral requirements have a natural origin. Nonetheless,Hume thinks natural impulses of humanity and dispositions to approvecannot entirely account for our virtue of justice; a correct analysisof that virtue reveals that mankind, an “inventivespecies,” has cooperatively constructed rules of property andpromise. Thus he takes an intermediate position: some virtues arenatural, and some are the products of convention.

Linked with these meta-ethical controversies is the dilemma ofunderstanding the ethical life either as the “ancients” do,in terms of virtues and vices of character, or as the“moderns” do, primarily in terms of principles of duty ornatural law. While even so law-oriented a thinker as Hobbes has a gooddeal to say about virtue, the ethical writers of the seventeenth andeighteenth centuries predominantly favor a rule- or law-governedunderstanding of morals, giving priority to laws of nature orprinciples of duty. The chief exception here is the moral sense school,which advocates an analysis of the moral life more like that of theGreek and Hellenistic thinkers, in terms of settled traits of character— although they too find a place for principles in their ethics.Hume explicitly favors an ethic of character along“ancient” lines. Yet he insists on a role for rules of dutywithin the domain of what he calls the artificial virtues.

Hume’s predecessors famously took opposing positions on whether humannature was essentially selfish or benevolent, some arguing that manwas so dominated by self-interested motives that for moralrequirements to govern us at all they must serve our interests in someway, and others arguing that uncorrupted human beings naturally careabout the weal and woe of others and here morality gets its hold. Humeroundly criticizes Hobbes for his insistence on psychological egoismor something close to it, and for his dismal, violent picture of astate of nature. Yet Hume resists the view of Hutcheson that all moralprinciples can be reduced to our benevolence, in part because hedoubts that benevolence can sufficiently overcome our perfectly normalacquisitiveness. According to Hume’s observation, we are both selfishand humane. We possess greed, and also “limitedgenerosity” — dispositions to kindness and liberalitywhich are more powerfully directed toward kin and friends and lessaroused by strangers. While for Hume the condition of humankind in theabsence of organized society is not a war of all against all, neitheris it the law-governed and highly cooperative domain imagined byLocke. It is a hypothetical condition in which we would care for ourfriends and cooperate with them, but in which self-interest andpreference for friends over strangers would make any wider cooperationimpossible. Hume’s empirically-based thesis that we are fundamentallyloving, parochial, and also selfish creatures underlies his politicalphilosophy.

In the realm of politics, Hume again takes up an intermediateposition. He objects both to the doctrine that a subject mustpassively obey his government no matter how tyrannical it is and tothe Lockean thesis that citizens have a natural right to revolutionwhenever their rulers violate their contractual commitments to thepeople. He famously criticizes the notion that all political dutiesarise from an implicit contract that binds later generations who werenot party to the original explicit agreement. Hume maintains that theduty to obey one’s government has an independent origin that parallelsthat of promissory obligation: both are invented to enable people tolive together successfully. On his view, human beings can create asociety without government, ordered by conventional rules ofownership, transfer of property by consent, and promise-keeping. Wesuperimpose government on such a pre-civil society when it grows largeand prosperous; only then do we need to use political power to enforcethese rules of justice in order to preserve social cooperation. So theduty of allegiance to government, far from depending on the duty tofulfill promises, provides needed assurance that promises of all sortswill be kept. The duty to submit to our rulers comes into beingbecause reliable submission is necessary to preserve order. Particulargovernments are legitimate because of their usefulness in preservingsociety, not because those who wield power were chosen by God orreceived promises of obedience from the people. In a long-establishedcivil society, whatever ruler or type of government happens to be inplace and successfully maintaining order and justice is legitimate,and is owed allegiance. However, there is some legitimate recourse forvictims of tyranny: the people may rightly overthrow any governmentthat is so oppressive as not to provide the benefits (peace andsecurity from injustice) for which governments are formed. In hispolitical essays Hume certainly advocates the sort of constitutionthat protects the people’s liberties, but he justifies it not based on individual natural rights or contractual obligations but based on thegreater long-range good of society.

2. The Passions and the Will

According to Hume’s theory of the mind, the passions (what we todaywould call emotions, feelings, and desires) are impressions ratherthan ideas (original, vivid and lively perceptions that are not copiedfrom other perceptions). The direct passions, which include desire,aversion, hope, fear, grief, and joy, are those that “ariseimmediately from good or evil, from pain or pleasure” that weexperience or think about in prospect (T 2.1.1.4, T 2.3.9.2); howeverhe also groups with them some instincts of unknown origin, such as thebodily appetites and the desires that good come to those we love andharm to those we hate, which do not proceed from pain and pleasure butproduce them (T 2.3.9.7). The indirect passions, primarily pride,humility (shame), love and hatred, are generated in a more complexway, but still one involving either the thought or experience of painor pleasure. Intentional actions are caused by the direct passions(including the instincts). Of the indirect passions Hume says thatpride, humility, love and hatred do not directly cause action; it isnot clear whether he thinks this true of all the indirectpassions.

Hume is traditionally regarded as a compatibilist about freedom anddeterminism, because in his discussion in theEnquiry concerningHuman Understanding he argues that if we understand thedoctrines of liberty and necessity properly, all mankind consistentlybelieve both that human actions are the products of causal necessityand that they are free. In theTreatise, however, heexplicitly repudiates the doctrine of liberty as “absurd... inone sense, and unintelligible in any other” (T 2.3.2.1). The twotreatments, however, surprisingly enough, are entirely consistent. Hume construes causal necessityto mean the same as causal connection (or rather,intelligiblecausal connection), as he himself analyzes this notion in his owntheory of causation: either the “constant union and conjunctionof like objects,” or that together with “the inference ofthe mind from the one to the other” (ibid.). In both works heargues that just as we discover necessity (in this sense) to holdbetween the movements of material bodies, we discover just as muchnecessity to hold between human motives, character traits, andcircumstances of action, on the one hand, and human behavior on theother. He says in theTreatise that the liberty ofindifference is the negation of necessity in this sense; this is thenotion of liberty that he there labels absurd, and identifies withchance or randomness (which can be no real power in nature) both in theTreatise and the first (epistemological)Enquiry. Humanactions are not free in this sense. However, Hume allows in theTreatise that they are sometimes free in the sense of‘liberty’ which is opposed to violence or constraint. This is thesense on which Hume focuses in EcHU: “a power of acting ornot acting, according to the determinations of the will;”which everyone has “who is not a prisoner and in chains”(EcHU 8.1.23, Hume’s emphasis). It is this that is entirely compatiblewith necessity in Hume’s sense. So the positions in the two works arethe same, although the polemical emphasis is so different —iconoclastic toward the libertarian view in theTreatise, andconciliatory toward “all mankind” in the firstEnquiry.

Hume argues, as well, that the causal necessity of human actions isnot only compatible with moral responsibility but requisite to it. Tohold an agent morally responsible for a bad action, it is not enoughthat the action be morally reprehensible; we must impute the badness ofthe fleeting act to the enduring agent. Not all harmful or forbiddenactions incur blame for the agent; those done by accident, for example,do not. It is only when, and because, the action’s cause is someenduring passion or trait of character in the agent thatsheis to blame for it.

3. The Influencing Motives of the Will

According to Hume, intentional actions are the immediate product ofpassions, in particular the direct passions, including the instincts.He does not appear to allow that any other sort of mental state could,on its own, give rise to an intentional action except by producing apassion, though he does not argue for this. The motivating passions,in their turn, are produced in the mind by specific causes, as we seeearly in theTreatise where he first explains the distinctionbetween impressions of sensation and impressions of reflection:

An impression first strikes upon the senses, and makes usperceive heat or cold, thirst or hunger, pleasure or pain, of somekind or other. Of this impression there is a copy taken by the mind,which remains after the impression ceases; and this we call an idea.This idea of pleasure or pain, when it returns upon the soul, producesthe new impressions of desire and aversion, hope and fear, which mayproperly be called impressions of reflection, because derived from it.(T 1.1.2.2)

Thus ideas of pleasure or pain are the causes of these motivatingpassions. Not just any ideas of pleasure or pain give rise tomotivating passions, however, but only ideas of those pleasures orpains we believe exist or will exist (T 1.3.10.3). More generally,the motivating passions of desire and aversion, hope and fear, joy andgrief, and a few others are impressions produced by the occurrence inthe mind either of afeeling of pleasure or pain, whetherphysical or psychological, or of abelieved idea of pleasureor pain to come (T 2.1.1.4, T 2.3.9.2). These passions, together withthe instincts (hunger, lust, and so on), are all the motivatingpassions that Hume discusses.

The will, Hume claims, is an immediate effect of pain or pleasure (T2.3.1.2) and “exerts itself” when either pleasure or theabsence of pain can be attained by any action of the mind or body (T2.3.9.7). The will, however, is merely that impression we feel whenwe knowingly give rise to an action (T 2.3.1.2); so while Hume is notexplicit (and perhaps not consistent) on this matter, he seems not toregard the will as itself a (separate) cause of action. Thecauses of action he describes are those he has already identified: theinstincts and the other direct passions.

Hume famously sets himself in opposition to mostmoral philosophers, ancient and modern, who talk of the combat ofpassion and reason, and who urge human beings to regulate theiractions by reason and to grant it dominion over their contrarypassions. He claims to prove that “reason alone can never be amotive to any action of the will,” and that reason alone“can never oppose passion in the direction of the will” (T413). His view is not, of course, that reason plays no role in thegeneration of action; he grants that reason provides information, inparticular about means to our ends, which makes a difference to thedirection of the will. His thesis is that reasonalone cannotmove us to action; the impulse to act itself must come frompassion. The doctrine that reason alone is merely the “slave ofthe passions,” i.e., that reason pursues knowledge of abstractand causal relations solely in order to achieve passions’ goals andprovides no impulse of its own, is defended in theTreatise, but not in the secondEnquiry, although in thelatter he briefly asserts the doctrine without argument. Hume gives threearguments in theTreatise for the motivational“inertia” of reason alone.

The first is a largely empirical argument based on the two rationalfunctions of the understanding. The understanding discovers theabstract relations of ideas by demonstration (a process of comparingideas and finding congruencies and incongruencies); and it alsodiscovers the causal (and other probabilistic) relations of objectsthat are revealed in experience. Demonstrative reasoning is never thecause of any action by itself: it deals in ideas rather thanrealities, and we only find it useful in action when we have somepurpose in view and intend to use its discoveries to inform ourinferences about (and so enable us to manipulate) causes andeffects. Probable or cause-and-effect reasoning does play a role indeciding what to do, but we see that it only functions as anauxiliary, and not on its own. When we anticipate pain or pleasurefrom some source, we feel aversion or propensity to that object and“are carry’d to avoid or embrace what will give us” thepain or pleasure (T 2.3.3.3). Our aversion or propensity makes us seekthe causes of the expected source of pain or pleasure, and we usecausal reasoning to discover what they are. Once we do, our impulsenaturally extends itself to those causes, and we act to avoid orembrace them. Plainly the impulse to act does not arise from thereasoning but is only directed by it. “’Tis from the prospect ofpain or pleasure that the aversion or propensity arises...”(ibid.). Probable reasoning is merely the discovering of causalconnections, and knowledge that A causes B never concerns us if we areindifferent to A and to B. Thus, neither demonstrative nor probablereasoning alone causes action.

The second argument is a corollary of the first. It takes as a premise the conclusion just reached, that reason alone cannot produce an impulse to act. Given that, can reason prevent action or resist passion in controlling the will? To stop a volition or retard the impulse of an existing passion would require a contrary impulse. If reason alone could give rise to such a contrary impulse, it would have an original influence on the will (acapacity to cause intentional action, when unopposed); which,according to the previous argument, it lacks. Therefore reasonalone cannot resist any impulse to act. Therefore, what offers resistance to our passions cannot be reason of itself. Hume later proposes that when we restrainimprudent or immoral impulses, the contrary impulse comes also frompassion, but often from a passion so “calm” that weconfuse it with reason.

The third or Representation argument is different in kind. Hume offersit initially only to show that apassion cannot be opposed byor be contradictory to “truth and reason”; later (T3.1.1.9), he repeats and expands it to argue that volitions andactions as well cannot be so. One might suppose he means to giveanother argument to show that reason alone cannot provide a force toresist passion. Yet the Representation Argument is notempirical, and does not talk of forces or impulses. Passions (andvolitions and actions), Hume says, do not refer to other entities;they are “original existence[s],” (T 2.3.3.5),“original facts and realities” (T3.1.1.9), not mental representations of other things. Since Hume here understandsrepresentation in terms of copying, he says a passion has no“representative quality, which renders it a copy of any otherexistence or modification” (T 2.3.3.5). Contradiction to truthand reason, however, consists in “the disagreement of ideas,consider’d as copies, with those objects, which they represent”(ibid.). Therefore, a passion (or volition or action), not having thisfeature, cannot be opposed by truth and reason. Theargument allegedly proves two points: first, that actions cannot be reasonable or unreasonable; second, that “reason cannot immediately prevent or produce any action by contradicting or approving of it” (T3.1.1.10). The point here isnot merely the earlier, empirical observation that the rationalactivity of the understanding does not generate an impulse in theabsence of an expectation of pain or pleasure. The main point is that, becausepassions, volitions, and actions have no content suitable forassessment by reason, reason cannot assess prospective motives oractions as rational or irrational; and therefore reason cannot, by soassessing them, create or obstruct them. By contrast,reasoncan assess a potential opinion as rational orirrational; and by endorsing the opinion, reason will (that is, wewill) adopt it, while by contradicting the opinion, reason willdestroy our credence in it. The Representation Argument, then, makes apointa priori about the relevance of the functions of theunderstanding to the generation of actions. Interpreters disagreeabout exactly how to parse this argument, whether it is sound, and itsimportance to Hume’s project.

Hume allows that, speaking imprecisely, we often say a passion isunreasonable because it arises in response to a mistaken judgment oropinion, either that something (a source of pleasure or uneasiness)exists, or that it may be obtained or avoided by a certain means. Injust these two cases a passion may be called unreasonable, but strictlyspeaking even here it is not the passion but the judgment that is so.Once we correct the mistaken judgment, “our passions yield to ourreason without any opposition,” so there is still no combat ofpassion and reason (T 2.3.3.7). And there is no other instance ofpassion contrary to reason. Hume famously declaims, “’Tis notcontrary to reason to prefer the destruction of the whole world to thescratching of my finger. ‘Tis not contrary to reason for me tochuse my total ruin, to prevent the least uneasiness of an Indian orperson wholly unknown to me. ‘Tis as little contrary to reason toprefer even my own acknowledg’d lesser good to my greater, and have amore ardent affection for the former than for the latter.”(2.3.3.6)

Interpreters disagree as to whether Hume is an instrumentalist or askeptic about practical reason. Either way, Hume denies that reasoncan evaluate the ends people set themselves; only passions can selectends, and reason cannot evaluate passions. Instrumentalistsunderstand the claim that reason is the slave of the passions to allowthat reason not only discovers the causally efficacious means to ourends (a task of theoretical causal reasoning) but also requires us totake them. If Hume regards the failure to take the known means toone’s end as contrary to reason, then on Hume’s view reason has agenuinely practical aspect: it can classify some actions asunreasonable. Skeptical interpreters read Hume, instead, as denyingthat reason imposes any requirements on action, even the requirementto take the known, available means to one’s end. They point to thelist of extreme actions that are not contrary to reason (such aspreferring one’s own lesser good to one’s greater), and to theRepresentation Argument, which denies that any passions, volitions, oractions are of such a nature as to be contrary to reason. Hume neversays explicitly that failing to take the known means to one’s end iseither contrary to reason or not contrary to reason (it is not one ofthe extreme cases in his list). The classificatory point in theRepresentation Argument favors the reading of Hume as a skeptic aboutpractical reason; but that argument is absent from the moralEnquiry.

4. Ethical Anti-rationalism

Hume claims that moraldistinctions are not derived from reason but rather fromsentiment. His rejection of ethical rationalism is at leasttwo-fold. Moral rationalists tend to say, first, that moral propertiesare discovered by reason, and also that what is morally good is in accordwith reason (even that goodness consists in reasonableness) and what is morallyevil is unreasonable. Hume rejects both theses. Some of his argumentsare directed to one and some to the other thesis, and in places it is unclear which he means to attack.

In theTreatise he argues against the epistemic thesis (thatwediscover good and evil by reasoning) by showing thatneither demonstrative nor probable/causal reasoning has vice andvirtue as its proper objects. Demonstrative reasoning discoversrelations of ideas, and vice and virtue are not identical with any ofthe four philosophical relations (resemblance, contrariety, degrees inquality, or proportions in quantity and number) whose presence can bedemonstrated. Nor could they be identical with any other abstractrelation; for such relations can also obtain between items such astrees that are incapable of moral good or evil. Furthermore, weremoral vice and virtue discerned by demonstrative reasoning, suchreasoning would reveal their inherent power to produce motivesin all who discern them; but no causal connections can bediscovereda priori. Causal reasoning, by contrast, doesinfer matters of fact pertaining to actions, in particular theircauses and effects; but the vice of an action (its wickedness) is notfound in its causes or effects, but is only apparent when we consultthe sentiments of the observer. Therefore moral good and evil are notdiscovered by reason alone.

Hume also attempts in theTreatise to establish the otheranti-rationalist thesis, that virtue is not the same as reasonablenessand vice is not contrary to reason. He gives two arguments for this. The first, very short, argument he claims follows directly from the RepresentationArgument, whose conclusion was that passions, volitions, and actionscan be neither reasonable nor unreasonable. Actions, he observes,can be laudable orblamable. Since actions cannot be reasonable or against reason, itfollows that “[l]audable and blameable are not the same withreasonable or unreasonable” (T 458). The properties are notidentical.

The second and more famous argument makes use of the conclusiondefended earlier that reason alone cannot move us to act. As we haveseen, reason alone “can never immediately prevent or produce anyaction by contradicting or approving of it” (T 458). Morality— this argument goes on — influences our passions andactions: we are often impelled to or deterred from action by ouropinions of obligation or injustice. Therefore morals cannot bederived from reason alone. This argument is first introduced asshowing it impossible “from reason alone... to distinguishbetwixt moral good and evil” (T 457) — that is, it isbilled as establishing the epistemic thesis. But Hume also says that,like the little direct argument above, it proves that “actionsdo not derive their merit from a conformity to reason, nor their blamefrom a contrariety to it” (T458): it is not the reasonablenessof an action that makes it good, or its unreasonableness that makes itevil.

This argument about motives concludes that moral judgments orevaluations are not the products of reason alone. From this many drawthe sweeping conclusion that for Hume moral evaluations are notbeliefs or opinions of any kind, but lack all cognitive content. Thatis, they take the argument to show that Hume holds a non-propositionalview of moral evaluations — and indeed, given hissentimentalism, that he is an emotivist: one who holds that moraljudgments are meaningless ventings of emotion that can be neither truenor false. Such a reading should be met with caution, however. ForHume, to say that something is not a product of reason alone is notequivalent to saying it is not a truth-evaluable judgment orbelief. Hume does not consider all our (propositional) beliefs andopinions to be products of reason; some arise directly from senseperception, for example, and some from sympathy. Also, perhaps thereare (propositional) beliefs we acquire via probable reasoning but notby such reasoningalone. One possible example is the beliefthat some object is a cause of pleasure, a belief that depends uponprior impressions as well as probable reasoning.

Another concern about the famous argument about motives is how it could besound. In order for it to yield its conclusion, it seems that its premise thatmorality (or a moral judgment) influences the will must be construed tosay that moral evaluationsalone move us to action,without the help of some (further) passion. This is a controversialclaim and not one for which Hume offers any support. The premise thatreason alone cannot influence action is also difficult to interpret.It would seem, given his prior arguments for this claim (e.g. that themere discovery of a causal relation does not produce an impulse toact), that Hume means by it not only that the faculty of reason or theactivity of reasoning alone cannot move us, but also that theconclusions of such activity alone (such as recognition of a relationof ideas or belief in a causal connection) cannot produce a motive. Yet it ishard to see how Hume, given his theory of causation, can argue that nomental item of a certain type (such as a causal belief) can possiblycause motivating passion or action. Such a claim could not be supporteda priori. And inTreatise 1.3.10, “Of theinfluence of belief,” he seems to assert very plainly that somecausal beliefs do cause motivating passions, specifically beliefsabout pleasure and pain in prospect. It is possible that Hume onlymeans to say, in the premise that reason alone cannot influenceaction, that reasoning processes cannot generate actions as theirlogical conclusions; but that would introduce an equivocation, sincehe surely does not mean to say, in the other premise, that moralevaluations generate actions as their logical conclusions. Thetransition from premises to conclusion also seems to rely on aprinciple of transitivity (If A alone cannot produce X and B producesX, then A alone cannot produce B), which is doubtful but receives nodefense.

Commentators have proposed various interpretations to avoid thesedifficulties. One approach is to construe ‘reason’ as thename of a process or activity, the comparing of ideas (reasoning), andto construe ‘morals’ as Hume uses it in this argument tomean the activity of moral discrimination (making a moraldistinction). If we understand the terms this way, the argument canbe read not as showing that the faculty of reason (or the beliefs itgenerates) cannot cause us to make moral judgments, but rather asshowing that the reasoning process (comparing ideas) is distinct fromthe process of moral discrimination. This interpretation does notrely on an assumption about the transitivity of causation and isconsistent with Hume’s theory of causation.

5. Is and ought

Hume famously closes the section of theTreatise that arguesagainst moral rationalism by observing that other systems of moralphilosophy, proceeding in the ordinary way of reasoning, at some pointmake an unremarked transition from premises whose parts are linkedonly by “is” to conclusions whose parts are linked by“ought” (expressing a new relation) — a deductionthat seems to Hume “altogether inconceivable”(T3.1.1.27). Attention to this transition would “subvert all thevulgar systems of morality, and let us see, that the distinction ofvice and virtue is not founded merely on the relations of objects, noris perceiv’d by reason” (ibid.).

Few passages in Hume’s work have generated more interpretivecontroversy.

According to the dominant twentieth-century interpretation, Hume sayshere that no ought-judgment may be correctly inferred from a set ofpremises expressed only in terms of ‘is,’ and the vulgarsystems of morality commit this logical fallacy. This is usuallythought to mean something much more general: that no ethical or indeedevaluative conclusion whatsoever may be validly inferred from any setof purely factual premises. A number of present-day philosophers,including R. M. Hare, endorse this putative thesis of logic, callingit “Hume’s Law.” (As Francis Snare observes, on thisreading Hume must simply assume that no purely factual propositionsare themselves evaluative, as he does not argue for this.) Someinterpreters think Hume commits himself here to a non-propositional ornoncognitivist view of moral judgment — the view that moraljudgments do not state facts and are not truth-evaluable. (If Hume hasalready used the famous argument about the motivational influence ofmorals to establish noncognitivism, then the is/ought paragraph maymerely draw out a trivial consequence of it. If moral evaluations aremerely expressions of feeling without propositional content, then ofcourse they cannot be inferred from any propositional premises.) Somesee the paragraph as denying ethical realism, excluding values fromthe domain of facts.

Other interpreters — the more cognitivist ones — see theparagraph about ‘is’ and ‘ought’ as doing none of theabove. Some read it as simply providing further support for Hume’sextensive argument that moral properties are not discernible bydemonstrative reason, leaving open whether ethicalevaluations may be conclusions of cogent probable arguments. Othersinterpret it as making a point about the original discovery of virtueand vice, which must involve the use of sentiment. On this view, onecannot make the initial discovery of moral properties by inferencefrom nonmoral premises using reason alone; rather, one requires someinput from sentiment. It is not simply by reasoning from the abstractand causal relations one has discovered that one comes to have theideas of virtue and vice; one must respond to such information withfeelings of approval and disapproval. Note that on this reading it iscompatible with the is/ought paragraph that once a person has themoral concepts as the result of prior experience of the moralsentiments, he or she may reach some particular moral conclusions byinference from causal, factual premises (stated in terms of‘is’) about the effects of character traits on thesentiments of observers. They point out that Hume himself makes suchinferences frequently in his writings.

6. The Nature of Moral Judgment

On Hume’s view, what is a moral evaluation? Four main interpretationshave significant textual support. First, as we have seen, thenonpropositional view says that for Hume a moral evaluation does notexpress any proposition or state any fact; either it gives vent to afeeling, or it is itself a feeling (Flew, Blackburn, Snare,Bricke). (A more refined form of this interpretation allows that moralevaluations have some propositional content, but claims that for Humetheir essential feature, as evaluations, is non-propositional.) Thesubjective description view, by contrast, says that for Hume moralevaluations describe the feelings of the spectator, or the feelings aspectator would have were she to contemplate the trait or action fromthe common point of view. Often grouped with the latter view is thethird, dispositional interpretation, which understands moralevaluations as factual judgments to the effect that the evaluatedtrait or action is so constituted as to cause feelings of approval ordisapproval in a (suitably characterized) spectator (Mackie, in one ofhis proposals). On the dispositional view, in saying some trait isgood we attribute to the trait the dispositional property of beingsuch as to elicit approval. A fourth interpretation distinguishes twopsychological states that might be called a moral evaluation: anoccurrent feeling of approval or disapproval (which is not truth-apt),and a moral belief or judgment that is propositional. Versions ofthis fourth interpretation differ in what they take to be the contentof that latter mental state. One version says that the moraljudgments, as distinct from the moral feelings, are factual judgmentsabout the moral sentiments (Capaldi). A distinct version, the moralsensing view, treats the moral beliefs as ideas copied from theimpressions of approval or disapproval that represent a trait ofcharacter or an action as having whatever quality it is that oneexperiences in feeling the moral sentiment (Cohon). This last viewemphasizes Hume’s claim that moral good and evil are like heat, cold,and colors as understood in “modern philosophy,” which areexperienced directly by sensation, but about which we formbeliefs.

7. Sympathy, and the Nature and Origin of the Moral Sentiments

Our moral evaluations of persons and their character traits, onHume’s positive view, arise from our sentiments. The virtues and vicesare those traits the disinterested contemplation of which producesapproval and disapproval, respectively, in whoever contemplates thetrait, whether the trait’s possessor or another. These moralsentiments are emotions (in the present-day sense of that term) with aunique phenomenological quality, and also with a special set ofcauses. They are caused by contemplating the person or action to beevaluated without regard to our self-interest, and from a common orgeneral perspective that compensates for certain likely distortions inthe observer’s sympathies, as explained inSection 8. Approval (approbation) is a pleasure, and disapproval(disapprobation) a pain or uneasiness. The moral sentiments aretypically calm rather than violent, although they can be intensifiedby our awareness of the moral responses of others. Theyare types of pleasure and uneasiness that are associated with thepassions of pride and humility, love and hatred: when we feel moralapproval of another we tend to love or esteem her, and when weapprove a trait of our own we are proud of it. Some interpretersanalyze the moral sentiments as themselves forms of these fourpassions; others argue that Hume’s moral sentiments tend tocause the latter passions. We distinguish whichtraits are virtuous and which are vicious by means of ourfeelings of approval and disapproval toward the traits; our approval ofactions is derived fromapproval of the traits we suppose to have given rise to them. We candetermine, by observing the various sorts of traits toward which wefeel approval, that every such trait — every virtue — hasat least one of the following four characteristics: it is eitherimmediately agreeable to the person who has it or to others, or it isuseful (advantageous over the longer term) to its possessor or toothers. Vices prove to have the parallel features: they are eitherimmediately disagreeable or disadvantageous either to the person whohas them or to others. These are not definitions of‘virtue’ and ‘vice’ but empiricalgeneralizations about the traits as first identified by their effectson the moral sentiments.

In theTreatise Hume details the causes of the moralsentiments, in doing so explaining why agreeable and advantageoustraits prove to be the ones that generate approval. He claims that thesentiments of moral approval and disapproval are caused by some of theoperations of sympathy, which is not a feeling but rather apsychological mechanism that enables one person to receive bycommunication the sentiments of another (more or less what we wouldcall empathy today).

Sympathy in general operates as follows. First, observation of theoutward expression of another person’s “affection”(feeling or sentiment) in his “countenance andconversation” conveys the idea of his passion into my mind. Sodoes observing the typical cause of a passion: for example, viewingthe instruments laid out for another’s surgery will evoke ideasin me of fear and pain. We at all times possess a maximally vivid andforceful impression of ourselves. According to Hume’sassociationism, vivacity of one perception is automaticallytransferred to those others that are related to it by resemblance,contiguity, and cause and effect. Here resemblance and contiguity areprimary. All human beings, regardless of their differences, aresimilar in bodily structure and in the types and causes of theirpassions. The person I observe or consider may further resemble me inmore specific shared features such as character ornationality. Because of the resemblance and my contiguity to theobserved person, the idea of his passion is associated in my mind withmy impression of myself, and acquires great vivacity from it. The soledifference between an idea and an impression is the degree ofliveliness or vivacity each possesses. So great is this acquiredvivacity that the idea of his passion in my mind becomes animpression, and I actually experience the passion. When I come toshare in the affections of strangers, and feel pleasure because theyare pleased, as I do when I experience an aesthetic enjoyment of awell-designed ship or fertile field that is not my own, my pleasurecan only be caused by sympathy (T 2.2.2–8,3.3.1.7–8). Similarly, Hume observes, when we reflect upon acharacter or mental quality knowing its tendency either to the benefitor enjoyment of strangers or to their harm or uneasiness, we come tofeel enjoyment when the trait is beneficial or agreeable to thosestrangers, and uneasiness when the trait is harmful or disagreeable tothem. This reaction of ours to the tendency of a character trait toaffect the sentiments of those with whom we have no specialaffectionate ties can only be explained by sympathy.

We greatly approve the artificial virtues (justice with respect toproperty, allegiance to government, and dispositions to obey the lawsof nations and the rules of modesty and good manners), which (Humeargues) are inventions contrived solely for the interest ofsociety. We approve them in all times and places, even where our owninterest is not at stake, solely for their tendency to benefit thewhole society of that time or place. This instance confirms that“the reflecting on the tendency of characters and mentalqualities, is sufficient to give us the sentiments of approbation andblame” (T 3.3.1.9). The sympathy-generated pleasure,then,is the moral approbation we feel toward these traits ofcharacter. We find the character traits — the causes —agreeable because they are the means to ends we find agreeable as aresult of sympathy. Hume extends this analysis to the approval of mostof the natural virtues. Those traits of which we approve naturally(without any social contrivance), such as beneficence, clemency, andmoderation, also tend to the good of individuals or all of society. Soour approval of those can be explained in precisely the same way, viasympathy with the pleasure of those who receive benefit. And sincethe imagination is more struck by what is particular than by what isgeneral, manifestations of the natural virtues, which directly benefitany individual to whom they are directed, are even more apt to givepleasure via sympathy than are the manifestations of justice, whichmay harm identifiable individuals in some cases though they contributeto a pattern of action beneficial to society as a whole (T3.3.1.13).

8. The Common Point of View

As we saw, the moral sentiments are produced by sympathy with thoseaffected by a trait or action. Such sympathetically-acquired feelingsare distinct from our self-interested responses, and an individual ofdiscernment learns to distinguish her moral sentiments (which aretriggered by contemplating another’s character trait “ingeneral”) from the pleasure or uneasiness she may feel whenresponding to a trait with reference to her “particularinterest,” for example when another’s strength of character makeshim a formidable opponent (T 3.1.2.4).

However, the sympathetic transmission of sentiments can vary ineffectiveness depending upon the degree of resemblance and contiguitybetween the observer and the person with whom he sympathizes. Ireceive the sentiments of someone very much like me or very close tome in time or place far more strongly than I do those of someoneunlike me or more remote from me in location or in history. Yet themoral assessments we make do not vary depending upon whether theperson we evaluate resembles us in language, sex, or temperament, oris near or far. Indeed, our moral assessments of people remain stableeven though our position with respect to them changes over time.Furthermore, sympathy only brings us people’s actual sentiments orwhat we believe to be their actual sentiments; yet we feel moralapproval of character traits that we know produce no real happinessfor anyone, because, for example, their possessor is isolated in aprison. To handle these objections to the sympathy theory, and toexplain more generally how, on a sentiment-based ethical theory, moralevaluations made by one individual at different times and manyindividuals in a community tend to be fairly uniform, Hume claims thatpeople do not make their moral judgments from their own individualpoints of view, but instead select “some common point of view,from which they might survey their object, and which might cause it toappear the same to all of them” (T 3.3.1.30). At least withrespect to natural virtues and vices, this common point of view iscomposed of the intimate perspectives of the various individuals whohave direct interactions with the person being evaluated. To make amoral evaluation I must sympathize with each of these persons in theirdealings with the subject of my evaluation; the blame or praise I giveas a result of this imaginative exercise is my genuine moralassessment of the subject’s character. In that assessment I alsooverlook the small external accidents of fortune that might render anindividual’s trait ineffectual, and respond to traits that render acharacter typically “fitted to be beneficial to society,”even if circumstances do not permit it to cause that benefit (T3.3.1.20). Thus I acquire by sympathy the pleasure or uneasiness thatI imagine peoplewould feel were the trait able to operate as it ordinarilydoes. “Experience soon teaches us this method of correcting oursentiments, or at least, of correcting our language, where thesentiments are more stubborn and inalterable” (T 3.3.1.16).

9. Artificial and Natural Virtues

The standard object of moral evaluation is a “quality ofmind,” a character trait. (As we have seen, for Hume evaluationof anaction is derived from evaluation of the inner qualitywe suppose to have given rise to it.) The typical moral judgment isthat some trait, such as a particular person’s benevolence orlaziness, is a virtue or a vice. A character trait, for Hume, is apsychological disposition consisting of a tendency to feel a certainsentiment or combination of sentiments, ones that often move theirpossessor to action. We reach a moral judgment by feeling approval ordisapproval upon contemplating someone’s trait in a disinterested wayfrom the common point of view. So moral approval is a favorablesentiment in the observer elicited by the observed person’sdisposition to have certain motivating sentiments. Thus moralapproval is a sentiment that is directed toward sentiments, or thedispositions to have them.

In theTreatise Hume emphasizes that “our sense ofevery kind of virtue is not natural; but … there are somevirtues, that produce pleasure and approbation by means of an artificeor contrivance, which arises from the circumstances and necessities ofmankind” (T 3.2.1.1). He divides the virtues into those that arenatural — in that our approval of them does not depend upon anycultural inventions or jointly-made social rules — and thosethat are artificial (dependent both for their existence as charactertraits and for their ethical merit on the presence of conventionalrules for the common good), and he gives separate accounts of the twokinds. The traits he calls natural virtues are more refined andcompleted forms of those human sentiments we could expect to find evenin people who belonged to no society but cooperated only within smallfamilial groups. The traits he calls artificial virtues are the oneswe need for successfulimpersonal cooperation; our naturalsentiments are too partial to give rise to these withoutintervention. In theTreatise Hume includes among theartificial virtues honesty with respect to property (which he oftencalls equity or “justice,” though it is a strangely narrowuse of the term), fidelity to promises (sometimes also listed under“justice”), allegiance to one’s government, conformity tothe laws of nations (for princes), chastity (refraining fromnon-marital sex) and modesty (both primarily for women and girls), andgood manners. A great number of individual character traits are listedas natural virtues, but the main types discussed in detail aregreatness of mind (“a hearty pride, or self-esteem, ifwell-concealed and well-founded,” T 3.2.2.11), goodness orbenevolence (an umbrella category covering generosity, gratitude,friendship, and more), and such natural abilities as prudence and wit,which, Hume argues, have a reasonably good claim to be included underthe title moral virtue, though traditionally they are not. Hume doesnot explicitly draw a distinction between artificial and naturalvirtues in the moralEnquiry.

In theTreatise Hume argues in turn that the virtues ofmaterial honesty and of faithfulness to promises and contracts areartificial, not natural virtues. Both arguments fall into at least twostages: one to show that if we suppose the given character trait toexist and to win our approval without help from any cooperative socialarrangement, paradoxes arise; and another, longer stage to explain howthe relevant convention might have come into being and to refute thosewith a different genetic story. He also explains the socialconstruction of the other artificial virtues and what social good theyserve.

10. Honesty with respect to Property

10.1 The Circle

Hume offers a rather cryptic argument to show that our approval ofmaterial honesty must be the product of collaborative human effort(convention). When we approve an action, he says, we regard it merelyas the sign of the motivating passion in the agent’s “mind andtemper” that produced it; our evaluation of the action is derivedfrom our assessment of this inner motive. Therefore all actions deemedvirtuous derive their goodness only from virtuous motives —motives we approve. It follows from this that the motive thatoriginally “bestows a merit on any action” can never bemoral approval of that action (awareness of its virtue), but must be anon-moral, motivating psychological state — that is, a statedistinct from the “regard to the virtue” of an action(moral approval or disapproval) (T 3.2.1.4). For if thevirtue-bestowing motive of the action were the agent’s sense that theact would be virtuous to do — if that were why he did it, and whywe approved it — then we would be reasoning in a vicious circle:we would approve of the action derivatively, because we approve of theagent’s motive, and this motive would consist of approval of theaction, which can only be based on approval of a motive... The basis ofour approval could not be specified. For every virtue, therefore, theremust be some non-moral motive that characteristicallymotivates actions expressive of that virtue, which motive, by elicitingour approval, makes the actions so motivated virtuous. The virtue of anaction of this species would be established by its being done from thisnon-moral motive, and only then could an agent also or alternatively bemoved so to act by her derivative concern for the virtue of the act.However, Hume observes that thereis no morally approved (andso virtue-bestowing), non-moral motive of honest action. The onlyapproved, reliable motive that we can find for acts of“equity” is a moral one, the sense of virtue or “regard tothe honesty” of the actions. The honest individual repays a loan not(merely) out of self-interest or concern for the well-being of thelender (who may be a “profligate debauchee” who will reaponly harm from his possessions), but from a “regard to justice,and abhorrence of villainy and knavery” (T 3.2.1.9, 13). This,however, is “evident sophistry and reasoning in acircle…” Now nature cannot have “establish’d asophistry, and render’d it necessary and unavoidable…”;therefore, “the sense of justice and injustice is not deriv’dfrom nature, but arises artificially… from education, and humanconventions” (T 3.2.1.17). Whatever, exactly, the logic of thisargument is supposed to be, Hume’s intent is to show that if we imagineequity to be a natural virtue we commit ourselves to a sophistry, andtherefore honesty is instead man-made.

Hume offers an account of the genesis of the social convention thatcreates honesty with respect to property, and this is meantto cope in some way with the circularity he identifies. How it does sois a matter of interpretive controversy, as we will see.

10.2 The Origin of Material Honesty

Hume next poses two questions about the rules of ownership of propertyand the associated virtue of material honesty: what is the artifice bywhich human beings create them, and why do we attribute moral goodnessand evil to the observance and neglect of these rules?

By nature human beings have many desires but are individuallyill-equipped with strength, natural weapons, or natural skills tosatisfy them. We can remedy these natural defects by means of socialcooperation: shared strength, division of labor, and mutualaid in times of individual weakness. It occurs to people to form asociety as a consequence of their experience with the small familygroups into which they are born, groups united initially by sexualattraction and familial love, but in time demonstrating the manypractical advantages of working together with others. However, in theconditions of moderate scarcity in which we find ourselves, and giventhe portable nature of the goods we desire, our untrammeled greed andnaturally “confined generosity” (generosity to those dearto us in preference to others) tends to create conflict or underminecooperation, destroying collaborative arrangements among people whoare not united by ties of affection, and leaving us all materiallypoor. No remedy for this natural partiality is to be found in“our natural uncultivated ideas of morality” (T 3.2.2.8);an invention is needed.

Hume argues that we create the rules of ownership of propertyoriginally in order to satisfy our avidity for possessions forourselves and our loved ones, by linking material goods more securelyto particular individuals so as to avoid conflict. Within small groupsof cooperators, individuals signal to one another a willingness toconform to a simple rule: to refrain from the material goods otherscome to possess by labor or good fortune, provided those others willobserve the same restraint toward them. (This rule will in time requiremore detail: specific rules determining who may enjoy which goodsinitially and how goods may be transferred.) This signalling is not apromise (which cannot occur without another, similar convention), butan expression of conditional intention. The usefulness of such a customis so obvious that others will soon catch on and express a similarintention, and the rest will fall in line. The convention developstacitly, as do conventions of language and money. When an individualwithin such a small society violates this rule, the others are aware ofit and exclude the offender from their cooperative activities. Once theconvention is in place, justice (of this sort) is defined as conformitywith the convention, injustice as violation of it; indeed, theconvention defines property rights, ownership, financial obligation,theft, and related concepts, which had no application before theconvention was introduced. So useful and obvious is this invention thathuman beings would not live for long in isolated family groups or influctuating larger groups with unstable possession of goods; theiringenuity would quickly enable them to invent property, so as to reapthe substantial economic benefits of cooperation in larger groups inwhich there would be reliable possession of the product, and they wouldthus better satisfy their powerful natural greed by regulating it withrules of justice.

Greed, and more broadly, self-interest, is the motive for inventingproperty; but we need a further explanation why we think of justice(adherence to the rules of ownership) as virtuous, and injustice (theirviolation) as vicious. Hume accounts for the moralization of propertyas follows. As our society grows larger, we may cease to see our ownproperty violations as a threat to the continued existence of a stableeconomic community, and this reduces our incentive to conform. Butwhen we consider violations by others, we partake by sympathy in theuneasiness these violations cause to their victims and all of society. Suchdisinterested uneasiness, and the concomitant pleasure we feel oncontemplating the public benefits of adherence, are instances of moraldisapproval and approval. We extend these feelings to our own behavior as aresult of general rules. This process is “forwarded by theartifice of politicians” (T 3.2.2.25), who assist nature bycultivating widespread esteem for justice and abhorrence of injustice in order togovern more easily. Private education assists in this further artifice.Thus material honesty becomes a virtue.

10.3 The Motive of Honest Actions

Does this account resolve the circularity problem? Is there anynon-moral motive of honest action? Some interpreters say yes, it isgreed redirected, which removes the circle. But this presents twodifficulties: first, our greed is not in fact best satisfied by justaction in every case, and second, Hume denies that this motive isapproved. Some interpret Hume as coping with the first difficulty bysupposing that politicians and parents deceive us into thinking,falsely, that every individual just act advances the interests of theagent; or they claim that Hume himself mistakenly thought so, at leastin theTreatise (see Baron, Haakonssen, and Gauthier). Othersclaim that Hume identifies a non-moral motive of honest action (albeitan artificial one) other than redirected greed, such as a dispositionto treat the rules of justice as themselves reason-giving (Darwall) orhaving a policy of conforming to the rules of justice as a system(Garrett). Still others say there is no non-moral motive of honestaction, and Hume escapes from the circle by relaxing this ostensiblyuniversal requirement on virtuous types of behavior, limiting it tothe naturally virtuous kinds. These interpreters either claim thatthere is no particular motive needed to evoke approval for conformityto the rules of property — mere behavior is enough (Mackie)— or that we approve of a motivating form of the moral sentimentitself, the sense of duty (Cohon).

Hume’s genetic account of property is striking for its lack ofpatriarchal assumptions about the family, its explicit denial that thecreation of ownership does or can depend on any promise or contract,and its concept of convention as an informal practice of mutualcompromise for mutual advantage that arises incrementally and entirelyinformally, without the use of central authority or force.

11. Fidelity to Promises

Fidelity is the virtue of being disposed to fulfill promises andcontracts. Hume has in mind promises made “at arm’s length”that parties undertake to promote their own interest, not affectionateexchanges of favors between friends. While he identifies the same circularitypuzzle about the approved motive of fidelity that he tackles at lengthin connection with honesty, in the case of fidelity he concentrates on adifferent conundrum that arises with the misguided attempt to analyzefidelity as a non-conventional (natural) virtue. Unlike Hobbes andLocke, who help themselves to the concept of a promise or contract intheir imagined state of nature, Hume argues that the performativeutterance “I promise” would be unintelligible in theabsence of background social conventions, and that the moral obligationof a promise is dependent upon such conventions as well.

Suppose the practice of giving and receiving promises did not dependon a socially-defined convention. In that case, what could we mean bythe utterances we use to make them, and what would be the origin of ourobligation to fulfill them? Where the words are used(uncharacteristically) in a way that does notpurport toreveal the agent’s will (as when the person is joking or play-acting), we do not think a promise is really beingmade; we only take a speaker to have promised, and so to be bound toperform, if he understands the words he uses, in particular aspurporting to obligate him. Thus for effective use there must be someact of the speaker’s mind expressed by the special phrase “Ipromise” and its synonyms, and our moral obligation results fromthis act of the mind. (This seems to be Hobbes’s assumption inLeviathan, where the implicit signs of covenant — asdistinct from the explicit ones — are clear signs of the person’swill.) The requisite mental act or mental state, though, could not beone of mere desire or resolution to act, since it does not follow fromour desiring or resolving to act that we are morally obligated to doso; nor could it be the volition to act, since that does not come intobeing ahead of time when we promise, but only when the time comes toact. And of course, one can promise successfully (incur obligation bypromising) even though one has no intention to perform; so the mentalact requisite to obligation is not the intention to perform. The only likely act ofmind that might be expressed in a promise is a mental act ofwilling to be obligated to perform the promised action, asthis conforms to our common view that we bind ourselves by choosing tobe bound.

But, Hume argues, it is absurd to think that one can actually bringan obligation into existence by willing to be obligated. What makes anaction obligatory is that its omission is disapproved by unbiasedobservers. But no act of will within an agent can directly change a previouslyneutral act into one that provokes moral disapproval in observers(even in the agent herself). Sentiments are not subject to suchvoluntary control. Even on a moral rationalist view the thesis would be absurd: to create a new obligation would be to change the abstractrelations in which actions and persons stand to oneanother, and one cannot do this by performing in one’s own mind an act of willing such a relation toexist. Thus, there is no such act of the mind. Even if people in theirnatural (pre-conventional) condition “cou’d perceive each other’sthoughts by intuition,” they could not understand one another tobind themselves by any act of promising, and could not be obligatedthereby. Since the necessary condition for a natural obligation ofpromises cannot be fulfilled, we may conclude that this obligation isinstead the product of group invention to serve the interests ofsociety.

Promises are invented in order to build upon the advantages affordedby property. The invention of mere ownership suffices tomake possession stable. The introduction of transfer by consent permits some trade, but so far only simultaneous swapping of visiblecommodities. Great advantages could be gained by all if people couldbe counted on to provide goods or services later for benefits givennow, or exchange goods that are distant or described generically. Butfor people without the capacity to obligate themselves to futureaction, such exchanges would depend upon the party who performs second doing soout of gratitude alone; and that motive cannot generally berelied on in self-interested transactions. However, we can devise betterways to satisfy our appetites “in an oblique and artificial manner...”(T 3.2.5.9). First, people can easily recognize that additional kindsof mutual exchanges would serve their interests. They need onlyexpress this interest to one another in order to encourage everyone toinvent and to keep such agreements. They devise a form of words tomark these new sorts of exchanges (and distinguish them from thegenerous reciprocal acts of friendship and gratitude). When someoneutters this form of words, he is understood to express a resolution todo the action in question, and he “subjects himself to the penalty ofnever being trusted again in case of failure” (T 3.2.5.10), a penaltymade possible by the practice of the group, who enforce therequirement to keep promises by the simple expedient of refusing tocontract with those whose word cannot be trusted . This “concert orconvention” (ibid.) alters human motives to act. One is moved byself-interest to give the promising sign (in order to obtain the otherparty’s cooperation), and once one has given it, self-interest demandsthat one do what one promised to do so as to insure that people willexchange promises with one in the future. Some interpreters say thatthis enlightened self-interest remains the only motive for keepingone’s promise, once the practice of promising has been created. ButHume says the sentiment of morals comes to play the same role inpromise-keeping that it does in the development of honesty withrespect to property (T 3.2.5.12); so there is evidence he thinks themoral sentiment not only becomes “annex’d” topromise-keeping but further motivates it. In larger, more anonymouscommunities, a further incentive is needed besides the fear ofexclusion; and a sentiment of moral approval of promise-keeping arisesas the result of sympathy with all who benefit from the practice,aided by a “second artifice,” the well-meaningpsychological manipulation of the people by parents and politicians,which yields a near-universal admiration of fidelity and shame atbreaking one’s word (T 3.2.5.12). This may provide a moral motive forpromise-keeping even in anonymous transactions.

12. Allegiance to Government

A small society can maintain a subsistence-level economy without anydominion of some people over others, relying entirely on voluntarycompliance with conventions of ownership, transfer of goods, andkeeping of agreements, and relying on exclusion as the sole means ofenforcement. But an increase in population and/or materialproductivity, Hume thinks, tends to stimulate a destabilizing rate ofdefection from the rules: more luxury goods greatly increase thetemptation to act unjustly, and more anonymous transactions make itseem likely that one will get away with it. Though people are awarethat injustice is destructive of social cooperation and so ultimatelydetrimental to their own interests, this knowledge will not enablethem to resist such strong temptation, because of an inherent humanweakness: we are more powerfully drawn to a near-term good even whenwe know we will pay for it with the loss of a greater long-termgood. This creates the need for government to enforce the rules ofproperty and promise (the “laws of nature,” as Humesometimes rather ironically calls them, since on his view they are notnatural). This is the reason for the invention of government. Once inpower, rulers can also make legitimate use of their authority toresolve disputes over just what the rules of justice require inparticular cases, and to carry out projects for the common good suchas building roads and dredging harbors.

Hume thinks it unnecessary to prove that allegiance to government isthe product of convention and not mere nature, since governments areobviously social creations. But he does need to explain the creation ofgovernments and how they solve the problem he describes. He speculatesthat people who are unaccustomed to subordination in daily life mightdraw the idea for government from their experience of wars with othersocieties, when they must appoint a temporary commander. To overcomethe preference for immediate gain over long-term security, the peoplewill need to arrange social circumstances so that the conformity tojustice is in people’s immediate interest. This cannot be done withrespect to all the people, but it can be done for a few. So the peopleselect magistrates (judges, kings, and the like) and so position them(presumably with respect to rank and wealth) that it will be in thosemagistrates’ immediate interest not only to obey but to enforce therules of justice throughout society. Hume is vague about the incentives of the magistrates, but apparently they are so pleased with their ownshare of wealth and status that they are not tempted by the possessionsof others; and since they are “indifferent… to thegreatest part of the state,” they have no incentive to assistanyone in any crimes (T3.2.7.6). Thus the magistrates’ most immediateinterest lies in preserving their own status and wealth by protectingsociety. (Perhaps more directly, they stand to lose their favoredstatus if they are found by the people not to enforce the rules ofjustice.)

It is possible for the people to agree to appoint magistrates in spiteof the incurable human attraction to the proximal good even whensmaller than a remote good, because this predilection only takeseffect when the lesser good is immediately at hand. When consideringtwo future goods, people always prefer the greater, and make decisionsaccordingly. So looking to the future, people can decide now toempower magistrates to force them to conform to the rules of justicein the time to come so as to preserve society. When the time comes toobey and individuals are tempted to violate the rules, the long-rangethreat this poses to society may not move them to desist, but theimmediate threat of punishment by the magistrates will.

We initially obey our magistrates from self-interest. But oncegovernment is instituted, we come to have a moral obligation to obeyour governors; this is another artificial duty that needs to beexplained. On Hume’s view it is independent of the obligation of promises.We are bound to our promises and to obey the magistrates’ commands onparallel grounds: because both kinds of conformity are so manifestly beneficial for all. Governors merely insure that the rules of justice are generallyobeyed in the sort of society where purely voluntary conventions wouldotherwise break down. As in the case of fidelity to promises, thecharacter trait of allegiance to our governors generates sympathy withits beneficiaries throughout society, making us approve the trait as avirtue.

The duty of allegiance to our present governors does not depend upontheir or their ancestors’ divine right to govern, Hume says, nor onany promise we have made to them or any contract that transfers rightsto them, but rather on the general social value of having agovernment. Rulers thus need not be chosen by the people in order tobe legitimate. Consequently, who is the ruler will often be a matterof salience and imaginative association; and it will be no ground forlegitimate rebellion that a ruler was selected arbitrarily. Rulersidentified by long possession of authority, present possession,conquest, succession, or positive law will be suitably salient and solegitimate, provided their rule tends to the common good. Althoughgovernments exist to serve the interests of their people, changingmagistrates and forms of government for the sake of small advantagesto the public would yield disorder and upheaval, defeating the purposeof government; so our duty of allegiance forbids this. A governmentthat maintains conditions preferable to what they would be without itretains its legitimacy and may not rightly be overthrown. Butrebellion against a cruel tyranny is no violation of our duty ofallegiance, and may rightly be undertaken.

Hume does advocate some forms of government as being preferable toothers, particularly in hisEssays. Governments structured bylaws are superior to those controlled by the edicts of rulers or rulingbodies (“That Politics May Be Reduced to a Science”).Representative democracy is superior to direct democracy, and“free” (popular) governments are more hospitable to tradethan “absolute” governments (ibid.). Hume speculates that aperfect government would be a representative democracy ofproperty-holders with division of powers and some features offederalism (“Idea of a Perfect Commonwealth”). He defendshis preferences by arguing that certain forms of government are lessprone to corruption, faction (with the concomitant threat of civilwar), and oppressive treatment of the people than others; that is, theyare more likely to enforce the rules of justice, adjudicate fairly, andencourage peace and prosperity.

Hume famously criticizes the social contract theory of politicalobligation. According to his own theory, our duty to obey ourgovernors is not reducible to an instance of our duty to fulfillpromises, but arises separately though in a way parallel to thegenesis of that duty. Hume denies that any native citizen or subjectin his own day has made even a tacit promise to obey the government,given that citizens do not think they did any such thing, but ratherthink they are born to obey it. Even a tacit contract requires thatthe will be engaged, and we have no memory of this; nor do governmentsrefrain from punishing disloyalty in citizens who have given no tacitpromise.

13. The Natural Virtues

In theTreatise Hume’s principle interest in the naturalvirtues lies in explaining the causes that make us approve them. Themechanism of sympathy ultimately accounts for this approval and thecorresponding disapproval of the natural vices. Sympathy also explainsour approval of the artificial virtues; the difference is that weapprove of those as a result of sympathy with the cumulative effectsproduced by the general practice of the artificial virtues on thewhole of society (individual acts of justice not always producingpleasure for anyone); whereas we approve each individual exercise ofsuch natural virtues as gratitude and friendship because we sympathizewith those who are affected by each such action when we consider itfrom the common point of view. As we saw, he argues that the traits ofwhich we approve fall into four groups: traits immediately agreeableto their possessor or to others, and traits advantageous to theirpossessor or to others. In these four groups of approved traits, ourapproval arises as the result of sympathy bringing into our minds thepleasure that the trait produces for its possessor or for others (withone minor exception). This is especially clear with suchself-regarding virtues as prudence and industry, which we approve evenwhen they occur in individuals who provide no benefit to us observers;this can only be explained by our sympathy with the benefits thatprudence and industry bring to their possessors.

According to Hume, different levels and manifestations of the passionsof pride and humility make for virtue or for vice. An obvious and“over-weaning conceit” is disapproved by any observer (isa vice) (T 3.3.2.1); while a well-founded but concealed self-esteem isapproved (is a virtue). Hume explains these opposite reactions to suchclosely related character traits by means of the interplay of theobserver’s sympathy with a distinct psychological mechanism he callscomparison. The mechanism of comparison juxtaposes asympathetically-communicated sentiment with the observer’s owninherent feeling, causing the observer to feel a sentiment opposite tothe one she observes in another (pleasure if the other is suffering,pain if the other is pleased) when the sympathetically-communicatedsentiment is not too strong. A person who displays excessive prideirritates others because, while others come to feel this person’spleasant sentiment of pride (to some degree) via sympathy, they alsofeel a greater uneasiness as a result of comparing that great pride(in whose objects they do not believe) with their own lesser pride inthemselves; this is why conceit is a vice. Self-esteem founded on anaccurate assessment of one’s strengths and politely concealed fromothers, though, is both agreeable and advantageous to its possessorwithout being distressing to others, and so is generallyapproved. (Thus the professed preference of Christians for humilityover self-esteem does not accord with the judgments of mostobservers.) Although excessive pride is a natural vice and self-esteema natural virtue, human beings in society createtheartificial virtue of good breeding (adherence to customsof slightly exaggerated mutual deference in accordance with socialrank) to enable us each to conceal our own pride easily so that itdoes not shock the pride of others.

Courage and military heroism are also forms of pride. Though thestudent of history can see that military ambition has mostly beendisadvantageous to human society, when we contemplate the“dazling” character of the hero, immediate sympathyirresistibly leads us to approve it (T 3.3.2.15).

Our approval of those traits that may be grouped together under theheading of goodness and benevolence, such as generosity, humanity,compassion, and gratitude, arises from sympathy with people in theindividual’s “narrow circle” of friends and associates,since, given natural human selfishness, we cannot expect people’sconcerns to extend farther (T 3.3.3.2). By adopting the common point ofview we correct for the distortions of sympathy by entering into thefeelings of those close to the person being evaluated even if they areremote from us. The vice of cruelty is most loathed because thesuffering of the person’s victims that reaches us via sympathy readilybecomes hatred of the perpetrator.

Although natural abilities of the mind are not traditionallyclassified as moral virtues and vices, the difference between thesetypes of traits is unimportant, Hume argues. Intelligence, goodjudgment, application, eloquence, and wit are also mental qualitiesthat bring individuals the approbation of others, and their absence isdisapproved. As is the case with many of the traditionally-recognizedvirtues, the various natural abilities are approved either because theyare useful to their possessor or because they are immediately agreeableto others. It is sometimes argued that moral virtues are unlike naturalabilities in that the latter are involuntary, but Hume argues that manytraditional moral virtues are involuntary as well. The sole differenceis that the prospect of reward or punishment can induce people to actas the morally virtuous would (as justice requires, for example), butcannot induce them to act as if they had the natural abilities.

14. Differences between theTreatise and the MoralEnquiry

Late in his life Hume deemed theEnquiry concerning the Principlesof Morals his best work, and in style it is a model of eleganceand subtlety. His method in that work differs from that of theTreatise: instead of explicating the nature of virtue and viceand our knowledge of them in terms of underlying features of the humanmind, he proposes to collect all the traits we know from common senseto be virtues and vices, observe what those in each group have incommon, and from that observation discover the “foundation ofethics” (EPM 1.10). The conclusions largely coincide with thoseof theTreatise. Some topics in theTreatise arehandled more fully in the moralEnquiry; for example Hume’saccount of the motive to just action is enriched by his discussionof a challenge from a “sensible knave.” However, withoutthe detailed background theories of the mind, the passions, motivationto action, and social convention presented in theTreatise,and without any substitute for them, some of the conclusions of themoralEnquiry stand unsupported.

In the latter work, Hume’s main argument that reason alone is notadequate to yield moral evaluations (in Appendix 1) depends onhis having demonstrated throughout the book that at least onefoundation of moral praise lies in the usefulness to society of thepraised character trait. We use reason extensively to learn the effectsof various traits and to identify the useful and pernicious ones. Bututility and disutility are merely means; were we indifferent to theweal and woe of mankind, we would feel equally indifferent to thetraits that promote those ends. Therefore there must be some sentimentthat makes us favor the one over the other. This could only behumanity, “a feeling for the happiness of mankind, and resentmentof their misery” (EPM App. 1.3). This argument presupposes thatthe moral evaluations we make are themselves the expression ofsentiment rather than reason alone. (The alternative position would bethat while of course we do feel approval and disapproval for vice andvirtue, the judgment as to which is which is itself the deliverance ofreason.) So Hume appends some arguments directed against the hypothesisof moral rationalism. One of these is an enriched version of theargument of Treatise 3.1.1 that neither demonstrative nor causalreasoning has moral distinctions as its proper object, since moral viceand virtue cannot plausibly be analyzed as either facts or relations.He adds that while in our reasonings we start from the knowledge ofrelations or facts and infer some previously-unknown relation or fact,moral evaluation cannot proceed until all the relevant facts andrelations are already known. At that point, there is nothing furtherfor reason to do; therefore moral evaluation is not the work of reasonalone but of another faculty. He bolsters this line of argument byexpanding hisTreatise analogy between moral and aestheticjudgment, arguing that just as our appreciation of beauty awaits fullinformation about the object but requires the further contribution oftaste, so in moral evaluation our assessment of merit or villainyawaits full knowledge of the person and situation but requires thefurther contribution of approbation or disapprobation. He also offersthe argument that since the chain of reasons why one acts must finallystop at something that is “desirable on its own account…because of its immediate accord or agreement withsentiment…” (EPM App.1.19), sentiment is needed to accountfor ultimate human ends; and since virtue is an end, sentiment and notreason alone must distinguish moral good and evil.

In the moralEnquiry Hume omits all arguments to show thatreason alone does not move us to act; so the RepresentationArgument about the irrelevance of reason to passions and actions is absent.Without it he has no support for his directargument that moral goodness and evil are not identical withreasonableness and unreasonableness, which relies on it for its keypremise; and that too is absent from EPM. On the whole in EPM Humedoes not appeal to the thesis that reason cannot produce motives inorder to show that morals are not derived from reason alone, butlimits himself to the epistemic and descriptive arguments showing thatreason alone cannot discern virtue and vice in order to reject ethicalrationalism in favor of sentimentalism. However, at Appendix I.21 hedoes assert (without support) that “Reason, being cool and disengaged,is no motive to action,” and perhaps this is intended to be a premisein a revised version of the famous argument that reason cannot producemotives but morals can, though what he writes here is tantalizinglydifferent from that argument as it appears repeatedly intheTreatise.

Why did Hume omit the more fundamental arguments for themotivational inertia of reason? He may have reconsidered and rejectedthem. For example, he may have given up his undefended claim thatpassions have no representative character, a premise of theRepresentation Argument on which, as we saw, some of his fundamentalanti-rationalist arguments depend. Or he may have retained these viewsbut opted not to appeal to anything so arcane in a work aimed at abroader audience and intended to be as accessible as possible. ThemoralEnquiry makes no use of ideas and impressions, and so noarguments that depend on that distinction can be offered there,including the Representation Argument. Apparently Hume thought he couldshow that reason and sentiment rule different domains without usingthose arguments.

Thus, not surprisingly, the causal analysis of sympathy as a mechanismof vivacity-transferal from the impression of the self to the ideas ofthe sentiments of others is entirely omitted from the moralEnquiry. Hume still appeals to sympathy there to explain theorigin of all moral approval and disapproval, but he explains oursympathy with others simply as a manifestation of the sentiment ofhumanity, which is given more prominence. He is still concerned about the objection that sympathetically-acquired sentiments vary with spatial and temporal distance from theobject of evaluation while moral assessments do not; so he addresses it in the moralEnquiry as well,and resolves it by appealing once again to the common point of view.In theEnquiry he places more emphasis on sympathy with the interests of the whole of society, in part achieved by conversation using shared moral vocabulary,as a way to correct our initial sentiments to make them genuinely moral (Taylor 2002). He also attends more explicitly to the role of reason and reflection in moral evaluation. Some interpreters see him as offering an account of how to arrive at reliable moral judgment superior to that in theTreatise (Taylor 2015).

The distinction between artificial and natural virtues thatdominates the virtue ethics of theTreatise is almost entirelyabsent from the moralEnquiry; the term ‘artificial’occurs in the latter only once in a footnote. Gone are the paradoxes ofproperty and promises intended to prove that particular virtues aredevised on purpose; also missing is what some commentators think Hume’smost original contribution to the theory of justice, his account ofconvention. Yet Hume briefly sketches part of the same quasi-historical account of the origin of justice that he gives in theTreatise; and while the emphasis has shifted, Hume not onlytries to show that justice has merit only because of its beneficialconsequences, but that “public utility is thesoleorigin of justice” — were we not to find it useful (and insome conditions we might not) we would not even have such a thing (EPM3.1.1). While any explanation of this shift and these omissions ismerely speculative, here it seems that Hume does not change his mindabout the arguments of theTreatise but chooses to lead thereader to the same conclusions by more subtle and indirect means whileavoiding provocative claims.

In the moralEnquiry Hume is more explicit about what hetakes to be the errors of Christian (or, more cautiously, RomanCatholic) moralists. Not only have they elevated craven humility tothe status of a virtue, which he hints in theTreatise is amistake, but they also favor penance, fasting, and other“monkish virtues” that are in fact disapproved by allreasonable folk for their uselessness and disagreeableness, and so arein fact vices.

Bibliography

Primary Sources: Works by Hume

• A Treatise of Human Nature: A CriticalEdition, David Fate Norton and Mary J. Norton (eds.), Oxford,Clarendon Press, 2007. (References to this work start withTand are followed by Book, Part, Section and paragraph number, inparentheses in the text.)
• A Treatise of Human Nature (Oxford Philosophical Texts),David Fate Norton and Mary J. Norton (eds.), Oxford, Clarendon Press,2000.
• A Treatise of Human Nature, L. A. Selby-Bigge (ed.),2nd ed. revised by P.H. Nidditch, Oxford: Clarendon Press, 1975.
• An Enquiry Concerning the Principles of Morals, TomL. Beauchamp (ed.) (The Claredon Edition of the Works of David Hume),Oxford, Oxford University Press, 1998. (References to this work startwithEPM and are followed by Part, Section (if any), andparagraph number, in parentheses within the text.)
• Enquiry concerning Human Understanding, inEnquiries concerning Human Understanding and concerning thePrinciples of Morals, L. A. Selby-Bigge (ed.), 3rd ed revised byP. H. Nidditch, Oxford: Clarendon Press, 1975
• Essays Moral, Political, and Literary, Eugene Miller(ed.), Indianapolis: Liberty Fund, 1985.

Secondary Sources

• Abramson, Kate, 2001, “Sympathy and the Project of Hume’sSecondEnquiry,”Archiv für Geschichte derPhiosophie, 83(1): 45–80.
• Ainslie, Donald C. and Butler, Annemarie, 2015,The CambridgeCompanion to Hume’s Treatise, Cambridge: Cambridge UniversityPress.
• Árdal, Páll, 1966,Passion and Value in Hume’sTreatise, Edinburgh: Edinburgh University Press;2^nd edition, revised, 1989.
• –––, 1977a, “Another Look at Hume’sAccount of Moral Evaluation,”Journal of the History ofPhilosophy, 15: 405–421.
• –––, 1977b, “Convention andValue,” inDavid Hume: Bicentenary Papers, G.P. Morice(ed.), Austin: University of Texas Press, pp. 51–68.
• Baier, Annette C., 1988, “Hume’s Account of Social Artifice— Its Origins and Originality,”Ethics, 98:757–778.
• –––, 1991,A Progress of Sentiments,Cambridge, MA: Harvard University Press.
• –––, 2010,The Cautious Jealous Virtue: Hume onJustice, Cambridge, MA: Harvard University Press.
• Baron, Marcia, 1982, “Hume’s Noble Lie: An Account of HisArtificial Virtues,”Canadian Journal of Philosophy,12: 539–55.
• Blackburn, Simon, 1993, “Hume on the MezzanineLevel,”Hume Studies, 19 (2): 273–288.
• Botros, Sophie, 2006,Hume, Reason and Morality: A legacy ofcontradiction, London and New York: Routledge.
• Bricke, John, 1996,Mind and Morality, Oxford: ClarendonPress.
• Brown, Charlotte, 1988, “Is Hume anInternalist?”Journal of the History of Philosophy, 26:69–87.
• Capaldi, Nicholas, 1975,David Hume: The Newtonian Philosopher, Boston:Twayne Publishing.
• –––, 1989,Hume’s Place in Moral Philosophy, NewYork: Peter Lang.
• Chappell, V.C. (ed.), 1996,Hume: A Collection of CriticalEssays, Garden City, NY: Doubleday.
• Cohon, Rachel, 1997a, “The Common Point of View in Hume’sEthics,”Philosophy and Phenomenological Research, 57(4): 827–850.
• –––, 1997b, “Hume’s Difficulty with the Virtue ofHonesty,”Hume Studies, 23 (1): 91–112.
• –––, 2001a, “The Shackles of Virtue: Hume onAllegiance to Government,”History of PhilosophyQuarterly, 18 (4): 393–413.
• ––– (ed.), 2001b,Hume: Moral and PoliticalPhilosophy, Aldershot, England and Burlington, Vermont:Dartmouth/Ashgate.
• –––, 2008,Hume’s Morality: Feeling and Fabrication,Oxford: Oxford University Press.
• Coventry, Angela and Sager, Alexander, 2019,The HumeanMind, New York: Routledge.
• Darwall, Stephen, 1995,The British Moralists and the Internal‘Ought’, Cambridge: Cambridge University Press.
• Dees, Richard H., 1997, “Hume on the Characters of Virtue,”Journal of the History of Philosophy, 35 (1): 45–65.
• Falk, W.D., 1976, “Hume on Is and Ought,”Canadian Journalof Philosophy, 6: 359–378.
• Flew, Antony, 1963, “On the Interpretation ofHume,”Philosophy, 38: 178–181.
• Foot, Philippa, 1963, “Hume on Moral Judgment,” inD. Pears (ed.),David Hume: A Symposium, London: St. Martin’sPress, pp. 74–80.
• Forbes, Duncan, 1975,Hume’s Philosophical Politics, CambridgeUniversity Press.
• –––, 1977, “Hume’s Science of Politics” in David Hume: Bicentenary Papers, G.P. Morice (ed.), Austin, TX:University of Texas Press, pp. 39–50.
• Garrett, Don, 1997,Cognition and Commitment in Hume’sPhilosophy, Oxford: Oxford University Press.
• –––, 2007, “The First Motive to Justice: Hume’s CircleArgument Squared,”Hume Studies, 33 (2): 257–288.
• Gauthier, David, 1979, “David Hume, Contractarian,”ThePhilosophical Review, 88: 3–38.
• –––, 1992, “Artificial Virtues and the SensibleKnave,”Hume Studies, 18 (2): 401–427.
• Gill, Michael, 2000, “Hume’s Progressive View of HumanNature,”Hume Studies, 26 (1): 87–108.
• –––, 2006,The British Moralists on Human Nature and theBirth of Secular Ethics, Cambridge: Cambridge University Press.
• Haakonssen, Knud, 1981,The Science of a Legislator: The NaturalJurisprudence of David Hume and Adam Smith, Cambridge: CambridgeUniversity Press.
• Hampton, Jean, 1995, “Does Hume Have an InstrumentalConception of Practical Reason?”,Hume Studies,21 (1): 57–74.
• Harris, James A., 2015,Hume: An Intellectual Biography,Cambridge: Cambridge University Press.
• Harrison, Jonathan, 1981,Hume’s Theory of Justice, Oxford,Clarendon Press.
• –––, 1976,Hume’s Moral Epistemology, Oxford:Clarendon Press.
• Hudson, W.D., 1964, “Hume on Is and Ought,”ThePhilosophical Quaterly, 14: 246–252.
• Hunter Geoffrey, 1962, “Hume on Is and Ought,”Philosophy, 37: 148–152.
• –––, 1963, “Reply to Professor Flew,”Philosophy, 38: 182–184.
• Hutcheson, Francis, 1742,Illustrations on the Moral Sense,Bernard Peach (ed.), Cambridge: Harvard University Press.
• Jensen, Henning, 1977, “Hume on Moral Agreement,”Mind, 86: 497–513.
• Kail, P. J. E., 2007,Projection and Realism in Hume’sPhilosophy, Oxford: Oxford University Press.
• Korsgaard, Christine M., 1999, “The General Point of View: Love andMoral Approval in Hume’s Ethics,”Hume Studies, 25(1/2): 3–41.
• Livingston, Donald W., 1984,Hume’s Philosophy of Common Life,Chicago: University of Chicago Press.
• Loeb, Louis, 1977, “Hume’s Moral Sentiments and the Structure ofthe Treatise,”Journal of the History of Philosophy, 15:395–403.
• MacIntyre, A.C., 1959, “Hume on ‘Is’ and‘Ought’,”Philosophical Review, 68: 451–468.
• Mackie, J.L., 1980,Hume’s Moral Theory, London: Routledge.
• Magri, Tito, 1996, “Natural Obligation and Normative Motivation inHume’s Treatise,”Hume Studies, 22 (2): 231–253.
• McIntyre, Jane, 1990, “Character: A Humean Account,”History of Philosophy Quarterly, 7: 193–206.
• Miller, David, 1981,Philosophy and Ideology in Hume’s PoliticalThought, Oxford: Clarendon Press.
• Millgram, Elijah, 1995, “Was Hume a Humean?,”HumeStudies, 21 (1): 75–93.
• Morice, G. P. (ed.), 1977,David Hume: Bicentenary Papers, Austin, TX:University of Texas Press.
• Norton, David Fate, 1982,David Hume: Common-Sense Moralist,Sceptical Metaphysician, Princeton: Princeton University Press.
• –––, 1985, “Hume’s MoralOntology,”Hume Studies, 11 (Supplement):189–214.
• ––– (ed.), 1993,The Cambridge Companion toHume, Cambridge: Cambridge University Press.
• –––, 1993a, “Hume, Human Nature, and the Foundationsof Morality,” in Fate 1993, pp. 148–182.
• Norton, David Fate, and Jacqueline Taylor (eds.), 2008,The CambridgeCompanion to Hume, second edition, Cambridge: CambridgeUniversity Press.
• Persson, Ingmar, 1997, “Hume — Not a‘Humean’ about Motivation,”History ofPhilosophy Quarterly, 14 (2): 189–206.
• Radcliffe, Elizabeth S., 1996, “How Does the Humean Sense ofDuty Motivate?,”Journal of the History of Philosophy,34 (3): 383–407.
• –––, 1997, “Kantian Tunes on a HumeanInstrument: Why Hume is not Really a Skeptic about PracticalReasoning,”Canadian Journal of Philosophy, 27 (2):247–270.
• ––– (ed.), 2008,A Companion to Hume,Malden, MA: Blackwell Publishing.
• Raphael, D.D. (ed.), 1991,British Moralsits 1650 —1800, Volume I, Indianapolis: Hackett.
• Schneewind, J. B., 1998,The Invention of Autonomy,Cambridge: Cambridge Univ. Press.
• Setiya, Kieran, 2016, “Hume on Practical Reason,”reprinted inPractical Knowledge: Selected Essays, Oxford:Oxford University Press.
• Smith, Norman Kemp, 1941,The Philosophy of David Hume, London:Macmillian.
• Snare, Francis, 1991,Morals, Motivation and Convention, Cambridge: Cambridge University Press.
• Stewart, John B., 1992,Opinion and Reform in Hume’s PoliticalPhilosophy, Princeton, NJ: Princeton University Press.
• Stroud, Barry, 1977,Hume, London: Routledge.
• –––, 1993, “Gliding or ‘Staining’ The Worldwith ‘Sentiments’ and ‘Phantasms,’”Hume Studies, 19 (2): 253–272.
• Sturgeon, Nicholas, 2001, “Moral Skepticism and MoralNaturalism in Hume’s Treatise,”Hume Studies, 21 (1):3–83.
• Swain, Corliss, 1992, “Passionate Objectivity,”Noûs, 26 (4): 465–490.
• Taylor, Jacqueline, 1998, “Justice and the Foundations ofSocial Morality in Hume’s Treatise,”Hume Studies, 24(1): 5–30.
• –––, 2002, “Hume on the Standard ofVirtue,”Journal of Ethics, 6: 43–62.
• –––, 2015,Reflecting Subjects: Passion,Sympathy and Society in Hume’s Philosophy, Oxford: OxfordUniversity Press.
• Traiger, Saul (ed.), 2006,The Blackwell Guide to Hume’sTreatise, Malden, MA: Blackwell Publishing.
• Whelan, Frederick, 1981,Order and Artifice in Hume’s PoliticalPhilosophy, Princeton: Princeton University Press.
• Wiggins, David, 1998, “A Sensible Subjectivism?” inNeeds, Values, Truth, 3^rd edition, Oxford: ClarendonPress, pp. 185–210.

Academic Tools

	How to cite this entry.
	Preview the PDF version of this entry at theFriends of the SEP Society.
	Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
	Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

• The Hume Society
• “Moral Theory”, section of the entry on Hume, by James Fieser (U. Tennessee/Martin),in theInternet Encyclopedia of Philosophy.

Related Entries

Hume, David |property and ownership